Homebrew PCBs

Aaron's site is back up by the way.
http://www.aaroncake.net/circuits/

I'm going over his schematic getting myself familiar with how things work
and I have a question concerning the motor wiring.
http://www.aaroncake.net/circuits/stepper.htm

There are 6 wires on a unipolar motor, 2 sets of 3 wires.  The center wire
of each set appears to be a common power source.  But how do you go about
deciding the order of the other wires in relation with the schematic?

Ok, I've uploaded a picture in the photo section of the group.  I've created
my photo album and I can't figure out how to add a picture.  It doesn't give
me the add photo icon like it does in the main photo section.  It's a pic of
a unipolar stepper motor wiring.  It should be easy to see, it's the only
pic there 'cause everyone else was smart enough to know how to upload pics
properly.

It's the order of the yellow/orange and black/brown wires that I'd like to
connect properly.  I'd like to avoid creating any magic smoke if at all
possible.

Robert
:)




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

An ohmmeter is the easy way.  Make a table of 6 rows and 6 columns. 
Each row and column represents a colored wire.  Take resistance
readings between each combination of wire pairs.  Some readings will
be open.  Other readings will show a resistance.  The three wires
sharing a reading are a pair of windings.  One wire will be a center
tap.  Lets suppose two of the readings are about 100 ohms and the
third reading is 200 ohms.  The two wires with the 200 ohm readings
are the end taps, and the other color shared by the 100 ohm readings
is the center tap.


--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@v...>
wrote:
> Aaron's site is back up by the way.
> http://www.aaroncake.net/circuits/
> 
> 
> There are 6 wires on a unipolar motor, 2 sets of 3 wires.  The
center wire
> of each set appears to be a common power source.  But how do you go
about
> deciding the order of the other wires in relation with the
schematic?
> 
> Yahoo! Groups Links

Nevermind, I've just noticed something on the schematic that matches the
wires with the motor wiring.  Instead of trying to make heads or tails of
the colours, it's the drawing on the right of the motor wiring that held the
answer.

The 1st 4027 has Q and /Q, so that matches with A and /A.
The 2nd 4027 has Q and /Q, so that matches with B and /B.

Robert
:)




-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de Robert Hedan
Envoyé : juin 17 2005 21:40
À : Homebrew_PCBs@yahoogroups.com
Objet : [Homebrew_PCBs] Unipolar stepper motor driver wiring


Aaron's site is back up by the way. http://www.aaroncake.net/circuits/

I'm going over his schematic getting myself familiar with how things work
and I have a question concerning the motor wiring.
http://www.aaroncake.net/circuits/stepper.htm

There are 6 wires on a unipolar motor, 2 sets of 3 wires.  The center wire
of each set appears to be a common power source.  But how do you go about
deciding the order of the other wires in relation with the schematic?

Ok, I've uploaded a picture in the photo section of the group.  I've created
my photo album and I can't figure out how to add a picture.  It doesn't give
me the add photo icon like it does in the main photo section.  It's a pic of
a unipolar stepper motor wiring.  It should be easy to see, it's the only
pic there 'cause everyone else was smart enough to know how to upload pics
properly.

It's the order of the yellow/orange and black/brown wires that I'd like to
connect properly.  I'd like to avoid creating any magic smoke if at all
possible.

Robert
:)




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links



 





Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

Once the two pairs have been identified, hooked them up to a
controller.  Send the motor a control sequence activating each of the
coils.  If the motor rotates, its hooked up correctly.  If it doesn't
rotate correctly, reverse the end taps on one pair of the windings and
it should work properly. A full step (wave drive) sequence is probably
the easiest to use for testing.


--- In Homebrew_PCBs@yahoogroups.com, "bob_ledoux" <bobledoux@p...>
wrote:
> An ohmmeter is the easy way.  Make a table of 6 rows and 6 columns. 
> Each row and column represents a colored wire.  Take resistance
> readings between each combination of wire pairs.  Some readings will
> be open.  Other readings will show a resistance.  The three wires
> sharing a reading are a pair of windings.  One wire will be a center
> tap.  Lets suppose two of the readings are about 100 ohms and the
> third reading is 200 ohms.  The two wires with the 200 ohm readings
> are the end taps, and the other color shared by the 100 ohm readings
> is the center tap.
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, Robert Hedan
<robert.hedan@v...>
> wrote:
> > Aaron's site is back up by the way.
> > http://www.aaroncake.net/circuits/
> > 
> > 
> > There are 6 wires on a unipolar motor, 2 sets of 3 wires.  The
> center wire
> > of each set appears to be a common power source.  But how do you
go

> about
> > deciding the order of the other wires in relation with the
> schematic?
> > 
> > Yahoo! Groups Links

--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan
<robert.hedan@v...> 
wrote:
> There are 6 wires on a unipolar motor, 2 sets of 3 wires.  The 
center wire
> of each set appears to be a common power source.  But how do you go 
about
> deciding the order of the other wires in relation with the
schematic?

There is a lot of information about stepper motor wireing including
an 
online "expert" system that helps you figure out the wireing based on 
resistance measurements between the wires at

http://www.piclist.com/techref/io/steppers.htm

I've done what I believe are improvements to the 12V stepper motor driver
schematic I had in the Photo section.  It's no longer there, I moved it to
the CNC folder of the Files section.

I've added opto-couplers between the CMOS chips and the transistors, lowered
the CMOS voltage down to 5 volts and increased the motor voltage to 24 volts
(which is what my motors are rated).

Comments please on the modifications.

Robert
:)


Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

On Mon, 20 Jun 2005 14:59:04 +0200, Robert Hedan  
<robert.hedan@...> wrote:

>
> I've added opto-couplers between the CMOS chips and the transistors,  
> lowered
> the CMOS voltage down to 5 volts and increased the motor voltage to 24  
> volts
> (which is what my motors are rated).


24V motors? they are going to have a lot of inductance and resistance.

Probably OK for simple drivers, but if you ever plan to use chopper  
drivers you are looking for low voltage low ohms motors.


(The idea is to drive like a 3V motor with 30V, to quickly increase the  
current in the motor to the level you want, and then reduce the drive  
voltage. For speed.)

ST

On Monday 20 June 2005 08:59 am, Robert Hedan wrote:
> I've done what I believe are improvements to the 12V stepper motor driver
> schematic I had in the Photo section.  It's no longer there, I moved it to
> the CNC folder of the Files section.
>
> I've added opto-couplers between the CMOS chips and the transistors,
> lowered the CMOS voltage down to 5 volts and increased the motor voltage to
> 24 volts (which is what my motors are rated).
>
> Comments please on the modifications.

What are you using for driver transistors there?

TIP31, I've included a parts list on the schematic in the CNC folder of the
Files Section.

Transformer 24V 1.5A
4070 XOR
4027 Flip-flop
MCT6 opto-coupler
1K 1/4W resistor
TIP31 100V 3A
1N4002 100V 1A
Steppers 24V, 36/72ohms, unknown amperage, obsolete model SP57B by COPAL.

Robert
:)







What are you using for driver transistors there?








Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

maybe this has been discussed but why not use mosfets instead of the
NPNs?  If you can turn them fully on they will run a lot cooler with
little voltage drop. 

Also, how are you limiting current?  

Will the 1N400x be fast enough to supress the spikes?  I thought one
needed to have schottkys for that.

Reasons to use a PIC (or other micros) is so you can do microstepping
which makes for a much smoother running stepper and you can use PWM
(aka chopping) on the coils to allow higher voltage and thus more
power.  But I'm sure that's all been discussed before.  by the way,
some of the newer 8 pin Z8s have a tansimpedance amp build in for
current sense - perfect for a stepper controller.  I'm thinking of
prototyping with one.

The allegro chips do this pretty well also and allegro is very easy
when it comes to samples.  I'd look at the A3982, it's new and comes
in a soic package which is much easier to solder than the older A3977.
 They'll send 3 units if you ask nice.

Phil

--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@v...>
wrote:
> TIP31, I've included a parts list on the schematic in the CNC folder
of the
> Files Section.
> 
> Transformer 24V 1.5A
> 4070 XOR
> 4027 Flip-flop
> MCT6 opto-coupler
> 1K 1/4W resistor
> TIP31 100V 3A
> 1N4002 100V 1A
> Steppers 24V, 36/72ohms, unknown amperage, obsolete model SP57B by
COPAL.
> 
> Robert
> :)
> 
> 
> 
> 
> 
> 
> 
> What are you using for driver transistors there?
> 
> 
> 
> 
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, Files, and
Photos:

> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
> Yahoo! Groups Links

Those puppies are faster than grease lightning, they use the structure used 
by pin diodes. I use the 1N4007 to switch rf signals by controlling the bias.

At 05:31 PM 6/20/2005, you wrote:
>maybe this has been discussed but why not use mosfets instead of the
>NPNs?  If you can turn them fully on they will run a lot cooler with
>little voltage drop.
>
>Also, how are you limiting current?
>
>Will the 1N400x be fast enough to supress the spikes?  I thought one
>needed to have schottkys for that.
>
>Reasons to use a PIC (or other micros) is so you can do microstepping
>which makes for a much smoother running stepper and you can use PWM
>(aka chopping) on the coils to allow higher voltage and thus more
>power.  But I'm sure that's all been discussed before.  by the way,
>some of the newer 8 pin Z8s have a tansimpedance amp build in for
>current sense - perfect for a stepper controller.  I'm thinking of
>prototyping with one.
>
>The allegro chips do this pretty well also and allegro is very easy
>when it comes to samples.  I'd look at the A3982, it's new and comes
>in a soic package which is much easier to solder than the older A3977.
>  They'll send 3 units if you ask nice.
>
>Phil
>
>--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@v...>
>wrote:
> > TIP31, I've included a parts list on the schematic in the CNC folder
>of the
> > Files Section.
> >
> > Transformer 24V 1.5A
> > 4070 XOR
> > 4027 Flip-flop
> > MCT6 opto-coupler
> > 1K 1/4W resistor
> > TIP31 100V 3A
> > 1N4002 100V 1A
> > Steppers 24V, 36/72ohms, unknown amperage, obsolete model SP57B by
>COPAL.
> >
> > Robert
> > :)
> >
> >
> >
> >
> >
> >
> >
> > What are you using for driver transistors there?
> >
> >
> >
> >
> >
> >
> >
> >
> > Be sure to visit the group home and check for new Links, Files, and
>Photos:
> > http://groups.yahoo.com/group/Homebrew_PCBs
> >
> > If Files or Photos are running short of space, post them here:
> > http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> > Yahoo! Groups Links
>
>
>
>
>Be sure to visit the group home and check for new Links, Files, and Photos:
>http://groups.yahoo.com/group/Homebrew_PCBs
>
>If Files or Photos are running short of space, post them here:
>http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
>Yahoo! Groups Links
>
>
>
>
>
>
>
>
>
>--
>No virus found in this incoming message.
>Checked by AVG Anti-Virus.
>Version: 7.0.323 / Virus Database: 267.7.8/22 - Release Date: 6/17/2005

Cecil Bayona
KD5NWA
www.qrpradio.com

'Never argue with an idiot. They drag you down to their level then beat you 
with experience.'

why do most hbridges and stepper controllers use schottkys?

