AVR-Chat

Hi,

I am a 15 year old high school student using a tiny 26 on an odyssey
of the mind project (we have a competition next week :) ). we are
building a vehicle based on an 18 volt drill motor. i have built an H
bridge based on the design on page 263 of  "The Robot Builder's
Bonanza". I am using BUZ11 n channel mosfets. I am also using a k2232
(also a n channel mosfet) to drive the h bridge between ground and the
rest of the H bridge. i am using diodes from source to drain. I am
using PWM on the avr to control speed at just under 4 kHz and about a
40% duty cycle. everything works well in testing using a 7.2 V nicad
and a small radio shack motor. when i try to connect to the drill
motor and the 18V motor the cmos buffer i am using between the avr and
the h bridge ( 74hc244e i think) smokes and has visible melting after
the circuit works for a couple seconds. The drill motor draws just
under 10 amps under load.  i am going to try adding a capacitor
tomorrow to reduce electrical noise but i was wondering if any one
else has any suggestions for me?


Thanks in advance for any suggestions

Ben Matthews
New York, USA

At 09:11 PM 2/18/05 -0500, Ben Matthews wrote:
>I am a 15 year old high school student using a tiny 26 on an odyssey
>of the mind project (we have a competition next week :) ). we are
>building a vehicle based on an 18 volt drill motor. i have built an H
>bridge based on the design on page 263 of  "The Robot Builder's
>Bonanza". I am using BUZ11 n channel mosfets. I am also using a k2232
>(also a n channel mosfet) to drive the h bridge between ground and the
>rest of the H bridge. i am using diodes from source to drain. I am
>using PWM on the avr to control speed at just under 4 kHz and about a
>40% duty cycle. everything works well in testing using a 7.2 V nicad
>and a small radio shack motor. when i try to connect to the drill
>motor and the 18V motor the cmos buffer i am using between the avr and
>the h bridge ( 74hc244e i think) smokes and has visible melting after
>the circuit works for a couple seconds. The drill motor draws just
>under 10 amps under load.  i am going to try adding a capacitor
>tomorrow to reduce electrical noise but i was wondering if any one
>else has any suggestions for me?


Do you have a link to the schematic and layout we can look at?  It will be 
hard to offer much more than generalities without that.

The first thing that occurs to me though is to ask what you are using to 
drive the FETs.  IE what is between the MOSFET gates and the hc244?  I'd be 
particularly concerned about the high sides.  I suspect the capacitor is 
unlikely to help. Depending on the problem and where you place it, it may 
make things worse.

A couple of other questions
         Why are you using different FETs for the high and low sides?  They 
have to deal with the same voltage and currents in most H bridges.
         Why are you adding diodes?  The FETs integral body diode is rated 
as high as the FETs forward capacity (normally for power MOSFETs 
anyway).  The diodes will only take up room and as a result may give you 
inductance problems.  It's a minor point in some respects and you won't 
want to try to change the layout at this stage anyway.
         Have you tried the circuit at different voltages without the motor 
in place?  Use something like a 1 K resistor and measure the current 
draw.  You may be getting shoot through.
         Do you have access to an oscilloscope (and know how to use 
it)?  It sounds like it may be a switching problem and an oscilloscope will 
make a big difference in trying to diagnose something like that.

Hmm, reading back through your note I realize I'm not clear as to what 
voltage you tested the 18V motor at.  Was that with the 7.2V NiCd as well?

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

At 09:11 PM 2/18/05 -0500, Ben Matthews wrote:
>I am a 15 year old high school student using a tiny 26 on an odyssey
>of the mind project (we have a competition next week :) ). we are
>building a vehicle based on an 18 volt drill motor. i have built an H
>bridge based on the design on page 263 of  "The Robot Builder's
>Bonanza". I am using BUZ11 n channel mosfets. I am also using a k2232
>(also a n channel mosfet) to drive the h bridge between ground and the
>rest of the H bridge. i am using diodes from source to drain. I am
>using PWM on the avr to control speed at just under 4 kHz and about a
>40% duty cycle. everything works well in testing using a 7.2 V nicad
>and a small radio shack motor. when i try to connect to the drill
>motor and the 18V motor the cmos buffer i am using between the avr and
>the h bridge ( 74hc244e i think) smokes and has visible melting after
>the circuit works for a couple seconds. The drill motor draws just
>under 10 amps under load.  i am going to try adding a capacitor
>tomorrow to reduce electrical noise but i was wondering if any one
>else has any suggestions for me?


Another question occurred to me.  Where are you getting the supply for the 
logic portion of your setup (The AVR, HC244 etc..)?  Also where does the 
supply for the MOSFET gate drivers come from?

If they are being fed off the same 7.2V NiCd as the H-bridge are they diode 
protected against droop?  IE a diode from battery positive to the positive 
side of a large filter cap on the input of your power regulators.

