MOTM

My MOTM-700 was working fine except that it was effecting other
modules through the power supply.

It caused an audible glitch in other modules.  This happened whenever a router
switched from one state to the other.  This happened even if the 700
was only connected by its power connector.  The problem was easily
heard  by using a 320 in the audible range.  Both routers within the 700 were
identical in effect, and the problem seemed twice as bad when both
routers switched at the same time.

Has anyone reported this or a similar problem??  One could easily be
fooled into thinking that a noise heard when switching audio is
caused by the VCA circuitry.  My guess is that there is a transient
from the LED's (perhaps both LED's are lit briefly exceeding the
capacity of either the 12V or 5V regulator).

C1 is a capacitor going to ground between the 5V regulator and LED's.
The one from my kit was a 10uF 50V cap. I didn't have any 50V caps
that would fit, and the schematic shows a 10uF 25V.  So in order to
provide more local storage, I put in in a 47uF 25V cap.  The problem
went away, but after a couple of minutes the 700 started behaving
erratically, and then stopped working altogether.  The cap burned
out.  It's now a perfect DC conductor sitting in my trash bin.

So I put in a 47uF 35V cap.  I had it running all night, and it is
still working fine.  No more PS glitch.

I still have no idea how 5 volts destroyed a 25V cap.

Are you absolutely certain beyond any doubt that you didn't have it
in backwards? Electrolytics leak more with temperature, and heat up
as they leak, so if they're ever put in backwards, you could have
from 5 minutes to several hundred minutes before they burn out.
Some of them burn out most spectacularly, waiting until the temp
is higher than the flashpoint of the electrolyte, in which case
miniature fireballs are launched from the printed circuit card.

-bobC

-----Original Message-----
From: media.nai@... [mailto:media.nai@...]
Sent: Sunday, April 27, 2003 9:08 AM
To: motm@yahoogroups.com
Subject: [motm] 700 problem and solution



My MOTM-700 was working fine except that it was effecting other
modules through the power supply.

It caused an audible glitch in other modules.  This happened whenever a
router
switched from one state to the other.  This happened even if the 700
was only connected by its power connector.  The problem was easily
heard  by using a 320 in the audible range.  Both routers within the 700
were
identical in effect, and the problem seemed twice as bad when both
routers switched at the same time.

Has anyone reported this or a similar problem??  One could easily be
fooled into thinking that a noise heard when switching audio is
caused by the VCA circuitry.  My guess is that there is a transient
from the LED's (perhaps both LED's are lit briefly exceeding the
capacity of either the 12V or 5V regulator).

C1 is a capacitor going to ground between the 5V regulator and LED's.
The one from my kit was a 10uF 50V cap. I didn't have any 50V caps
that would fit, and the schematic shows a 10uF 25V.  So in order to
provide more local storage, I put in in a 47uF 25V cap.  The problem
went away, but after a couple of minutes the 700 started behaving
erratically, and then stopped working altogether.  The cap burned
out.  It's now a perfect DC conductor sitting in my trash bin.

So I put in a 47uF 35V cap.  I had it running all night, and it is
still working fine.  No more PS glitch.

I still have no idea how 5 volts destroyed a 25V cap.





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

Yes, the resonance knob on the 490 is "very touchy".  Mine started to
self-oscillate at around 3.5 ticks.  The manual suggests changing R39
to 51K to 68K.  I had both, and went with 51.1K  At the high end it
starts to oscillate at  a resonance setting of about 4.5.  With the frequency
knob in the middle, it starts around 6.  It begins to oscillate
around 8.5. at the low end (around 90Hz).  Is there any way to get it
to self-oscillate below that??  The 440 goes down to around 20Hz.

Does it sound like a Moog??  While I can't do an AB comparison, it
certainly exhibits the safe goofy response to FM modulation, so I'm
happy :)  I don't know if it does the rising cutoff thing
-- it's very hard to tell misusing a 320 as a VCO (the 320 overshoots with
quick CV changes).

Also, in an attempt to minimize gratuitous drift, I pushed the two
transistors together with a bit of heat sink, and cinched them
together with a wire tie.  Now if certain people want to call me
names because that's what makes a Moog a Moog, I'll just have to live
with that :)

If I could change the 490 interface, I would have only one input jack
which would leave room for an input level knob.  Imho, passive mixers
aren't very useful unless you have a VCA for each input. I haven't
decided if I want to try lowering resistors to increase the input
level.

While I'm on the subject of swapping resistors.  Does anyone know a
good way to unsolder MOTM PCB's??  They seem to delaminate very
easily.  Just removing R39 removed the foil around both pads.  You
don't even want to know about the 700 (I had to use silver to get the
last repair to stick)  I've tried using a Paladin solder sucker
(spring-loaded vacuum) and a braid (Chemwick).  Any ideas??

> It begins to oscillate
> around 8.5. at the low end (around 90Hz).  Is there any way to get 
> it to self-oscillate below that??  

That's a limitation of the Moog filter design. Paul worked very hard 
to keep that behavior in to maintain the authentic sound. 

> While I'm on the subject of swapping resistors.  Does anyone know a
> good way to unsolder MOTM PCB's??  They seem to delaminate very
> easily. 

The first thing you should ALWAYS do is to cut the component out. 
Clip all the legs with a sharp pair of dykes. It's not worth saving 
that component worth a few cents (worst case a few dollars) to ruin 
your board. Then it becomes much simpler to work on one pad at a 
time. I like to heat from one side and use a vacuum sucker from the 
other. Properly timed, you can heat the pad just until the solder 
liquifies again, then suck the leg out the other side.

Moe

This thing:

http://www.howardelectronics.com/den-on/sc7000.html

  Don't let the price scare you (too much).  I've had one for 18 months 
now, and I cannot imagine doing solder re-work without it.