--- In Homebrew_PCBs@yahoogroups.com, KD5NWA <KD5NWA@c...> wrote:
> Those puppies are faster than grease lightning, they use the
structure used 
> by pin diodes. I use the 1N4007 to switch rf signals by controlling
the bias.
> 
> At 05:31 PM 6/20/2005, you wrote:
> >maybe this has been discussed but why not use mosfets instead of the
> >NPNs?  If you can turn them fully on they will run a lot cooler with
> >little voltage drop.
> >
> >Also, how are you limiting current?
> >
> >Will the 1N400x be fast enough to supress the spikes?  I thought one
> >needed to have schottkys for that.
> >
> >Reasons to use a PIC (or other micros) is so you can do microstepping
> >which makes for a much smoother running stepper and you can use PWM
> >(aka chopping) on the coils to allow higher voltage and thus more
> >power.  But I'm sure that's all been discussed before.  by the way,
> >some of the newer 8 pin Z8s have a tansimpedance amp build in for
> >current sense - perfect for a stepper controller.  I'm thinking of
> >prototyping with one.
> >
> >The allegro chips do this pretty well also and allegro is very easy
> >when it comes to samples.  I'd look at the A3982, it's new and comes
> >in a soic package which is much easier to solder than the older A3977.
> >  They'll send 3 units if you ask nice.
> >
> >Phil
> >
> >--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan <robert.hedan@v...>
> >wrote:
> > > TIP31, I've included a parts list on the schematic in the CNC folder
> >of the
> > > Files Section.
> > >
> > > Transformer 24V 1.5A
> > > 4070 XOR
> > > 4027 Flip-flop
> > > MCT6 opto-coupler
> > > 1K 1/4W resistor
> > > TIP31 100V 3A
> > > 1N4002 100V 1A
> > > Steppers 24V, 36/72ohms, unknown amperage, obsolete model SP57B by
> >COPAL.
> > >
> > > Robert
> > > :)
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > > What are you using for driver transistors there?
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > > Be sure to visit the group home and check for new Links, Files, and
> >Photos:
> > > http://groups.yahoo.com/group/Homebrew_PCBs
> > >
> > > If Files or Photos are running short of space, post them here:
> > > http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> > > Yahoo! Groups Links
> >
> >
> >
> >
> >Be sure to visit the group home and check for new Links, Files, and
Photos:
> >http://groups.yahoo.com/group/Homebrew_PCBs
> >
> >If Files or Photos are running short of space, post them here:
> >http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> >Yahoo! Groups Links
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >--
> >No virus found in this incoming message.
> >Checked by AVG Anti-Virus.
> >Version: 7.0.323 / Virus Database: 267.7.8/22 - Release Date: 6/17/2005
> 
> Cecil Bayona
> KD5NWA
> www.qrpradio.com
> 
> 'Never argue with an idiot. They drag you down to their level then
beat you 
> with experience.'

"maybe this has been discussed but why not use mosfets instead of the NPNs?
If you can turn them fully on they will run a lot cooler with little voltage
drop."

'Cause the original schematic was configured with NPNs, 'cause MOSFETs
weren't included in alternatives and 'cause I lack experience in
substituting NPNs for MOSFETs. :)  If you have a suggestion, please post a
pic, I'm a visual person.


"Also, how are you limiting current?"

See, this is something else I need more information.  I have no idea what
you mean by limitting current.

I started with a schematic from Aaron's site which he had gotten from
someone else.  I'm just trying to make the circuit stable to use the 24V
steppers that I happen to have on hand.


Robert
:)





Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

On Tue, 21 Jun 2005 01:57:34 +0200, KD5NWA <KD5NWA@...> wrote:

> Those puppies are faster than grease lightning, they use the structure  
> used
> by pin diodes. I use the 1N4007 to switch rf signals by controlling the  
> bias.


That would be the first time i hear that. I was much rather believing the  
1n400x is a slow rectifier diode.
Using the bias of a diode to switch rf does not require a fast diode, if  
you don't need to switch fast.

I would like to know where you got that information, and maybe you should  
look at the datasheets again.

ST

On Tue, 21 Jun 2005 05:36:54 +0200, Phil <phil1960us@...> wrote:

> why do most hbridges and stepper controllers use schottkys?


because they are faster. all switchmode/pwm circuits use faster diodes.
In the simple controller suggested there is no chopping, the frequency is  
low, a 1n400x could be ok.

ST

> In the simple controller suggested 

A blast from the past....Is the obsolete Allegro UCN5804 unipolar 
driver that difficult to get these days? I found 3 in my parts box and 
4 installed in boards here, using anything other than that for a simple 
unipolar driver is a waste of time IMO.

http://www.allegromicro.com/datafile/archive/5804.pdf


--- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan" 
<stefan_trethan@g...> wrote:
> On Tue, 21 Jun 2005 05:36:54 +0200, Phil <phil1960us@y...> wrote:
> 
> > why do most hbridges and stepper controllers use schottkys?
> 
> 
> because they are faster. all switchmode/pwm circuits use faster 
diodes.
> In the simple controller suggested there is no chopping, the 
frequency is  

> low, a 1n400x could be ok.
> 
> ST

Jameco, DigiKey and Mouser, my favorite suppliers, have none.

Gonna check the surplus places now.

Robert
:(



A blast from the past....Is the obsolete Allegro UCN5804 unipolar 
driver that difficult to get these days?



Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

This site has "pulled" UCN5804s for $3.50 each;

http://www.alltronics.com/linear_ic_list.htm

--- In Homebrew_PCBs@yahoogroups.com, "derekhawkins" 
<derekhawkins@y...> wrote:
> > In the simple controller suggested 
> 
> A blast from the past....Is the obsolete Allegro UCN5804 unipolar 
> driver that difficult to get these days? I found 3 in my parts box 
and 
> 4 installed in boards here, using anything other than that for a 
simple 

> unipolar driver is a waste of time IMO.
> 
> http://www.allegromicro.com/datafile/archive/5804.pdf
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan" 
> <stefan_trethan@g...> wrote:
> > On Tue, 21 Jun 2005 05:36:54 +0200, Phil <phil1960us@y...> wrote:
> > 
> > > why do most hbridges and stepper controllers use schottkys?
> > 
> > 
> > because they are faster. all switchmode/pwm circuits use faster 
> diodes.
> > In the simple controller suggested there is no chopping, the 
> frequency is  
> > low, a 1n400x could be ok.
> > 
> > ST

Check here;

http://www.alltronics.com/linear_ic_list.htm

--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
<robert.hedan@v...> wrote:
> Jameco, DigiKey and Mouser, my favorite suppliers, have none.
> 
> Gonna check the surplus places now.
> 
> Robert
> :(
> 
> 
> 
> A blast from the past....Is the obsolete Allegro UCN5804 unipolar 
> driver that difficult to get these days?
> 
> 
> 
> Be sure to visit the group home and check for new Links, Files, and 
Photos:

> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
> Yahoo! Groups Links

Make that $6.50 each...

Robert
:(




This site has "pulled" UCN5804s for $3.50 each;

http://www.alltronics.com/linear_ic_list.htm





Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

Look at the bottom of that page;

UCN5804B pulls
Famous stepper motor driver integrated circuit (spec sheet available 
see our 93I002) these chips have been discontinued and are hard to 
find. We lucked onto a pile of solder pulls and they are offered here 
at a greatly reduced price. These trims have all been tested. Inquire 
about 100 lot pricing.
25I001 Unit Price : $3.50 



--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
<robert.hedan@v...> wrote:
> Make that $6.50 each...
> 
> Robert
> :(
> 
> 
> 
> 
> This site has "pulled" UCN5804s for $3.50 each;
> 
> http://www.alltronics.com/linear_ic_list.htm
> 
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, Files, and 
Photos:

> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
> Yahoo! Groups Links

I'm blind, can't see the pulls.  :D

But I did see another model: L6506D.

What do you think of that chip?  Maybe it's more readily available, haven't
checked yet.  I'm busy unpacking my pillow blocks.  Woohoo!

Robert
:)


-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de derekhawkins
Envoyé : juin 21 2005 10:45
À : Homebrew_PCBs@yahoogroups.com
Objet : Re: RE : RE : [Homebrew_PCBs] Unipolar stepper motor driver wiring


Look at the bottom of that page;

UCN5804B pulls
Famous stepper motor driver integrated circuit (spec sheet available 
see our 93I002) these chips have been discontinued and are hard to 
find. We lucked onto a pile of solder pulls and they are offered here 
at a greatly reduced price. These trims have all been tested. Inquire 
about 100 lot pricing.
25I001 Unit Price : $3.50 



--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
<robert.hedan@v...> wrote:
> Make that $6.50 each...
> 
> Robert
> :(
> 
> 
> 
> 
> This site has "pulled" UCN5804s for $3.50 each;
> 
> http://www.alltronics.com/linear_ic_list.htm
> 
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, Files, and
Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here: 
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> Yahoo! Groups Links




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

<http://www.jameco.com/webapp/wcs/stores/servlet/CatalogSearchResultView?lan
gId=-1&storeId=10001&catalogId=10001&searchType=k&searchValue=l6506d&categor
yId=106093>

If the link screws up, search for "l6506d" on the Jameco web site:

$4.25 DIP / $4.49 SOIC, new.