If not there is a very good chance those supplies are being cannibalized to 
drive the H-bridge.  The resulting voltage drop could cause almost anything 
to happen on the H-bridge.  One of the more likely scenarios is the whole 
thing would turn into a high power oscillator at some frequency depending 
on pack resistance stall current, lead inductance and SW response times.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

--- In AVR-Chat@yahoogroups.com, Robert Adsett <subscriptions@a...> 
wrote:
> At 09:11 PM 2/18/05 -0500, Ben Matthews wrote:
> >I am a 15 year old high school student using a tiny 26 on an 
odyssey
> >of the mind project (we have a competition next week :) ).

This is Ben's father.  A little more info.

Ben's coach is the person in charge of the program and is about to 
lose it.  She is coaching 3 teams (balsa, technical, and vehicle) 
and all of them are having major problems.  I am trying to help the 
vehicle and technical teams.  I used to do analog hardware (Nike 
Herc missle) but have been a software guy for over 20 years now. 

The two types of MOSFETs are in use because I made a Digikey order 
for the first batch.  Between Ben's team and another with an 
impulsive teenager the supply went down in a hurry.  The local 
electronics surplus store has BUZ11s in stock and they looked 
adequate.  Thus the mix of types.

The buffer outputs are connected directly to the MOSFET gates.

The diodes are there for spike supprssion as called for in the book.

He has access to a 65MHz dual trace scope and knows the basics of 
using it.  I can get him through the rest of it.

Logic power is supplied by a 4 cell nicad pack.  Drill power is the 
15 cell pack that came with the unit.  Negative is common to both.

The drill is the Harbor Freight 18v unit.  About $17 US complete 
(cheap junk).  There are no noise supprssion caps on the motor.

He had blown a transistor earlier in the day from a backwards 
diode.  That one was found with low volts testing.

The 7.2 v pack was a nicad "9v" battery, about 120 mah.  I handed 
him that because it can't source much over a couple of amps. 

Oscillation is an interesting guess.  I hadn't thought of that one. 
The only way that I could see getting enough gate leakage to wipe 
the buffer was tranistor failure and since they were not getting 
warm it did not make a lot of sense to me.  I am guessing that when 
the failure occurs he is dumping one battery pack into the other.

Plan for today:
Check the transistors.
Put a 300-500 ohm resistor between the buffer and the MOSFET gate to 
limit current when the failure occurs.
Put an auto light bulb in place of the motor in the circuit (dummy 
load).  That will get rid of the motor electrical noise while 
testing.
Fuse it all at 5 amps
Test with 12 volts (or less).
I am going to look through the junk box for a low ohm resistor to 
toss in series with the battery as a limiter for testing.

Odyessey is a great program.  The vehicle and technical problems are 
really difficult this year.  We are in a small public school (100 
graduates/year) in upstate NY.  Lots of community involement for 
coaching and skills training.

Dave Matthews




> 
> Another question occurred to me.  Where are you getting the supply 
for the 
> logic portion of your setup (The AVR, HC244 etc..)?  Also where 
does the 
> supply for the MOSFET gate drivers come from?
> 
> If they are being fed off the same 7.2V NiCd as the H-bridge are 
they diode 
> protected against droop?  IE a diode from battery positive to the 
positive 
> side of a large filter cap on the input of your power regulators.
> 
> If not there is a very good chance those supplies are being 
cannibalized to 
> drive the H-bridge.  The resulting voltage drop could cause almost 
anything 
> to happen on the H-bridge.  One of the more likely scenarios is 
the whole 
> thing would turn into a high power oscillator at some frequency 
depending 
> on pack resistance stall current, lead inductance and SW response 
times.
> 
> Robert
> 
> " 'Freedom' has no meaning of itself.  There are always 
restrictions,
> be they legal, genetic, or physical.  If you don't believe me, try 
to

> chew a radio signal. "
> 
>                          Kelvin Throop, III

On Feb 19, 2005, at 8:00 AM, velo1_4mb wrote:

> Put a 300-500 ohm resistor between the buffer and the MOSFET gate to
> limit current when the failure occurs.

Not only that but it will provide a shunt which allows one to measure 
the current.

Most any time I think the circuit can stand it I put 100 ohm resistors 
inline. Provides a bit of protection but also provides a location I can 
easily measure voltage across the resistor to "see" the current. Also 
because I often draw the schematic, lay out the PCB, and have a few 
boards built before powering the circuit the first time.

--
David Kelly N4HHE, dkelly@HiWAAY.net
========================================================================
Whom computers would destroy, they must first drive mad.

Thanks for all the help so far. i am going to add the resistors in a
couple minutes, all my electronics stuff is over in our Odyssey of the
mind room. i have a hand drawn scematic which i scanned.
http://benlinuxisbest1.tripod.com/hbridge.pdf if any one wants to take
a look.

Ben Matthews

On Feb 19, 2005, at 8:00 AM, velo1_4mb wrote:

  

Put a 300-500 ohm resistor between the buffer and the MOSFET gate to
limit current when the failure occurs.
    

Not only that but it will provide a shunt which allows one to measure 
the current.

Most any time I think the circuit can stand it I put 100 ohm resistors 
inline. Provides a bit of protection but also provides a location I can 
easily measure voltage across the resistor to "see" the current. Also 
because I often draw the schematic, lay out the PCB, and have a few 
boards built before powering the circuit the first time.