Is this IC as robust as the UCN5804?

Robert
:)




-----Message d'origine-----
De : Homebrew_PCBs@yahoogroups.com [mailto:Homebrew_PCBs@yahoogroups.com] De
la part de derekhawkins
Envoyé : juin 21 2005 10:45
À : Homebrew_PCBs@yahoogroups.com
Objet : Re: RE : RE : [Homebrew_PCBs] Unipolar stepper motor driver wiring


Look at the bottom of that page;

UCN5804B pulls
Famous stepper motor driver integrated circuit (spec sheet available 
see our 93I002) these chips have been discontinued and are hard to 
find. We lucked onto a pile of solder pulls and they are offered here 
at a greatly reduced price. These trims have all been tested. Inquire 
about 100 lot pricing.
25I001 Unit Price : $3.50 



--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
<robert.hedan@v...> wrote:
> Make that $6.50 each...
> 
> Robert
> :(
> 
> 
> 
> 
> This site has "pulled" UCN5804s for $3.50 each;
> 
> http://www.alltronics.com/linear_ic_list.htm
> 
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, Files, and
Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here: 
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> Yahoo! Groups Links




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

Nevermind, it's not quite the same thing.  It seems as if they broke down
the UCN5804 into various chips.


"The L6506/D is a linear integrated circuit designed
to sense and control the current in stepping motors
and similar devices. When used in conjunction with
the L293, L298, L7150, L6114/L6115, the chip set
forms a constant current drive for an inductive load
and performs all the interface function from the control
logic thru the power stage.

Two or more devices may be synchronized using
the sync pin. In this mode of operation the oscillator
in the master chip sets the operating frequency in all
chips."

Robert
:(




Is this IC as robust as the UCN5804?



Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

--- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan"
<stefan_trethan@g...> wrote:
> On Tue, 21 Jun 2005 05:36:54 +0200, Phil <phil1960us@y...> wrote:
> 
> > why do most hbridges and stepper controllers use schottkys?
> 
> 
> because they are faster. all switchmode/pwm circuits use faster diodes.
> In the simple controller suggested there is no chopping, the
frequency is  
> low, a 1n400x could be ok.
> 
> ST

It was a rhetorical question, of course they are faster.  It's
completely up to the designer what he wants to do but I'd be careful
and engineer for reliability.  I wouldn't use 1N400x even though they
are practically free.  The 1N400x will probably work most of the time
but I'd worry about extreme cases like a high speed stall or direction
reverse.

Also, I wasn't aware that the kickback waveform speed was related to
the drive frequency.  I thought it was intrinsic to the driver and
windings (and mechanical design) of the motor.  To be good enough, the
diode needs to let less energy though than can damage the driver (er,
shunt it off).  The reverse recovery time of the diode needs to be
fast enough for that.  Some energy will get through so the issues are
driving voltage and switching time.  I've heard that typical
rectifiers have a Trr of about .5 uS.  Unfortunately, this isn't
spec'd.  Schottkys are in the range of 10s of nS so there is a big
gap.  Unfortunately no one seems to document Trr on the spec sheets
for barrier schottkys, either.  sigh.

On Tue, 21 Jun 2005 18:45:30 +0200, Phil <phil1960us@...> wrote:

>
> Also, I wasn't aware that the kickback waveform speed was related to
> the drive frequency.  I thought it was intrinsic to the driver and
> windings (and mechanical design) of the motor.  To be good enough, the
> diode needs to let less energy though than can damage the driver (er,
> shunt it off).  The reverse recovery time of the diode needs to be
> fast enough for that.  Some energy will get through so the issues are
> driving voltage and switching time.  I've heard that typical
> rectifiers have a Trr of about .5 uS.  Unfortunately, this isn't
> spec'd.  Schottkys are in the range of 10s of nS so there is a big
> gap.  Unfortunately no one seems to document Trr on the spec sheets
> for barrier schottkys, either.  sigh.


Hmm, interesting theory, i would still think there is some relation to  
drive speed.

For example, if you look at a relay, most wouldn't argue that a 400x is  
not ok.

But your point makes sense.
Well, hopefully the diodes in the FETs will catch the mess (if FETs are  
used)

ST

Allegro says they have replaced it with the 3967 and 3977. 

Have a look here: http://www.allegromicro.com/sf/3967/

They might even send you a sample. 

David

derekhawkins wrote:

>Look at the bottom of that page;
>
>UCN5804B pulls
>Famous stepper motor driver integrated circuit (spec sheet available 
>see our 93I002) these chips have been discontinued and are hard to 
>find. We lucked onto a pile of solder pulls and they are offered here 
>at a greatly reduced price. These trims have all been tested. Inquire 
>about 100 lot pricing.
>25I001 Unit Price : $3.50 
>
>  
>

> I'm blind, can't see the pulls.  :D

Then buy them new for $6.50.

http://www.alltronics.com/stepper_motors.htm

93I002    $6.50 each 


--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
<robert.hedan@v...> wrote:
> I'm blind, can't see the pulls.  :D
> 
> But I did see another model: L6506D.
> 
> What do you think of that chip?  Maybe it's more readily available, 
haven't
> checked yet.  I'm busy unpacking my pillow blocks.  Woohoo!
> 
> Robert
> :)
> 
> 
> -----Message d'origine-----
> De : Homebrew_PCBs@yahoogroups.com 
[mailto:Homebrew_PCBs@yahoogroups.com] De
> la part de derekhawkins
> Envoyé : juin 21 2005 10:45
> À : Homebrew_PCBs@yahoogroups.com
> Objet : Re: RE : RE : [Homebrew_PCBs] Unipolar stepper motor driver 
wiring
> 
> 
> Look at the bottom of that page;
> 
> UCN5804B pulls
> Famous stepper motor driver integrated circuit (spec sheet 
available 
> see our 93I002) these chips have been discontinued and are hard to 
> find. We lucked onto a pile of solder pulls and they are offered 
here 
> at a greatly reduced price. These trims have all been tested. 
Inquire 
> about 100 lot pricing.
> 25I001 Unit Price : $3.50 
> 
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
> <robert.hedan@v...> wrote:
> > Make that $6.50 each...
> > 
> > Robert
> > :(
> > 
> > 
> > 
> > 
> > This site has "pulled" UCN5804s for $3.50 each;
> > 
> > http://www.alltronics.com/linear_ic_list.htm
> > 
> > 
> > 
> > 
> > 
> > Be sure to visit the group home and check for new Links, Files, 
and
> Photos:
> > http://groups.yahoo.com/group/Homebrew_PCBs
> > 
> > If Files or Photos are running short of space, post them here: 
> > http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> > Yahoo! Groups Links
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, Files, and 
Photos:

> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
> Yahoo! Groups Links

> I'm blind, can't see the pulls.  :D

Then buy them new for $6.50.

http://www.alltronics.com/stepper_motors.htm

93I002    $6.50 each 


--- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
<robert.hedan@v...> wrote:
> I'm blind, can't see the pulls.  :D
> 
> But I did see another model: L6506D.
> 
> What do you think of that chip?  Maybe it's more readily available, 
haven't
> checked yet.  I'm busy unpacking my pillow blocks.  Woohoo!
> 
> Robert
> :)
> 
> 
> -----Message d'origine-----
> De : Homebrew_PCBs@yahoogroups.com 
[mailto:Homebrew_PCBs@yahoogroups.com] De
> la part de derekhawkins
> Envoyé : juin 21 2005 10:45
> À : Homebrew_PCBs@yahoogroups.com
> Objet : Re: RE : RE : [Homebrew_PCBs] Unipolar stepper motor driver 
wiring
> 
> 
> Look at the bottom of that page;
> 
> UCN5804B pulls
> Famous stepper motor driver integrated circuit (spec sheet 
available 
> see our 93I002) these chips have been discontinued and are hard to 
> find. We lucked onto a pile of solder pulls and they are offered 
here 
> at a greatly reduced price. These trims have all been tested. 
Inquire 
> about 100 lot pricing.
> 25I001 Unit Price : $3.50 
> 
> 
> 
> --- In Homebrew_PCBs@yahoogroups.com, Robert Hedan 
> <robert.hedan@v...> wrote:
> > Make that $6.50 each...
> > 
> > Robert
> > :(
> > 
> > 
> > 
> > 
> > This site has "pulled" UCN5804s for $3.50 each;
> > 
> > http://www.alltronics.com/linear_ic_list.htm
> > 
> > 
> > 
> > 
> > 
> > Be sure to visit the group home and check for new Links, Files, 
and
> Photos:
> > http://groups.yahoo.com/group/Homebrew_PCBs
> > 
> > If Files or Photos are running short of space, post them here: 
> > http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> > Yahoo! Groups Links
> 
> 
> 
> 
> Be sure to visit the group home and check for new Links, Files, and 
Photos:

> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
> Yahoo! Groups Links

Robert Hedan wrote:
> "maybe this has been discussed but why not use mosfets instead of the NPNs?
> If you can turn them fully on they will run a lot cooler with little voltage
> drop."

Mosfets will work just fine.  You'll need to assure
sufficient drive voltage in the case of an H-bridge
for the upper N channel mosfets.  This means needing
somewhere between 2-10V greater for the drive than
the mosfet is switching.  Same issue exists for NPN
bipolar drivers though not as severe.  For unipolar
steppers this issue doesn't exist.

BTW the discussion of external suppression diodes for
mosfets is likely moot for devices where an internal
parasitic diode is present.  The main reason to add
an external schottky in SMPS designs is that of
efficiency as the voltage drop across the internal
diode is typically greater than available via the
external schottky.

> "Also, how are you limiting current?"
>
> See, this is something else I need more information.  I have no idea what
> you mean by limitting current.

If you want to drive a stepper at maximum possible
velocity it requires a high drive voltage to the
coils.  This voltage is only needed initially in
the cycle and in fact if allowed to be imposed
statically on the windings will cause excessive
current consumption and quickly overheat the works.
One of the techniques to achieve this voltage profile
is a constant current scheme which limits the steady
state winding current to a nominal value but will
allow the highest available drive voltage to be
applied initially in the step cycle.