--
David Kelly N4HHE, dkelly@HiWAAY.net
========================================================================
Whom computers would destroy, they must first drive mad.



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/AVR-Chat/

<*> To unsubscribe from this group, send an email to:
    AVR-Chat-unsubscribe@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 




  

At 02:00 PM 2/19/05 +0000, velo1_4mb wrote:


>--- In AVR-Chat@yahoogroups.com, Robert Adsett <subscriptions@a...>
>wrote:
> > At 09:11 PM 2/18/05 -0500, Ben Matthews wrote:
> > >I am a 15 year old high school student using a tiny 26 on an
>odyssey
> > >of the mind project (we have a competition next week :) ).
>
>This is Ben's father.  A little more info.
>
>Ben's coach is the person in charge of the program and is about to
>lose it.  She is coaching 3 teams (balsa, technical, and vehicle)
>and all of them are having major problems.  I am trying to help the
>vehicle and technical teams.  I used to do analog hardware (Nike
>Herc missle) but have been a software guy for over 20 years now.
>
>The two types of MOSFETs are in use because I made a Digikey order
>for the first batch.  Between Ben's team and another with an
>impulsive teenager the supply went down in a hurry.  The local
>electronics surplus store has BUZ11s in stock and they looked
>adequate.  Thus the mix of types.


OK, but be aware that may effect the switching.  The two types may need 
different gate shaping circuits.


>The buffer outputs are connected directly to the MOSFET gates.


Um, how are you getting the 10V to drive the gates then?  I don't believe 
either of these devices are logic level devices and while they will sort of 
work at 5V it is asking for trouble.  More importantly the high sides need 
to run at 10V ABOVE the input voltage to turn on,  If they are only running 
at 5V you will definitely have problems.  The gate drive issue, especially 
the high sides, is what worries me the most about what I've heard of this 
setup so far.  Is their a possibility the original circuit called for P 
type FETs on the high side?  That might explain the odd drive 
circuitry.  Even in that case though driving a high side gate off of more 
than 5 or 6 Volts is asking for trouble.  Truthfully I'm rather surprised 
the high sides turn on for any appreciable amount of time.

>The diodes are there for spike supprssion as called for in the book.

Sounds like someone designed the circuit who knew a little bit but not 
enough.  The diodes at best are unnecessary for a MOSFET H-bridge.  The 
MOSFET integral body diode will work just fine.  In some case adding diodes 
will cause problems (including blown power sections).

>He has access to a 65MHz dual trace scope and knows the basics of
>using it.  I can get him through the rest of it.

Definitely needed.  With a low enough load not to cause problems you should 
measure
         - the low side gate
         - the high side gate
         - The gate drive power supply -what voltage is the gate drive 
power?, if it's less than 10V I'd be worried although it might work at 
these low currents.
         - The logic power supply

Both should get 10V with respect to the source of the FET they are driving 
to turn on. You can probably get away with 7V although I'd rather see 12-15 
V on a non-logic level gate drive.  On the off portion of the PWM they 
should be less than 1V.

Check the low and high side turn on and turn off relationships to make sure 
you are not getting shoot through.


>Logic power is supplied by a 4 cell nicad pack.  Drill power is the
>15 cell pack that came with the unit.  Negative is common to both.
>
>The drill is the Harbor Freight 18v unit.  About $17 US complete
>(cheap junk).  There are no noise supprssion caps on the motor.

Noise suppression caps really shouldn't matter very much

>He had blown a transistor earlier in the day from a backwards
>diode.  That one was found with low volts testing.
>
>The 7.2 v pack was a nicad "9v" battery, about 120 mah.  I handed
>him that because it can't source much over a couple of amps.
>
>Oscillation is an interesting guess.  I hadn't thought of that one.
>The only way that I could see getting enough gate leakage to wipe
>the buffer was tranistor failure and since they were not getting
>warm it did not make a lot of sense to me.  I am guessing that when
>the failure occurs he is dumping one battery pack into the other.


Remember the high side reference point.


>Plan for today:
>Check the transistors.
>Put a 300-500 ohm resistor between the buffer and the MOSFET gate to
>limit current when the failure occurs.

Careful, that might make things worse by slowing down the edges and 
increasing shoot through.  Also note that the reference point for the high 
side drive is the high side FET source not ground.  That means its 
reference will vary from 0V (or slightly below) to slightly above B+.  That 
means the output drive on the high side must float with it (and yes I think 
I've said this multiple times, it's important).  The off on the high side 
must float between 0V and perhaps slightly above B+. It should never be 
more than approximately a volt below the high side source voltage.

>Put an auto light bulb in place of the motor in the circuit (dummy
>load).  That will get rid of the motor electrical noise while
>testing.

Good idea if the bulb will draw less current than the motor.

>Fuse it all at 5 amps
>Test with 12 volts (or less).
>I am going to look through the junk box for a low ohm resistor to
>toss in series with the battery as a limiter for testing.

Also a good idea.

Another troubleshooting suggestion.  Hookup only one 1/2H rather than the 
full H.  The other side of the load you are using for testing can go to 
either B+ or B- depending on the test.  For example to test the high side 
turn on hook up the load (say 1K to start to B+ and one of the 1/2 
H's.  Then scope the high side gate with respect to the High side source 
(which will be close to B+) and see how it behaves.  In fact this should 
probably be one of the first tests you do.  This will allow you to verify 
the high side operation at low currents (particularly if you use a current 
limiting resistor on the B+.


One more question (for now), do you have a large capacitor across B+ close 
to the H-bridge (physically close with short trace lengths)?

>Odyessey is a great program.  The vehicle and technical problems are
>really difficult this year.  We are in a small public school (100
>graduates/year) in upstate NY.  Lots of community involement for
>coaching and skills training.


I just wish I had something like that available at that age.


Good luck,

I'll try to help you along.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

At 10:49 AM 2/19/05 -0500, Robert Adsett wrote:
>Um, how are you getting the 10V to drive the gates then?  I don't believe
>either of these devices are logic level devices and while they will sort of
>work at 5V it is asking for trouble.  More importantly the high sides need
>to run at 10V ABOVE the input voltage to turn on,  If they are only running
>at 5V you will definitely have problems.  The gate drive issue, especially
>the high sides, is what worries me the most about what I've heard of this
>setup so far.  Is their a possibility the original circuit called for P
>type FETs on the high side?  That might explain the odd drive
>circuitry.  Even in that case though driving a high side gate off of more
>than 5 or 6 Volts is asking for trouble.  Truthfully I'm rather surprised
>the high sides turn on for any appreciable amount of time.

That should be driving the power section off of more than 5 or 6 Volts is 
asking for trouble.  You want the gate drive voltage to be 10-15V above 
it's reference.

>One more question (for now), do you have a large capacitor across B+ close
>to the H-bridge (physically close with short trace lengths)?

That's the B+ for the power section, not the logic section BTW

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

Doug: 
> I am sure that one of the hardware gurus will be able to clarify
> things more and correct me if wrong, but on first looks the circuit in
> the pdf looks incorrect to me if all the mosfets are the same. If I
> remember correctly, the idea of the H configuration is to have one of
> the "top" mosfets on one side of the motor and one of the "bottom"
> mosfets on the other side on with the other two off and vice versa for
> the opposite direction of rotation. As things stand, my interpretation
> of the circuit would be that the top control line just switches the
> two "top" mosfets and the bottom line just switches the "bottom"
> mosfets and there is no complementary action with regards to driving
> the motor.
Looks that way, doesn't it?  If they're arranged like this:

1---2
|-m-|
4---3

The gates of 1 and 3 need to be driven together, as should 2 and 4.




Richard

---
"Iz dana u dan ona dolazi i odlazi u talasima"
   http://www.van-gogh.co.yu/

At 07:33 PM 2/19/05 +0000, Doug Locke wrote:
>I am sure that one of the hardware gurus will be able to clarify things 
>more and correct me if wrong, but on first looks the circuit in the pdf 
>looks incorrect to me if all the mosfets are the same. If I remember 
>correctly, the idea of the H configuration is to have one of the "top" 
>mosfets on one side of the motor and one of the "bottom" mosfets on the 
>other side on with the other two off and vice versa for the opposite 
>direction of rotation. As things stand, my interpretation of the circuit 
>would be that the top control line just switches the two "top" mosfets and 
>the bottom line just switches the "bottom" mosfets and there is no 
>complementary action with regards to driving the motor.
>
>Doug.
>
>
>-----Original Message-----
>From: Ben Matthews []
>Sent: 19 February 2005 14:51
>To: AVR-Chat@yahoogroups.com
>Subject: Re: [AVR-Chat] Re: Motor Control Problems - more info
>
>Thanks for all the help so far. i am going to add the resistors in a
>couple minutes, all my electronics stuff is over in our Odyssey of the
>mind room. i have a hand drawn scematic which i scanned.
><http://benlinuxisbest1.tripod.com/hbridge.pdf>http://benlinuxisbest1.tripod.com/hbridge.pdf 
>if any one wants to take
>a look.

For some reason the original post hasn't shown up here yet.  The vagaries 
of Yahoo and e-mail  I expect.  In any case this circuit will not work.  It 
also strikes me as too complex but that's a separate issue.

- Problem 1:
         Unless there is a couple of invertors you haven't drawn then what 
you have is a single 1/2 H with two FETs in parallel tied to both sides of 
the load (your motor).  The only way the motor gets current is if the FETs 
don't turn on at the same time  when commanded to.  Given the lack of gate 
resistors oscillation between the FETs  is a possibility.

- Problem 2:
         None of the Buz11's have sufficient drive voltage, unless your 
CMOS buffer is a quadruple high side FET driver.  That seems 
unlikely.  From an earlier post I believe you had a simple logic chip in there.