-- 
uhmgawa@...        www.gnu.org

In a message dated 6/21/2005 9:10:54 A.M. Central Daylight Time,  
robert.hedan@... writes:

Is the  obsolete Allegro UCN5804 unipolar 
driver that difficult to get these  days?



why not Google Allegro's web-address, and ask them for  distributors!


[Non-text portions of this message have been removed]

Using IN4007 diodes as pin diods is a quite a well known ham radio trick 
Stefan.

Bill
----- Original Message ----- 

From: "Stefan Trethan" <stefan_trethan@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Tuesday, June 21, 2005 4:18 PM
Subject: Re: RE : [Homebrew_PCBs] Unipolar stepper motor driver wiring


> On Tue, 21 Jun 2005 01:57:34 +0200, KD5NWA <KD5NWA@...> wrote:
>
>> Those puppies are faster than grease lightning, they use the structure
>> used
>> by pin diodes. I use the 1N4007 to switch rf signals by controlling the
>> bias.
>
>
> That would be the first time i hear that. I was much rather believing the
> 1n400x is a slow rectifier diode.
> Using the bias of a diode to switch rf does not require a fast diode, if
> you don't need to switch fast.
>
> I would like to know where you got that information, and maybe you should
> look at the datasheets again.
>
> ST
>
>
> Be sure to visit the group home and check for new Links, Files, and 
> Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
>
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> Yahoo! Groups Links
>
>
>
>
>
>
>

Thanks David,

But the 3967 and 3977 are designed for bipolar steppers.  I've requested a
sample of 3967 for testing on another application, but they don't help me
with my unipolar steppers.

Robert
:)



Allegro says they have replaced it with the 3967 and 3977. 

David




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

My surplus source has the UCN5804 in stock.  I'm waiting for a reply with
prices.

I know I'm going to get a batch of these babies.  I know unipolar steppers
are on their way out, but there are still a lot in circulation.

Robert
:)



Is the  obsolete Allegro UCN5804 unipolar 
driver that difficult to get these  days?



Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

I picked this up from the Gecko group.  We will have a problem with the
power cables to the motors and must take some preventive action.  Here's the
answer:



> I just found out today that whenever I have my Bridgeport with 201 
> drivers turned on  all AM radios in the building next door have a squeal 
> that is so loud that no one can listen to them .  OOOPS !  Talk about 
> being a bad neighbor !
>   I tried a jumper lead from the machine body to the metal box that the 
> drivers are in and it helped a good bit .
> Any good ideas for RFI .




Here's what will 'take you off of the air':

1) Run your motor leads thru a 5-conductor shielded cable. 4 leads are 
for the motor connections, the 5th conductor goes to the motor case, 
the other end goes to the drive's power supply GND (term 1).

Ground the shields to your control box 'single-point' ground location 
(usually the '-' terminal of your big filter cap). Leave the motor end 
of the shield unconnected.

2) Use an AC line filter (Corcomp or similar) on the 115 VAC going into 
your control box transformer.

It'll then be as quiet as a baby. No, wait; that's not a good analogy...



The Gecko Yahoo group focuses on the Gecko CNC controller/driver system.
You can get more information about them here:
http://www.geckodrive.com/products.htbml
and here:
http://finance.groups.yahoo.com/group/geckodrive/


Just remember to say I sent you so that I can collect my regular 50% cut on
all future sales.


Robert
:D




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

On Wed, 22 Jun 2005 01:55:06 +0200, Bill Maxwell <wrmaxwell@...>  
wrote:

> Using IN4007 diodes as pin diods is a quite a well known ham radio trick
> Stefan.
> Bill


I am not into radio, but i thought a PIN diode is used to get a variable  
capacity by changing the bias point.
This does NOT require a fast diode, the bias is changed quite slowly.
I do not doubt it can be used as pin diode, but this does not make it a  
fast diode.
a 1n4007 is a slow mains frequency rectifier diode.

ST

On Wed, 22 Jun 2005 08:06:00 +0200, Robert Hedan  
<robert.hedan@...> wrote:

> I know I'm going to get a batch of these babies.  I know unipolar  
> steppers
> are on their way out, but there are still a lot in circulation.
> Robert
>


I wouldn't be too sure 'bout that.

By the way, you can use a unipolar stepper in bipolar mode as 2 phase  
motor. I've heard it works well.

ST

On Wed, 22 Jun 2005 08:32:17 +0200, Robert Hedan  
<robert.hedan@...> wrote:

>
> Here's what will 'take you off of the air':
> 1) Run your motor leads thru a 5-conductor shielded cable. 4 leads are
> for the motor connections, the 5th conductor goes to the motor case,
> the other end goes to the drive's power supply GND (term 1).
> Ground the shields to your control box 'single-point' ground location
> (usually the '-' terminal of your big filter cap). Leave the motor end
> of the shield unconnected.
> 2) Use an AC line filter (Corcomp or similar) on the 115 VAC going into
> your control box transformer.
> It'll then be as quiet as a baby. No, wait; that's not a good analogy...
>


If you use the 24V steppers and no chopper driver you probably will have  
much less problems anyway.
I've pulled the shield from cables in the past, if you compress the mesh  
together it gets rather wide and you can slip it onto existing cables, if  
you can't find the right shielded one.

ST

That's true Stefan and I now see the point you were making.  My response was 
mistaken in thinking you had not heard of the IN4007's PIN function.  Sorry.

Bill
----- Original Message ----- 

From: "Stefan Trethan" <stefan_trethan@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Wednesday, June 22, 2005 4:56 PM
Subject: Re: RE : [Homebrew_PCBs] Unipolar stepper motor driver wiring


> On Wed, 22 Jun 2005 01:55:06 +0200, Bill Maxwell <wrmaxwell@...>
> wrote:
>
>> Using IN4007 diodes as pin diods is a quite a well known ham radio trick
>> Stefan.
>> Bill
>
>
> I am not into radio, but i thought a PIN diode is used to get a variable
> capacity by changing the bias point.
> This does NOT require a fast diode, the bias is changed quite slowly.
> I do not doubt it can be used as pin diode, but this does not make it a
> fast diode.
> a 1n4007 is a slow mains frequency rectifier diode.
>
> ST
>
>
> Be sure to visit the group home and check for new Links, Files, and 
> Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
>
> If Files or Photos are running short of space, post them here:
> http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> Yahoo! Groups Links
>
>
>
>
>
>
>

On Wed, 22 Jun 2005 10:35:35 +0200, Bill Maxwell <wrmaxwell@...>  
wrote:

> That's true Stefan and I now see the point you were making.  My response  
> was
> mistaken in thinking you had not heard of the IN4007's PIN function.   
> Sorry.
> Bill


Indeed i had not heard of it, and it is good to know.
But fast it isn't, which you already knew as i see now.

ST

Uh, change in plans.  I've been fiddling around with the 24V motors I have
on hand and they appear to be VERY demagnetized.  I didn't notice this until
I picked up a set of smaller steppers and noticed a drastic difference in
resistance.

So now I have to get myself decent steppers.  I'm not going back to the
local surplus store, I'll just get the same problem (dozens of steppers all
piled in the same box).  I'm going to buy larger motors for the CNC machine
later, but for now, I'm looking for decent intermediate motors that will
make my drilling station work as well as machine wood and plastics.

I was checking these pulls from Xylotex:
http://www.xylotex.com/DoubleStepper.htm

Problem is now I have to reconsider a totally new driver circuit 'cause
these are bipolar steppers.  Can I modify my schematic without TOO much
trouble?  Or am I better using a new design?

Argh, back to square one...

Robert
:(


(schematic in the CNC folder of the Files section of the group.



Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

Darn,

It seems the circuit I had for unipolar steppers can be wired for bipolar
steppers as well.  I found a copy of it here, but there's just one problem:

Figure 9.1: http://eio.com/jasstep.htm#translators

"I think when you reverse direction and continue stepping, the motor will
advance 1 more step in the previous direction it was going before
responding."

Ugh...

Robert
:(
 





Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

Stefan Trethan wrote:
> I am not into radio, but i thought a PIN diode is used to get a variable  
> capacity by changing the bias point.

You are likely thinking of a varicap which
has a change in capacitance of about 2x
when the DC bias imposed upon it varies over
a 0 to ~30V range.

Actually all diodes exhibit this "variable
capacitance" effect, though varicaps are
specifically designed to do so.

-- 
uhmgawa@...        www.gnu.org

Robert Hedan wrote:

> Uh, change in plans.  I've been fiddling around with the 24V motors I have
> on hand and they appear to be VERY demagnetized.  I didn't notice this until
> I picked up a set of smaller steppers and noticed a drastic difference in
> resistance.
> 


   LOL..  If by 'resistance' you mean cogging, that has nothing to do with being 
demagnetized.  Short all the wires on one of your motors together, and see if it 
isn't pretty damn hard to turn.

   You should forget these and get some 5V 1A or so motors, but for other 
reasons rather than this.


> So now I have to get myself decent steppers.  I'm not going back to the
> local surplus store, I'll just get the same problem (dozens of steppers all
> piled in the same box).  I'm going to buy larger motors for the CNC machine
> later, but for now, I'm looking for decent intermediate motors that will
> make my drilling station work as well as machine wood and plastics.
> 

   They're usually in a can, and even if they weren't just tossing them in a box 
isn't likely to demagnetize much of anything.  They're shipped in boxes of 50 or 
100, about as close as just tossed together.


> I was checking these pulls from Xylotex:
> http://www.xylotex.com/DoubleStepper.htm
> 
> Problem is now I have to reconsider a totally new driver circuit 'cause
> these are bipolar steppers.  Can I modify my schematic without TOO much
> trouble?  Or am I better using a new design?
> 

   Probably better to find some good solid 5V unipolar motors and get using them 
until you've learned a bit more about steppers in general.  Some of it simply 
comes from getting and using all types, start simple.