I believe you are on a pretty tight schedule.  The easiest way to solve 
your problem maybe to use relays instead of FETs to provide the H-bridge 
part of your design.  I can do up a quick schematic to give you a direction 
if you would like.   You would then still use the K2232 for speed 
control.  You could probably find automotive relays with sufficient 
capabilities.  They would be large than the FETs but a lot of the drive 
problems would be gone.  It would be a simple circuit to hook up.


Otherwise you need to do a complete re-design of the drive circuitry.  I'd 
use four high side drivers (IR2125s come to mind) and probably some charge 
pump circuitry to maintain the gate voltage.  Actually if I was doing that 
much work I'd eliminate the K2232 and use a pair of 1/2H driver like the 
2110 and as long as you don't require a full 100% in either direction no 
additional charge pump circuit would be required.  I'd really recommend you 
go the relay route and then look at an all MOSFET version after you get a 
relay version working if you still want to.

I suspect you also need a supply for you gate drive.  You want 10 - 15V for 
that and if you are driving off of CMOS logic you probably only have 5V

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

>  For some reason the original post hasn't shown up here yet.  The vagaries 
>  of Yahoo and e-mail  I expect.  In any case this circuit will not work.  It
>  also strikes me as too complex but that's a separate issue.
>  
>  - Problem 1:
>           Unless there is a couple of invertors you haven't drawn then what 
>  you have is a single 1/2 H with two FETs in parallel tied to both sides of 
>  the load (your motor).  The only way the motor gets current is if the FETs 
>  don't turn on at the same time  when commanded to.  Given the lack of gate 
>  resistors oscillation between the FETs  is a possibility.
>  
>  - Problem 2:
>           None of the Buz11's have sufficient drive voltage, unless your 
>  CMOS buffer is a quadruple high side FET driver.  That seems 
>  unlikely.  From an earlier post I believe you had a simple logic chip in
> there.
>  
>  I believe you are on a pretty tight schedule.  The easiest way to solve 
>  your problem maybe to use relays instead of FETs to provide the H-bridge 
>  part of your design.  I can do up a quick schematic to give you a direction
>  if you would like.   You would then still use the K2232 for speed 
>  control.  You could probably find automotive relays with sufficient 
>  capabilities.  They would be large than the FETs but a lot of the drive 
>  problems would be gone.  It would be a simple circuit to hook up.
>  
>  
>  Otherwise you need to do a complete re-design of the drive circuitry.  I'd 
>  use four high side drivers (IR2125s come to mind) and probably some charge 
>  pump circuitry to maintain the gate voltage.  Actually if I was doing that 
>  much work I'd eliminate the K2232 and use a pair of 1/2H driver like the 
>  2110 and as long as you don't require a full 100% in either direction no 
>  additional charge pump circuit would be required.  I'd really recommend you
>  go the relay route and then look at an all MOSFET version after you get a 
>  relay version working if you still want to.
>  
>  I suspect you also need a supply for you gate drive.  You want 10 - 15V for
>  that and if you are driving off of CMOS logic you probably only have 5V
>  
>  Robert
>  
>  " 'Freedom' has no meaning of itself.  There are always restrictions,
>  be they legal, genetic, or physical.  If you don't believe me, try to
>  chew a radio signal. "
>  
>                           Kelvin Throop, III
>  
>  
>  
>  Yahoo! Groups Sponsor 
>  
> 
> Get unlimited calls to 
> 
> U.S./Canada 
>  
>  
>  ________________________________
>  Yahoo! Groups Links
>  
> To visit your group on the web, go to:
> http://groups.yahoo.com/group/AVR-Chat/
>   
> To unsubscribe from this group, send an email to:
> AVR-Chat-unsubscribe@yahoogroups.com 
>   
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 


I actually was using relays originally but the ones we could find
couldn't deal with the current and welded quickly.
relooking at my drawing i agree that the gates are paired wrong, i do
have it built correctly, must have been caught as i soldered it
together.  i am going to prototype the circuit using higher voltage on
the gates, we both thought that the mosfets would work on 5 volts or
higher but looking at the datasheet again it makes sense although i
have also done circuity on another vehicle using only the k2232's
using 5 volts on the gate successfully but with out pwm and with out
an h bridge. using the oscilloscope i have noticed some noise coming
out of the buffer that may be turning on all four h bridge transistors
partially, about 1.5 volts at roughly my PWM frequency.

Well i am going to have dinner and try it out using some small npn
transistors suppling higer voltage to the gates,
by the way: the data sheets:
BUZ11's http://www.fairchildsemi.com/ds/BU/BUZ11.pdf
the buffer: http://focus.ti.com/lit/ds/symlink/cd74hct244.pdf
the odyssey of the mind website if any one is curious: www.odysseyofthemind.com

thanks for all the help so far
Ben Matthews

On Feb 19, 2005, at 10:06 AM, Robert Adsett wrote:

> That should be driving the power section off of more than 5 or 6 Volts 
> is
> asking for trouble.  You want the gate drive voltage to be 10-15V above
> it's reference.

Datasheet reference please for your 10-15V recommendation?

Rule of thumb which has worked well for me is to drive the gate with at 
least 2 volts more than the source. And remember the "source" is the 
output of the MOSFET while drain is the input.

Yes, drive the gate harder and "on" resistance will be lower. But the 
most bang-for-bucks is just over 2 volts.

One of my favorite parts is the IRF7307:
http://www.irf.com/product-info/datasheets/data/irf7307.pdf

Notice in figure 3 how they have about 9 amps flowing thru a 5A part, 
pulsing, with 2 volts GS. Figure 6 shows gate charge and you can see 
how the gate takes quite a charge before it rises above 2 volts.

I don't recommend the IRF7307 for this application as its SOIC-8 and 
therefore hard to prototype with. Also its only 20V.

Digikey labels the IRF7307 "logic level" but not the IRF7317:
http://www.irf.com/product-info/datasheets/data/irf7317.pdf

In Figure 5 of the IRF7317 datasheet we see a knee in the plot of "on" 
resistance vs. gate voltage at 3 volts.

One other point is that maximum Vgs is +-12 volts on these parts while 
the maximum Vds is 20. If one were to drive the gate +15 then it would 
be outside of its ratings.

The real reason for speaking up is to suggest the combination of P- and 
N-channel MOSFETs for the purpose of switching a higher voltage with a 
lower voltage.

--
David Kelly N4HHE, dkelly@HiWAAY.net
========================================================================
Whom computers would destroy, they must first drive mad.

At 05:12 PM 2/19/05 -0500, Ben Matthews wrote:
>I actually was using relays originally but the ones we could find
>couldn't deal with the current and welded quickly.


You do need a substantial relay.

>relooking at my drawing i agree that the gates are paired wrong, i do
>have it built correctly, must have been caught as i soldered it
>together.  i am going to prototype the circuit using higher voltage on
>the gates, we both thought that the mosfets would work on 5 volts or
>higher but looking at the datasheet again it makes sense although i
>have also done circuity on another vehicle using only the k2232's
>using 5 volts on the gate successfully but with out pwm and with out
>an h bridge. using the oscilloscope i have noticed some noise coming
>out of the buffer that may be turning on all four h bridge transistors
>partially, about 1.5 volts at roughly my PWM frequency.


Please note that the voltage required on the gate is referenced to the 
source so that worst case for all the BUZ11's is B+ + 10V or approx. 28V 
with a 18V battery pack!  Not only that but that potential MUST float with 
the source voltage to avoid blowing the gate when the source changes from 
18V to 0V.  It will take more than just a transistor.  Although a 
transistor may be enough for the K2232.

>Well i am going to have dinner and try it out using some small npn
>transistors suppling higer voltage to the gates,
>by the way: the data sheets:
>BUZ11's http://www.fairchildsemi.com/ds/BU/BUZ11.pdf
>the buffer: http://focus.ti.com/lit/ds/symlink/cd74hct244.pdf
>the odyssey of the mind website if any one is curious: 
>www.odysseyofthemind.com
>
>thanks for all the help so far
>Ben Matthews

You are entirely welcome Ben

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

At 04:29 PM 2/19/05 -0600, David Kelly wrote:
>On Feb 19, 2005, at 10:06 AM, Robert Adsett wrote:
>
> > That should be driving the power section off of more than 5 or 6 Volts
> > is
> > asking for trouble.  You want the gate drive voltage to be 10-15V above
> > it's reference.
>
>Datasheet reference please for your 10-15V recommendation?


Well take a look at the BUZ11's the OP is using as an instance.  The 
threshold voltage is 4 Volts.  RDSon is 40mOhm with a gate to source 
voltage of 10V.  That rises dramatically at 5V and it would be close to 
impossible to get reasonable currents through the device at that voltage, 
even 10 A takes the typical resistance well over 150m Ohms (it's actually 
off the chart, see figure 7, I don't think you could switch 10A on this 
device with a 5V gate).  Add in a little bit of noise and drops in the 
driver...

You also want to turn the FETs on quickly, that's usually easier to do with 
a larger supply voltage since a little droop from the initial current 
demand can be accommodated without falling below the threshold voltage.

Most drive circuits I've worked with use a 12-15 V gate drive supply for 
all of those reasons

>Rule of thumb which has worked well for me is to drive the gate with at
>least 2 volts more than the source. And remember the "source" is the
>output of the MOSFET while drain is the input.


That's way too low for consistently driving a standard power 
MOSFET.  That's low even for a lot of logic level MOSFETs, many have 
maximum thresholds of 2V (some are 2.5 or 3 IIRC).  Some of the newer logic 
level FETs might get away with that but I wouldn't want to try it on any of 
the logic level power FETs I've used.

>Yes, drive the gate harder and "on" resistance will be lower. But the
>most bang-for-bucks is just over 2 volts.


Don't drive it hard enough and you are either in the linear region or it 
doesn't turn on at all.

>One of my favorite parts is the IRF7307:
>http://www.irf.com/product-info/datasheets/data/irf7307.pdf


Well if you want to use a small logic level part :)  One of the devices I 
use at the moment is an ST 
part  http://www.st.com/stonline/books/pdf/docs/9231.pdf  They control a 
little more power ;)


>Notice in figure 3 how they have about 9 amps flowing thru a 5A part,
>pulsing, with 2 volts GS. Figure 6 shows gate charge and you can see
>how the gate takes quite a charge before it rises above 2 volts.