Alan

Robert Hedan wrote:

> Darn,
> 
> It seems the circuit I had for unipolar steppers can be wired for bipolar
> steppers as well.  I found a copy of it here, but there's just one problem:
> 
> Figure 9.1: http://eio.com/jasstep.htm#translators
> 
> "I think when you reverse direction and continue stepping, the motor will
> advance 1 more step in the previous direction it was going before
> responding."
> 


   That is "bi-directional", as in turning the motor both ways, vs only one way 
in the circuit above it.  Have to keep your 'bi-'s straight, nothing to do with 
running a bipolar motor, that takes two full H-bridges as in his fig 6.1.  Stick 
to unipolar until you have a better handle on what you're doing, bipolar has 
advantages but you don't need them to start and it takes twice as much to drive, 
plus other problems in the driving.




   Also note that anyone can read a few other people's web pages, draw up a 
circuit or two, and look like they know what they're doing in five minutes. 
Says more to their web page creation skills than anything else really.  7486 is 
typically a 5V chip.  The upper transistors will only pull up to 5V, with more 
power wasted in them than making it to the motor with a 12V supply.  In other 
words a crap draw up, likely it was copied from somewhere else with changes and 
only a partial understanding of what's going on.

   While 90% of the info on most of these basic stepper motor type web pages is 
correct, 90% or more of them also have a notable error or two.  It's easy to get 
the basics and make a web page, but relatively few of the people who do that 
tend to have a good understanding before they make their pages.  Don't count on 
any of them being 100%.

Alan

> I wouldn't be too sure 'bout that.
> 
> By the way, you can use a unipolar stepper in bipolar mode as 2 phase  
> motor. I've heard it works well.
> 
> ST
> 


   Have to be careful with this though, four coils on instead of two, so you 
have to make sure your motors are designed to run that way as well.  More than a 
few will overheat if you use full power in all four coils.  Definite improvement 
for the ones that can run this way though.

Alan

Robert Hedan wrote:
> Problem is now I have to reconsider a totally new driver circuit 'cause
> these are bipolar steppers.  Can I modify my schematic without TOO much
> trouble?  Or am I better using a new design?

IMHO steppers are a rather brute-force way to
address a limited set of applications.  If all
you need are simple open-loop indexed positioning,
minimal/moderate static holding torque for gratis,
but without much of a velocity/torque range,
steppers will fit the bill.

I'd spend some time looking at a DC motor approach
with shaft encoder positional feedback.  As a
back of the envelope estimate a 28TPI lead screw
combined with the optical encoder scavenged from
a $3 mouse can give you positional accuracy in
the 0.001" range.  Physical drive requirements
take the form of a simple H-bridge without the
gyrations needed by steppers.

You aren't likely to find anything off the shelf
to address the drive and closed loop control but
nearly any uC controller will due.  Direction and
velocity of the motor are proportional to the average
energy seen by the motor and is driven via PWM.
As with all but trivial stepper designs you'll need
to address velocity ramp up/down.  And if needed,
an algorithm to emulate static hold of the motor
position.  Though with a lead screw of sufficient
pitch and its inherent friction this may be
unneeded or at least trivial.

-- 
uhmgawa@...        www.gnu.org

>You should forget these and get some 5V 1A or so motors, but for 
>other reasons rather than this.

That's exactly what I found in one of my storage boxes in the garage 
over the weekend. Found three 5V 1A unipolar steppers that were 
hooked up to a board with three UCN5804 drivers. The power supply was 
30V and there were 25 ohm high wattage resistors in series with the 
coils.  A poor man's substitute for chopper drives. Built this around 
5 years ago. I'll probably dump these on E-bay someday.

--- In Homebrew_PCBs@yahoogroups.com, Alan King <alan@n...> wrote:
> Robert Hedan wrote:
> 
> > Uh, change in plans.  I've been fiddling around with the 24V 
motors I have
> > on hand and they appear to be VERY demagnetized.  I didn't notice 
this until
> > I picked up a set of smaller steppers and noticed a drastic 
difference in
> > resistance.
> > 
> 
> 
>    LOL..  If by 'resistance' you mean cogging, that has nothing to 
do with being 
> demagnetized.  Short all the wires on one of your motors together, 
and see if it 
> isn't pretty damn hard to turn.
> 
>    You should forget these and get some 5V 1A or so motors, but for 
other 
> reasons rather than this.
> 
> 
> > So now I have to get myself decent steppers.  I'm not going back 
to the
> > local surplus store, I'll just get the same problem (dozens of 
steppers all
> > piled in the same box).  I'm going to buy larger motors for the 
CNC machine
> > later, but for now, I'm looking for decent intermediate motors 
that will
> > make my drilling station work as well as machine wood and 
plastics.
> > 
> 
>    They're usually in a can, and even if they weren't just tossing 
them in a box 
> isn't likely to demagnetize much of anything.  They're shipped in 
boxes of 50 or 
> 100, about as close as just tossed together.
> 
> 
> > I was checking these pulls from Xylotex:
> > http://www.xylotex.com/DoubleStepper.htm
> > 
> > Problem is now I have to reconsider a totally new driver 
circuit 'cause
> > these are bipolar steppers.  Can I modify my schematic without 
TOO much
> > trouble?  Or am I better using a new design?
> > 
> 
>    Probably better to find some good solid 5V unipolar motors and 
get using them 
> until you've learned a bit more about steppers in general.  Some of 
it simply 

> comes from getting and using all types, start simple.
> 
> Alan

> BTW the discussion of external suppression diodes for
> mosfets is likely moot for devices where an internal
> parasitic diode is present.  The main reason to add
> an external schottky in SMPS designs is that of
> efficiency as the voltage drop across the internal
> diode is typically greater than available via the
> external schottky.


   The main reason for schottky is to make sure it turns on and takes the heat, 
if it is not schottky then the internal diode may turn on instead and the extra 
dissipation will go back into the transistor.  Very important it goes through 
the right one in a high powered circuit that's being run near the transistor's 
limit to start with.


   Oversize the FETs, make sure they have the intrinsic reverse diode, and 
forget about extra reverse diodes.  Big FETs will easily deal with 5 or 8 V, 1 
or 2 A steppers if everything else is correct, without any extra parts helping out.


   99% of general failure blown FETs come from blowing the gates either through 
poor handling or poor drive, and then get blamed on every other possible failure 
idea under the sun.  Difficult to even blow them on purpose after they're in a 
proper circuit.  Always install about everything else first, then the 
transistors.  You'll easily damage the gates in handling the board if they're 
installed first and the rest of the gate circuit isn't in place.

Alan

>I'd spend some time looking at a DC motor approach
>with shaft encoder positional feedback.

Just like those fooling around with toner transfer now will someday 
move on to photoetching so will those on steppers now move on to DC 
servos. Seems to be a natural progression, at least it was for me.

--- In Homebrew_PCBs@yahoogroups.com, uhmgawa <uhmgawa@m...> wrote:
> Robert Hedan wrote:
> > Problem is now I have to reconsider a totally new driver 
circuit 'cause
> > these are bipolar steppers.  Can I modify my schematic without 
TOO much

> > trouble?  Or am I better using a new design?
> 
> IMHO steppers are a rather brute-force way to
> address a limited set of applications.  If all
> you need are simple open-loop indexed positioning,
> minimal/moderate static holding torque for gratis,
> but without much of a velocity/torque range,
> steppers will fit the bill.
> 
> I'd spend some time looking at a DC motor approach
> with shaft encoder positional feedback.  As a
> back of the envelope estimate a 28TPI lead screw
> combined with the optical encoder scavenged from
> a $3 mouse can give you positional accuracy in
> the 0.001" range.  Physical drive requirements
> take the form of a simple H-bridge without the
> gyrations needed by steppers.
> 
> You aren't likely to find anything off the shelf
> to address the drive and closed loop control but
> nearly any uC controller will due.  Direction and
> velocity of the motor are proportional to the average
> energy seen by the motor and is driven via PWM.
> As with all but trivial stepper designs you'll need
> to address velocity ramp up/down.  And if needed,
> an algorithm to emulate static hold of the motor
> position.  Though with a lead screw of sufficient
> pitch and its inherent friction this may be
> unneeded or at least trivial.
> 
> -- 
> uhmgawa@m...        www.gnu.org

"That is "bi-directional", as in turning the motor both ways, vs only one
way 
in the circuit above it.  Have to keep your 'bi-'s straight, nothing to do
with 
running a bipolar motor, that takes two full H-bridges as in his fig 6.1."

D'oh!  I knew I should have gone to bed instead of kept on researching this.



"While 90% of the info on most of these basic stepper motor type web pages
is 
correct, 90% or more of them also have a notable error or two."

And that's why I only 'add' to existing circuits.  :)  Putting in
opto-couplers and protection diodes does nothing to the overall performance
of a circuit.  Either it worked to begin with, or it didn't.

I learned this real quick by reading a thread in Aaron's forum.  Check it
out for yourself:
<http://www.aaroncake.net/forum/topic.asp?TOPIC_ID=2996&FORUM_ID=18&CAT_ID=1
0&Forum_Title=Power+Supply&Topic_Title=12%2F120V+inverter+again>
I needed an inverter to power the backlight of a LCD and started reading
that to find a decent schematic.  13 pages later and I still don't have an
answer, the worst mumbojumbo I've ever read anywhere, it just goes on and
on...



And that's why I show you folks these circuits.  I'd rather listen to your
opinion, than buying things that won't work with money I don't have.

Robert
:)


Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

--- In Homebrew_PCBs@yahoogroups.com, "Phil" <phil1960us@y...> wrote:
> maybe this has been discussed but why not use mosfets instead of the
> NPNs?  If you can turn them fully on they will run a lot cooler with
> little voltage drop. 
> 
> Also, how are you limiting current?  
> 

Would it be easier to just use the L297/L298 combo? Together they do
PWM limited current and half-stepping. They are bipolar internally,
but it's all inside the package.

Steve Greenfield

"LOL..  If by 'resistance' you mean cogging, that has nothing to do with
being 
demagnetized.  Short all the wires on one of your motors together, and see
if it 
isn't pretty damn hard to turn."

Yeah, I guess I meant cogging.  None of the 6 steppers I have showed any
effect whatsoever from shorting all the wires, both unipolar and bipolar
(1).


"Probably better to find some good solid 5V unipolar motors and get using
them 
until you've learned a bit more about steppers in general.  Some of it
simply 
comes from getting and using all types, start simple."

I have some 3967 Microstepping Driver with Translator coming as well as
UCN5804s.  The 9V Stepper400 from Melnick is interesting.

robert
:)



Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

Hi Robert,

Did the same thing, ordered the Steppr400's and 5804's
from Nelnick.

Good price for them ($6.99), worth $20 bucks to try
them out.

Al C

--- Robert Hedan <robert.hedan@...> wrote:

> "LOL..  If by 'resistance' you mean cogging, that
> has nothing to do with
> being 
> demagnetized.  Short all the wires on one of your
> motors together, and see
> if it 
> isn't pretty damn hard to turn."
> 
> Yeah, I guess I meant cogging.  None of the 6
> steppers I have showed any
> effect whatsoever from shorting all the wires, both
> unipolar and bipolar
> (1).
> 
> 
> "Probably better to find some good solid 5V unipolar
> motors and get using
> them 
> until you've learned a bit more about steppers in
> general.  Some of it
> simply 
> comes from getting and using all types, start
> simple."
> 
> I have some 3967 Microstepping Driver with
> Translator coming as well as
> UCN5804s.  The 9V Stepper400 from Melnick is
> interesting.
> 
> robert
> :)
> 
> 
> 
> Be sure to visit the group home and check for new
> Links, Files, and Photos:
> http://groups.yahoo.com/group/Homebrew_PCBs
> 
> If Files or Photos are running short of space, post
> them here:
>
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/
> 
> Yahoo! Groups Links
> 
> 
> 
>  
> 
> 
> 
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com

"Have to be careful with this though, four coils on instead of two, so you 
have to make sure your motors are designed to run that way as well.  More
than a 
few will overheat if you use full power in all four coils.  Definite
improvement 
for the ones that can run this way though."  - Alan


Nah, gonna go with the herd  :D  and stick with unipolar steppers for this
first machine.  I'll probably move up to bipolar for the CNC version.  Right
now with the 'discovery' of the UCN5804, things seem to be a breeze to get
going, economical to start up and difficult to screw up.

This drilling station is to help me save time when I fabricate PCBs.  It was
not meant to be a project to take my attention for several weeks.  Yeah
right, famous last words...

Robert
:)




Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBs

If Files or Photos are running short of space, post them here:
http://groups.yahoo.com/group/Homebrew_PCBs_Archives/ 
Yahoo! Groups Links

Robert Hedan wrote:
> Problem is now I have to reconsider a totally new driver circuit 'cause
> these are bipolar steppers.  Can I modify my schematic without TOO much
> trouble?  Or am I better using a new design?

The discrete driver contained in "stepper_driver_mod.JPG"
can be modified for bipolar drive by creating a totem
pole structure of an upper PNP and lower NPN device.
E of PNP to V+, C of PNP + C of NPN becomes the drive
output, E of NPN to GND.  You will need a means to
drive B of the PNP devices opposite polarity w/r/t
the NPN devices.

One of these structures will be needed for each of the
four leads of the bipolar stepper resulting in a
total of 8 driver devices.  Do you actually intend to
drive 28V rated steppers off this design or are you
intending the higher voltage to enable faster stepping
rates?  If the latter you will need some means of
current control to avoid overheating the windings.

Concerning the transistor base drive you might want
to check the gain (hfe) available and verify if the
optocouplers and added series resistance results in
sufficient base drive current.

Though the circuitry is minimal from a gate count
perspective, I'd prefer to see a single uC package
displace the 2 SSI packages and buy the design
flexibility of control software.  IMHO I'd toss
the steppers along with half of the required driver
transistors, constant current source, add the
positional feedback and DC motor, and create if you
will a "digital motor".  You are going to need the
the control software to run somewhere and pushing
it to a dedicated uC for each drive motor can
greatly simplify the implementation.

-- 
uhmgawa@...        www.gnu.org

Please trim the quoted parts.

To be specific, that is true only of the 1N4007. It's my understanding
that the other lower voltage 1N400x series are not the same.

Steve Greenfield

--- In Homebrew_PCBs@yahoogroups.com, KD5NWA <KD5NWA@c...> wrote:
> Those puppies are faster than grease lightning, they use the
structure used 
> by pin diodes. I use the 1N4007 to switch rf signals by controlling
the bias.
> 
> At 05:31 PM 6/20/2005, you wrote:
...

> >Will the 1N400x be fast enough to supress the spikes?  I thought one
> >needed to have schottkys for that.
> >

derekhawkins wrote:
>>I'd spend some time looking at a DC motor approach
>>with shaft encoder positional feedback.
> 
> 
> Just like those fooling around with toner transfer now will someday 
> move on to photoetching so will those on steppers now move on to DC 
> servos. Seems to be a natural progression, at least it was for me.

Another benefit of closed loop control is the ability to
detect/correct positioning errors which isn't possible
with an open-loop stepper approach.  If the stepper-driven
drive mechanism hits a friction snag and misses a step(s)
you are going to be cutting somewhere you'd rather not.

Additionally it is possible (perhaps even preferable) to
avoid direct drive of the lead screw by the DC motor and
pick up low speed torque in the process.  This speed reduction
can be implemented as a "belt drive" from motor to lead
screw with no loss in positional accuracy when the encoder
remains on the lead screw.  Note depending upon the torque
required the "belts and pulleys" can be as low-tech as
those found in consumer electronic devices.  i.e.: scavenging
a few scrap VCRs will supply the needed prototype components.

-- 
uhmgawa@...        www.gnu.org

On Wed, 22 Jun 2005 16:30:02 +0200, derekhawkins <derekhawkins@...>  
wrote:

>
> Just like those fooling around with toner transfer now will someday
> move on to photoetching so will those on steppers now move on to DC
> servos. Seems to be a natural progression, at least it was for me.


Imagine that, i did photoetching for years and moved to toner transfer,  
and couldn't be happier about it.

DC servos are not the answer to a simple drilling machine for sure, as the  
controllers are unnecessary complicated.
But we all have our opinions.

ST

--- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan"
<stefan_trethan@g...> wrote:
...
> 
> For example, if you look at a relay, most wouldn't argue that a 400x
is  
> not ok.

Agreed. I use them for relays and haven't blown a driver yet (famous
last words).  I think the reason is because there is very little
mechanical energy in the relay.  A motor with permanent magnets acts
as a generator so you have both the field collapse emf and the rotor's
mechanical energy. Given a high speed stop, I'd bet there is more emf
from the mechanical energy than the field collapse.  Again, its not
just about switching speed but about the amount of energy that doesn't
get shunted off.

What I don't know is how to calculate when its safe to use slower
diodes and when not.  So I err on the side of safety.

On Tuesday 21 June 2005 03:49 pm, uhmgawa wrote:

> BTW the discussion of external suppression diodes for mosfets is likely moot
> for devices where an internal parasitic diode is present.  The main reason
> to add an external schottky in SMPS designs is that of efficiency as the
> voltage drop across the internal diode is typically greater than available
> via the external schottky.

This is the second post I've seen so far that seems to suggest that it's okay 
to rely on the internal diodes in mosfets.  The general consensus in other 
lists seems to be that this is _not_ a real good idea,  for a number of 
reasons.  One of those is that those internal diodes don't have very 
consistent characteristics,  and another is that they turn on too slow,  and 
turn off too slow,  which will tend to make the device run hot.  Since heat 
is an enemy of semiconductors,  I'd tend to go with the external diodes,  
myself.

There is no question about it, servo motors are better than 
steppers, but at a cost.  True servos are much more expensive than 
steppers of equal power.  Servo drives are much more complicated to 
build (or purchase) than stepper drivers.  Steppers natural 
attribute of "Stepping" is perfect for easy control via logic.  Most 
of the low-end servo drives try to replicate the functionality of a 
stepper drive (Step and direction).

I will convert my router to servos someday, but when I finally hit 
the limitations of my current stepper motor driven system.

For a simple drilling (and even milling) machine for circuit boards, 
you really can't beat a stepper system for simplicity and cost. 
Heck, if my $50.00 200 Oz. In. stepper motors can drive my router 
(chewing through hard maple), milling a circuit board or drilling 
one is a piece of cake.







-- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan" 
<stefan_trethan@g...> wrote:
> On Wed, 22 Jun 2005 16:30:02 +0200, derekhawkins 
<derekhawkins@y...>  
> wrote:
> 
> >
> > Just like those fooling around with toner transfer now will 
someday
> > move on to photoetching so will those on steppers now move on to 
DC
> > servos. Seems to be a natural progression, at least it was for 
me.
> 
> 
> Imagine that, i did photoetching for years and moved to toner 
transfer,  
> and couldn't be happier about it.
> 
> DC servos are not the answer to a simple drilling machine for 
sure, as the  

> controllers are unnecessary complicated.
> But we all have our opinions.
> 
> ST

that's a decent solution but I hate the L298's pin arrangement - makes
prototyping a PITA.   Prototyping with mosfets is pretty easy as long
as you get the drivers right.

By the way, I'm a big fan of using SPICE to figure out what this stuff
does.  I can read theory till I'm blue and not get it but fiddling
with a SPICE emulation gives me a real feel for how the theory
relates.  Highly recommended, in some way, better than putting
together real components - a LOT cheaper, changes are fast and you can
see how current and voltage interrelate in your circuit.   It helped
demystify mosfets for me.  Wish I knew how to simulate BEMF, though.

--- In Homebrew_PCBs@yahoogroups.com, "Steve" <alienrelics@y...> > 

> Would it be easier to just use the L297/L298 combo? Together they do
> PWM limited current and half-stepping. They are bipolar internally,
> but it's all inside the package.
> 
> Steve Greenfield

--- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan"
<stefan_trethan@g...> wrote:
... 
> DC servos are not the answer to a simple drilling machine for sure,
as the  
> controllers are unnecessary complicated.
> But we all have our opinions.