That's the normal turn-on knee.  The device isn't usually considered fully 
on until you rise above that point.

>I don't recommend the IRF7307 for this application as its SOIC-8 and
>therefore hard to prototype with. Also its only 20V.


Not to mention it won't handle 10A continuously.


>Digikey labels the IRF7307 "logic level" but not the IRF7317:
>http://www.irf.com/product-info/datasheets/data/irf7317.pdf


Regardless of what Digikey calls it, with a 0.7V threshold it's a logic 
level FET.  The biggest issue I'd have with these is getting them off and 
keeping them off.  I've seen problems even with higher threshold 
devices.  I notice the spec appears incomplete, I don't see the max 
threshold.  I may just have missed it I didn't spend much time looking.

>In Figure 5 of the IRF7317 datasheet we see a knee in the plot of "on"
>resistance vs. gate voltage at 3 volts.
>
>One other point is that maximum Vgs is +-12 volts on these parts while
>the maximum Vds is 20. If one were to drive the gate +15 then it would
>be outside of its ratings.


That's not unusual for logic level parts.

>The real reason for speaking up is to suggest the combination of P- and
>N-channel MOSFETs for the purpose of switching a higher voltage with a
>lower voltage.


That would work well for the high side FETs.  It won't work for the low 
side FETs in the H-bridge though.  They will see 0 - 18V on the source as 
the K2232 PWMs.  Hmm, you might actually get away with that since no 
current flows as the 2232 turns off anyway.  Funny things might happen at 
the gate though.  You would end up turning one of the low size BUZ11 on and 
off with the K2232 (that could get interesting), the other BUZ11 could get 
a negative gate voltage.  Ugly.  It would really make more sense to PWM the 
BUZ11's and remove the K2232.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

Hi again,

after an evening of experimentation i think we have found something
that may work:
we used a test circuit with one BUZ11, a 2N2904 npn transistor the
light bulb from before and a .3 ohm resistor in series with the bulb.
the npn transistor drives the buz11 with about an 18V gate voltage and
the buz11 drives the bulb. Dad drew the schematic and i put it at
http://benlinuxisbest1.tripod.com/testcircuit.pdf  . I drew a
schematic of an h bridge that i  think will work based on this
experiment, it is at http://benlinuxisbest1.tripod.com/hbridge2.pdf
(sorry it got a little messy). it seems like the problem was that the
gate voltage was too low, as every one said. Do you think this new h
bridge design will work? should we run the drill on a couple volts
less than the gate voltage? (16V on the drill, 18V on the gate
maybe?).

I didn't draw in the other k2232 drive mosfet, i would rather leave it
in but i can pull it out i think (tiny26 's can do 2 channels of PWM
right? as i have it now i have portB1 doing what i want and portB0
doing the inverse).

Thanks again for all your help, 
Ben Matthews

At 09:06 PM 2/19/05 -0500, Ben Matthews wrote:

>Hi again,
>
>after an evening of experimentation i think we have found something
>that may work:
>we used a test circuit with one BUZ11, a 2N2904 npn transistor the
>light bulb from before and a .3 ohm resistor in series with the bulb.
>the npn transistor drives the buz11 with about an 18V gate voltage and
>the buz11 drives the bulb. Dad drew the schematic and i put it at
>http://benlinuxisbest1.tripod.com/testcircuit.pdf  .

That looks good.  The turn on might be a little slow because of the 
resistor but as long as you don't try to run it too quickly it will work.

>I drew a
>schematic of an h bridge that i  think will work based on this
>experiment, it is at http://benlinuxisbest1.tripod.com/hbridge2.pdf
>(sorry it got a little messy). it seems like the problem was that the
>gate voltage was too low, as every one said. Do you think this new h
>bridge design will work? should we run the drill on a couple volts
>less than the gate voltage? (16V on the drill, 18V on the gate
>maybe?).


Closer but still not quite there.  Two problems.  In order of importance

Problem 1 - the high side drive.  The voltage to the gate on the high side 
needs to be 10V above the high side FET source.  Let's take the simplest 
case, 100% on (with the top left FET fully on and the bottom right FET 
fully on).  In that case the right hand side of the motor will be 0V and 
the left hand side will be 18V.  In that case the gate drive for the bottom 
right should be between 10V and 20V (if it goes over 20 you may damage the 
gate).  However, the gate drive for the top left FET should be 10-15 V 
above its source.  Since that is 18V the top left gate should be at least 
28V and no more than 38V.  As you can see there is no way for those values 
to be met by a single driver.  For an all N channel H-bridge you need four 
independent drivers.  The drivers can be driven in pairs but their outputs 
must be independent. It gets worse the voltage on the high side source can 
vary from 0 to 18V during operation (very quickly during PWM) so that 
driver must follow that voltage, it cannot simply have a fixed output 
voltage.  The easiest way to solve that (and keep your N-channel FETs) 
would be to use a high side driver such as an IR2125.  The second easy way 
to solve it would be to use a P Channel FET on the high side, P channel 
FETS are driven on by driving the gate below the drain.