While I'm reluctant to recommend servos for cost reasons, I think they
can be done at about the same complexity level provided one is
conversant with microcontroller design and programming.  They are more
reliable as open loop steppers can miss steps in some circumstances
and the encoder always tells you where you are.

--- In Homebrew_PCBs@yahoogroups.com, "Roy J. Tellason"
<rtellason@b...> wrote:
...
> This is the second post I've seen so far that seems to suggest that
it's okay 
> to rely on the internal diodes in mosfets.  The general consensus in
other 
> lists seems to be that this is _not_ a real good idea,  for a number of 
> reasons.  One of those is that those internal diodes don't have very 
> consistent characteristics,  and another is that they turn on too
slow,  and 
> turn off too slow,  which will tend to make the device run hot. 
Since heat 
> is an enemy of semiconductors,  I'd tend to go with the external
diodes,  
> myself.

agreed.  even if they switched fast, there would be dissipation IN THE
MOSFET.  Distributing the heat dissipation is a good thing.

For what it's worth I'm using unipolar (6 wire) steppers (the stock Sherline
steppers) on my milling machine, and they're being driven by a Xylotex board,
which uses the Allegro 3977 chip (bipolar driver) using 1/8 steps.

I have them wired up such that I'm only using 2 of the 4 coils. I'm currently
running the X & Y axis at 20 IPM which corresponds to 400 RPM (cause I have
20 tpi leadscrews). This works fine for machining purposes, so it should be
more than adequate for drilling.

If you decide to go with two coils in series, then you need to drop the
current by half (you wind up getting the same amount of torque, but the
increased resistance lowers your high end speed).

By adding some vibration dampeners (see the files section of
http://groups.yahoo.com/group/ComputerControlledMotion/) Jeff was able to get
90 IPM using a similar configuration.

> -----Original Message-----
> From: Homebrew_PCBs@yahoogroups.com 
> [mailto:Homebrew_PCBs@yahoogroups.com] On Behalf Of Alan King
> Sent: Wednesday, June 22, 2005 6:37 AM
> To: Homebrew_PCBs@yahoogroups.com
> Subject: Re: [Homebrew_PCBs] Unipolar stepper motor driver wiring
> 
> 
> 
> > I wouldn't be too sure 'bout that.
> > 
> > By the way, you can use a unipolar stepper in bipolar mode 
> as 2 phase  
> > motor. I've heard it works well.
> > 
> > ST
> > 
> 
> 
>    Have to be careful with this though, four coils on instead 
> of two, so you 
> have to make sure your motors are designed to run that way as 
> well.  More than a 
> few will overheat if you use full power in all four coils.  
> Definite improvement 
> for the ones that can run this way though.
> 
> Alan

Roy J. Tellason wrote:

> This is the second post I've seen so far that seems to suggest that it's okay 
> to rely on the internal diodes in mosfets.  The general consensus in other 
> lists seems to be that this is _not_ a real good idea...

It depends upon the design.  Certainly the intrinsic
diode isn't typically as well characterized as a
discrete diode (though I've seen exceptions to this)
nor typically as fast as schottky types.

But unless there is a overriding cost/performance
reason reason to add an external shunt diode the
intrinsic diode is often sufficient.

The energy being dissipated by any diode in an
inductive shunt role is going to be much less
than a diode used to free-wheel current in a
switching regulator application when the switch
device is off.  It is the latter case where the
total switch device package heat dissipation is
of consequence and indeed where external diodes
are routinely employed.

-- 
uhmgawa@...        www.gnu.org

>DC servos are not the answer to a simple drilling machine for sure, 
>as the controllers are unnecessary complicated.

I wouldn't consider three ICs complicated...Even simpler than most 
stepper drives. No doubt you're thinking of yesterday's controllers 
like the ever popular Geico 320.

--- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan" 
<stefan_trethan@g...> wrote:
> On Wed, 22 Jun 2005 16:30:02 +0200, derekhawkins 
<derekhawkins@y...>  
> wrote:
> 
> >
> > Just like those fooling around with toner transfer now will 
someday
> > move on to photoetching so will those on steppers now move on to 
DC
> > servos. Seems to be a natural progression, at least it was for me.
> 
> 
> Imagine that, i did photoetching for years and moved to toner 
transfer,  
> and couldn't be happier about it.
> 
> DC servos are not the answer to a simple drilling machine for sure, 
as the  

> controllers are unnecessary complicated.
> But we all have our opinions.
> 
> ST

Phil wrote:
> --- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan"
> <stefan_trethan@g...> wrote:
> ... 
> 
>>DC servos are not the answer to a simple drilling machine for sure,
> 
> as the  
> 
>>controllers are unnecessary complicated.
>>But we all have our opinions.
> 
> 
> While I'm reluctant to recommend servos for cost reasons, I think they
> can be done at about the same complexity level provided one is
> conversant with microcontroller design and programming.  They are more
> reliable as open loop steppers can miss steps in some circumstances
> and the encoder always tells you where you are.

Agreed.  I was assuming we are talking about
"midnight engineering" here rather than looking
for a $ervo motor off the shelf.  I'd expect the
lab-built "digital motor" approach to be less
than the cost of a roughly equivalent stepper
solution as viewed by total system cost.

-- 
uhmgawa@...        www.gnu.org

> I wouldn't consider three ICs complicated...Even simpler than most 
> stepper drives. No doubt you're thinking of yesterday's controllers 
> like the ever popular Geico 320.

Heck, if you can do a 3 IC circuit to drive servo motors, please post 
it!  That would be simple and cost effective and I for one would love 
to see it (make it & use it).

Thanks!

Chris

>I'd expect the lab-built "digital motor" approach to be less
>than the cost of a roughly equivalent stepper
>solution as viewed by total system cost.

A new shaft encoder can be had for $19.00 these days. Adding 
a "second shaft" to any surplus DC motor is a piece of cake if you 
have access to and know how to use a mini-lathe...Or buy a surplus 
motor with encoder. H-bridge ICs are around $5.00, microcontroller 
around $3.00...Here's what's missing....Good PID code for the 
micros....Right now it's under lock and key but will soon trickle 
into the public domain.


--- In Homebrew_PCBs@yahoogroups.com, uhmgawa <uhmgawa@m...> wrote:
> Phil wrote:
> > --- In Homebrew_PCBs@yahoogroups.com, "Stefan Trethan"
> > <stefan_trethan@g...> wrote:
> > ... 
> > 
> >>DC servos are not the answer to a simple drilling machine for 
sure,
> > 
> > as the  
> > 
> >>controllers are unnecessary complicated.
> >>But we all have our opinions.
> > 
> > 
> > While I'm reluctant to recommend servos for cost reasons, I think 
they
> > can be done at about the same complexity level provided one is
> > conversant with microcontroller design and programming.  They are 
more
> > reliable as open loop steppers can miss steps in some 
circumstances

> > and the encoder always tells you where you are.
> 
> Agreed.  I was assuming we are talking about
> "midnight engineering" here rather than looking
> for a $ervo motor off the shelf.  I'd expect the
> lab-built "digital motor" approach to be less
> than the cost of a roughly equivalent stepper
> solution as viewed by total system cost.
> 
> -- 
> uhmgawa@m...        www.gnu.org

>Heck, if you can do a 3 IC circuit to drive servo motors, please post
>it! That would be simple and cost effective and I for one would love
>to see it (make it & use it).

No way, there's money to be made. BTW, one of the ICs is just a chip 
resistor. 

--- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing" 
<lcdpublishing@y...> wrote:
> 
> > I wouldn't consider three ICs complicated...Even simpler than 
most 
> > stepper drives. No doubt you're thinking of yesterday's 
controllers 
> > like the ever popular Geico 320.
> 
> Heck, if you can do a 3 IC circuit to drive servo motors, please 
post 
> it!  That would be simple and cost effective and I for one would 
love 

> to see it (make it & use it).
> 
> Thanks!
> 
> Chris

Do you have a source for those shaft encoders that are $19.00 each?

The cheapest I have found is from US Digital and they are around 
$35.00 or so.

Thanks

Chris

> A new shaft encoder can be had for $19.00 these days. Adding 
> a "second shaft" to any surplus DC motor is a piece of cake if you 
> have access to and know how to use a mini-lathe...Or buy a surplus 
> motor with encoder. H-bridge ICs are around $5.00, microcontroller 
> around $3.00...Here's what's missing....Good PID code for the 
> micros....Right now it's under lock and key but will soon trickle 
> into the public domain.

Well I guess that won't help anyone else will it, so it's a mute 
point.


--- In Homebrew_PCBs@yahoogroups.com, "derekhawkins" 
<derekhawkins@y...> wrote:
> >Heck, if you can do a 3 IC circuit to drive servo motors, please 
post
> >it! That would be simple and cost effective and I for one would 
love
> >to see it (make it & use it).
> 
> No way, there's money to be made. BTW, one of the ICs is just a 
chip 

> resistor. 
> 
> --- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing" 
> <lcdpublishing@y...> wrote:
> > 
> > > I wouldn't consider three ICs complicated...Even simpler than 
> most 
> > > stepper drives. No doubt you're thinking of yesterday's 
> controllers 
> > > like the ever popular Geico 320.
> > 
> > Heck, if you can do a 3 IC circuit to drive servo motors, please 
> post 
> > it!  That would be simple and cost effective and I for one would 
> love 
> > to see it (make it & use it).
> > 
> > Thanks!
> > 
> > Chris

>Do you have a source for those shaft encoders that are $19.00 each?
>The cheapest I have found is from US Digital

Same place but they're limited to 300 CPR;

http://www.usdigital.com/products/e4p/

--- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing" 
<lcdpublishing@y...> wrote:

> Do you have a source for those shaft encoders that are $19.00 each?
> 
> The cheapest I have found is from US Digital and they are around 
> $35.00 or so.
> 
> Thanks
> 
> Chris
> 
> 
> > A new shaft encoder can be had for $19.00 these days. Adding 
> > a "second shaft" to any surplus DC motor is a piece of cake if you 
> > have access to and know how to use a mini-lathe...Or buy a surplus 
> > motor with encoder. H-bridge ICs are around $5.00, microcontroller 
> > around $3.00...Here's what's missing....Good PID code for the 
> > micros....Right now it's under lock and key but will soon trickle 
> > into the public domain.

derekhawkins wrote:
>>I'd expect the lab-built "digital motor" approach to be less
>>than the cost of a roughly equivalent stepper
>>solution as viewed by total system cost.
> 
> 
> A new shaft encoder can be had for $19.00 these days. Adding 
> a "second shaft" to any surplus DC motor is a piece of cake if you 
> have access to and know how to use a mini-lathe...Or buy a surplus 
> motor with encoder.