Problem 2- speed.  May not be an issue, but you need to keep an eye 
open.  The high resistance you are using will slow down the MOSFET turn 
on.  Think of the gate as a capacitor, you essentially have an RC filter, 
the higher the resistance to the gate the slower the turn-on.  Not much of 
an issue if you just turn it on and leave it on.  A real problem if the PWM 
period is only a small multiple of the turn-on time.  Similar issues can 
occur on turn-off by the transistor should give it a pretty fast turn-off.

If you can get a few P channel power MOSFET that would probably be simpler 
than getting and using high side drivers but both will work.  If you had 
higher current or voltage requirements I'd say you had no choice but to use 
high side drivers but I think you should be able to find appropriate P 
channel MOSFET.  The driver for those will be similar to that for the N 
channel devices, actually the identical driver should work, except that a 
high input will turn the FET on instead of off.

You are making good progress here.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

Do you Yahoo!?
Yahoo! Search presents - Jib Jab's 'Second Term'

At 05:04 AM 2/20/05 -0800, sami saqqa wrote:
>try to use the L298 IC
>it is a simple to use two h bridg , here is the data sheet
><http://www.mip.sdu.dk/~n2018/Aflevering/L298.pdf>http://www.mip.sdu.dk/~n2018/Aflevering/L298.pdf


But that's only a 4A peak device.  The OP has a 10A load.

Robert


" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

At 09:06 PM 2/19/05 -0500, Ben Matthews wrote:
>Dad drew the schematic and i put it at
>http://benlinuxisbest1.tripod.com/testcircuit.pdf  . I drew a
>schematic of an h bridge that i  think will work based on this
>experiment, it is at http://benlinuxisbest1.tripod.com/hbridge2.pdf
>(sorry it got a little messy).

Ben, this is an aside from what you are doing at the moment.

Presumably you will want to document this at some point.  It would be 
worthwhile looking at one of the freely available Schematic/PCB packages 
for doing so.  They make it much easier to produce professional looking 
schematics.  One possibility is Eagle http://www.cadsoftusa.com/ which has 
a limited free version.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

On Sun, 20 Feb 2005 15:04:48 -0500, Robert Adsett
<subscriptions@aeolusdevelopment.com> wrote:
>  At 09:06 PM 2/19/05 -0500, Ben Matthews wrote:
>  >Dad drew the schematic and i put it at
>  >http://benlinuxisbest1.tripod.com/testcircuit.pdf   . I drew a
>  >schematic of an h bridge that i  think will work based on this
>  >experiment, it is at http://benlinuxisbest1.tripod.com/hbridge2.pdf 
>  >(sorry it got a little messy).
>  
>  Ben, this is an aside from what you are doing at the moment.
>  
>  Presumably you will want to document this at some point.  It would be 
>  worthwhile looking at one of the freely available Schematic/PCB packages 
>  for doing so.  They make it much easier to produce professional looking 
>  schematics.  One possibility is Eagle http://www.cadsoftusa.com/ which has 
>  a limited free version.
>  
>  Robert
>  
>  " 'Freedom' has no meaning of itself.  There are always restrictions,
>  be they legal, genetic, or physical.  If you don't believe me, try to
>  chew a radio signal. "
>  
>                           Kelvin Throop, III
>  
>  
>  
>  
>  Yahoo! Groups Sponsor 
>  
> 
> Get unlimited calls to 
> 
> U.S./Canada 
>  
>  ________________________________
>  Yahoo! Groups Links
>  
> To visit your group on the web, go to:
> http://groups.yahoo.com/group/AVR-Chat/
>   
> To unsubscribe from this group, send an email to:
> AVR-Chat-unsubscribe@yahoogroups.com 
>   
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 

Yes, i probably should, i have looked into a couple open source
packages a little but haven't had time to really try them out. I think
what i am going to do for now is scrap the h bridge for now and use a
servo to toggle a switch, if we win and go on to the state competition
maybe i will be able to build one, but hopefully the servo will work
for now. (p channel mosfets aren't available locally at a reasonable
price and we don't have time for a digikey order).  i am going to
pulse with one mosfet on the low side of the circuit so i shouldn't
run into problems. now all i need to do is build it and finish the
software.

thanks again for all the help
Ben Matthews

At 05:50 PM 2/20/05 -0500, Ben Matthews wrote:
>I think
>what i am going to do for now is scrap the h bridge for now and use a
>servo to toggle a switch, if we win and go on to the state competition
>maybe i will be able to build one, but hopefully the servo will work
>for now. (p channel mosfets aren't available locally at a reasonable
>price and we don't have time for a digikey order).  i am going to
>pulse with one mosfet on the low side of the circuit so i shouldn't
>run into problems. now all i need to do is build it and finish the
>software.

That sounds like it is likely to work.  Remember the freewheeling diode. 
You can use one of your BUZ11's for that if you need to, tie the gate to 
the source and you have a 30A diode.

Good Luck


" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III