For lead screw applications I'd keep the encoder on
the screw.  Doing so allows the screw to be driven
indirectly and velocity/torque/rotational-precision
to be scaled via simple friction coupling without
introducing slippage error.

You can cut an encoder disk with a lathe but it would
be easier to print such on transparency film via laser
printer.  Postscript programs to do so are floating
around.  Tap Don Lancaster as he is a likely source
for such sundries.

Though I wouldn't bother initially.  36-segment photo
interrupter disks and associated dual element quadrature
IR detectors can be ransacked from a cast off PC mouse.
And two are employed for use in X and Y axis encoding.
While 36 segment seems rather coarse, it is quite usable
directly with a lead screw pitch of 24-28TPI.

Other sources for interrupter substitutes would be gears
reclaimed from defunct consumer items (mechanical clocks,
VCRs, walkmans, etc..).  The more difficult thing to
scrounge is the dual channel IR photo-interrupter used
for quadrature detection thus my suggestion of using
a mouse.

> H-bridge ICs are around $5.00, microcontroller 
> around $3.00...Here's what's missing....Good PID code for the 
> micros....Right now it's under lock and key but will soon trickle 
> into the public domain.

Don't hang onto that key too long.  Experience has
proven the benefit of community development and
support

-- 
uhmgawa@...        www.gnu.org

>For lead screw applications I'd keep the encoder on
>the screw. Doing so allows the screw to be driven
>indirectly and velocity/torque/rotational-precision
>to be scaled via simple friction coupling without
>introducing slippage error.

A DC gearhead motor is best for lead screw applications as it allows 
the screw to be driven "directly" without any complications provided 
speed and torque requirements are met. It doesn't get much simpler 
than this;

http://www.pbase.com/eldata/image/45200286

http://www.pbase.com/eldata/image/45200509

derekhawkins wrote:
>>For lead screw applications I'd keep the encoder on
>>the screw. Doing so allows the screw to be driven
>>indirectly and velocity/torque/rotational-precision
>>to be scaled via simple friction coupling without
>>introducing slippage error.
> 
> 
> A DC gearhead motor is best for lead screw applications as it allows 
> the screw to be driven "directly" without any complications provided 
> speed and torque requirements are met. It doesn't get much simpler 
> than this;
> 
> http://www.pbase.com/eldata/image/45200286
> 
> http://www.pbase.com/eldata/image/45200509

Depends upon what one is trying to accomplish.
A direct coupled gear head approach is certainly
straightforward if somewhat of an expensive option.

Then again there is the backlash in the gear train
which needs to be factored-in when making fine
adjustments via changing direction to minimize
over travel.  There is also the need to align both
axises within a given tolerance to minimize bearing
wear.  These tend to be non-issues with frictional
coupling.

BTW what type of rotational velocity are you
getting out of this setup?  Actually the linear
velocity is of more interest.  Hard to gauge from
the jpg but the lead screw appears to be ~10TPI?

-- 
uhmgawa@...        www.gnu.org

> A direct coupled gear head approach is certainly
> straightforward if somewhat of an expensive option.

I wouldn't consider $15.00 - $30.00 for a gearhead motor expensive. 
You can get one with encoder for $29.00.

> Then again there is the backlash in the gear train

A fraction of the lead screw backlash. Better than using belts 
(toothed or friction). Backlash at reversal is handled in software. 
This is a drilling setup, spurious movement in the backlash dead 
zone like with milling or routing isn't an issue.

>There is also the need to align both
>axises within a given tolerance to minimize bearing
>wear.

Use shims and a dial indicator or use a flexible coupling. I prefer 
proper alignment and rigid couplings. Yeah, machining and metal 
working savvy is beneficial.

>Actually the linear velocity is of more interest. 

20 IPM for a 12" X 8" travel. The milling table weighs 64 lbs. 
Drills 18 holes per minute (6" X 4" matrix).


--- In Homebrew_PCBs@yahoogroups.com, uhmgawa <uhmgawa@m...> wrote:
> derekhawkins wrote:
> >>For lead screw applications I'd keep the encoder on
> >>the screw. Doing so allows the screw to be driven
> >>indirectly and velocity/torque/rotational-precision
> >>to be scaled via simple friction coupling without
> >>introducing slippage error.
> > 
> > 
> > A DC gearhead motor is best for lead screw applications as it 
allows 
> > the screw to be driven "directly" without any complications 
provided 
> > speed and torque requirements are met. It doesn't get much 
simpler 

> > than this;
> > 
> > http://www.pbase.com/eldata/image/45200286
> > 
> > http://www.pbase.com/eldata/image/45200509
> 
> Depends upon what one is trying to accomplish.
> A direct coupled gear head approach is certainly
> straightforward if somewhat of an expensive option.
> 
> Then again there is the backlash in the gear train
> which needs to be factored-in when making fine
> adjustments via changing direction to minimize
> over travel.  There is also the need to align both
> axises within a given tolerance to minimize bearing
> wear.  These tend to be non-issues with frictional
> coupling.
> 
> BTW what type of rotational velocity are you
> getting out of this setup?  Actually the linear
> velocity is of more interest.  Hard to gauge from
> the jpg but the lead screw appears to be ~10TPI?
> 
> -- 
> uhmgawa@m...        www.gnu.org

derekhawkins wrote:
> Use shims and a dial indicator or use a flexible coupling. I prefer 
> proper alignment and rigid couplings. Yeah, machining and metal 
> working savvy is beneficial.

A flexible coupling will introduce error as
the encoder in this example is on the motor shaft
rather than the lead screw.

Coaxial alignment of two shafts is at best an
approximation irrespective of the machinist's
savvy or methods.

-- 
uhmgawa@...        www.gnu.org

>A flexible coupling will introduce error as
>the encoder in this example is on the motor shaft

That's why rigid couplings are preferred and being used in this case.

>Coaxial alignment of two shafts is at best an
>approximation irrespective of the machinist's
>savvy or methods.

Have you really looked at your advice and comments?....They're a 
hodge-podge of discrepancies....On one hand you're suggesting that 
one make their own encoder including the disk itself, and now here 
you seem to be critical of shafts aligned within the runout of the 
shafts themselves...Yes, plus or minus one or two thou is an 
approximation...In terms of specifications and tolerances, we live 
in an approximate world....LOL!

--- In Homebrew_PCBs@yahoogroups.com, uhmgawa <uhmgawa@m...> wrote:
> derekhawkins wrote:
> > Use shims and a dial indicator or use a flexible coupling. I 
prefer 

> > proper alignment and rigid couplings. Yeah, machining and metal 
> > working savvy is beneficial.
> 
> A flexible coupling will introduce error as
> the encoder in this example is on the motor shaft
> rather than the lead screw.
> 
> Coaxial alignment of two shafts is at best an
> approximation irrespective of the machinist's
> savvy or methods.
> 
> -- 
> uhmgawa@m...        www.gnu.org

> A DC gearhead motor is best for lead screw applications as it allows 
> the screw to be driven "directly" without any complications provided 
> speed and torque requirements are met. It doesn't get much simpler 
> than this;


Sure it does, eliminate the ball screw and coupling and switch over to 
a linear motor - that is much simpler!  However, in reality, much more 
expensive too :-(

derekhawkins wrote:
> Have you really looked at your advice and comments?

Yes.

> They're a hodge-podge of discrepancies....

They were intended to be a set of alternatives.

> On one hand you're suggesting that one make their own encoder
> including the disk itself

I offered that as a realistic possibility.  It is.
Not everyone in this forum is able to reach for
a shaft encoder off the shelf.  If you are able
to do so that's swell.

> and now here you seem to be critical of shafts
> aligned within the runout..

I didn't say it was not possible.  But rather was
contrasting the pros/cons of the available approaches.
You are obviously quite proud of your handiwork and
I did not intend to pooh-pooh your implementation.

On a general note, I'd like to point out the "homebrew"
nature of this group.  The intention here is to exchange
ideas and resources with the intent of finding viable
solutions within the means of the "midnight engineer".

I think most here would agree off-the-shelf
preengineered components when available make
for a fine solution.  But this is not always feasible
and perfectly suitable homebrew approaches do exist.
Finding them is the collective goal here.

-- 
uhmgawa@...        www.gnu.org

>Sure it does, eliminate the ball screw and coupling and switch over 
>to a linear motor - that is much simpler! However, in reality, much 
>more expensive too :-(

So you consider $25.00 for a 250 RPM gearhead motor with 250 oz-in 
stall torque expensive too? That's what was paid for the one's I'm 
using.


--- In Homebrew_PCBs@yahoogroups.com, "lcdpublishing" 
<lcdpublishing@y...> wrote:
> > A DC gearhead motor is best for lead screw applications as it 
allows 
> > the screw to be driven "directly" without any complications 
provided 
> > speed and torque requirements are met. It doesn't get much 
simpler 
> > than this;
> 
> 
> Sure it does, eliminate the ball screw and coupling and switch 
over to 
> a linear motor - that is much simpler!  However, in reality, much 
more 
> expensive too :-(

--- In Homebrew_PCBs@yahoogroups.com, "derekhawkins" 
<derekhawkins@y...> wrote:
> >Sure it does, eliminate the ball screw and coupling and switch over 
> >to a linear motor - that is much simpler! However, in reality, much 
> >more expensive too :-(
> 
> So you consider $25.00 for a 250 RPM gearhead motor with 250 oz-in 
> stall torque expensive too? That's what was paid for the one's I'm 
> using.

No, I don't think $25.00 for that motor is too expensive.  I am saying 
that your way is not the MOST simple way to create linear motion.  A 
linear motor has no moving parts- it doesn't get any simpler than that.