Crow
/**/

On Sun, 27 Apr 2003 media.nai@... wrote:

> While I'm on the subject of swapping resistors.  Does anyone know a good
> way to unsolder MOTM PCB's??  They seem to delaminate very easily.  
> Just removing R39 removed the foil around both pads.  You don't even
> want to know about the 700 (I had to use silver to get the last repair
> to stick)  I've tried using a Paladin solder sucker (spring-loaded
> vacuum) and a braid (Chemwick).  Any ideas??

At 11:21 AM -0700 4/27/03, Bob Colwell wrote:
>Are you absolutely certain beyond any doubt that you didn't have it
>in backwards?

Yes.

>Electrolytics leak more with temperature, and heat up
>as they leak, so if they're ever put in backwards, you could have
>from 5 minutes to several hundred minutes before they burn out.
>Some of them burn out most spectacularly, waiting until the temp
>is higher than the flashpoint of the electrolyte, in which case
>miniature fireballs are launched from the printed circuit card.

It died peacefully.

At 7:26 PM +0000 4/27/03, mate_stubb wrote:
>
>That's a limitation of the Moog filter design. Paul worked very hard
>to keep that behavior in to maintain the authentic sound.

OK :)

>The first thing you should ALWAYS do is to cut the component out.
>Clip all the legs with a sharp pair of dykes. It's not worth saving
>that component worth a few cents (worst case a few dollars) to ruin
>your board. Then it becomes much simpler to work on one pad at a
>time. I like to heat from one side and use a vacuum sucker from the
>other. Properly timed, you can heat the pad just until the solder
>liquifies again, then suck the leg out the other side.

That sounds like good advice.

At 3:36 PM -0400 4/27/03, The Old Crow wrote:
>
>
>   This thing:
>
>http://www.howardelectronics.com/den-on/sc7000.html
>
>   Don't let the price scare you (too much).  I've had one for 18 months
>now, and I cannot imagine doing solder re-work without it.

Thanks :)  That looks great, but I think it's way over my DIY budget.

In a message dated 4/27/2003 12:56:01 PM Pacific Daylight Time, 
media.nai@... writes:

> >The first thing you should ALWAYS do is to cut the component out.
> >Clip all the legs with a sharp pair of dykes. It's not worth saving
> >that component worth a few cents (worst case a few dollars) to ruin
> >your board. Then it becomes much simpler to work on one pad at a
> >time. I like to heat from one side and use a vacuum sucker from the
> >other. Properly timed, you can heat the pad just until the solder
> >liquifies again, then suck the leg out the other side.
> 
> That sounds like good advice.
> 
> At 3:36 PM -0400 4/27/03, The Old Crow wrote:
> >
> >
> >  This thing:
> >
> >http://www.howardelectronics.com/den-on/sc7000.html
> >
> >  Don't let the price scare you (too much).  I've had one for 18 months
> >now, and I cannot imagine doing solder re-work without it.
> 
> Thanks :)  That looks great, but I think it's way over my DIY budget.
> 


I know what you mean about that price. Even as much as I screw up, I can't 
justify a $400 eraser. 

Quite a bit less professional but MUCH more useful that the little bulbs: 
Larry and I have long been fans of the Radio Shack desoldering iron (it looks 
like an iron with a vacuum bulb in a sidecar kind of arrangement) -- I think 
it costs $10. I recently had to remove a capacitor from a double sided 
through plated PCB much like a MOTM (I think it was an Oakley). I think the 
iron touched each pad for 2-3 seconds before the solder was molten, and then 
just release the bulb -- then the component came out very easily.

The trick is (to put the component in right the first time! -- but failing 
that) to let the desoldering iron heat up before you try to use it -- I let 
it heat about five minutes. I have had a few problems with MOTM type PCBs 
(had to make a few attempts to get the component out), but I've never damaged 
a board.

JB

> That's a limitation of the Moog filter design. Paul worked very hard
> to keep that behavior in to maintain the authentic sound.

Well, it's not just a limitation, it's also a feature. Without this, it
won't sound like a Moog Modular filter.
Actually it's not only the point where oscillation starts; it's also
frequency dependent nonlinearities when the filter is on the
edge to oscillation, and similar things. It's all linked - you
don't get one feature without the other. (I must know, because
it was me who worked very hard to maintain the
authentic sound. (;->) )

(You might have heard of a different approach, as used in a certain
polysynth, where the Moog-type VCF does also oscillate at lower
frequencies, and where you are advised to scale the resonance
feedback loop with keyboard tracking or modulation or whatever,
if you want to get frequency dependent resonance.
This does _not_ produce the Moog Modular VCF sound, however,
because it dosn't even have the chance to emulate the frequency
dependent nonlinear effects.)

JH.

>I know what you mean about that price. Even as much as I screw up, I 
>can't justify a $400 eraser.

  :)

>Quite a bit less professional but MUCH more useful that the little 
>bulbs: Larry and I have long been fans of the Radio Shack 
>desoldering iron (it looks like an iron with a vacuum bulb in a 
>sidecar kind of arrangement) -- I think it costs $10. I recently had 
>to remove a capacitor from a double sided through plated PCB much 
>like a MOTM (I think it was an Oakley). I think the iron touched 
>each pad for 2-3 seconds before the solder was molten, and then just 
>release the bulb -- then the component came out very easily.

Since it's not regulated, I would worry that it would get too hot for 
PCB work.  I use an unregulated iron to desolder connectors, but then 
the extra heat is less of a risk.

>The trick is (to put the component in right the first time! -- but 
>failing that) to let the desoldering iron heat up before you try to 
>use it -- I let it heat about five minutes. I have had a few 
>problems with MOTM type PCBs (had to make a few attempts to get the 
>component out), but I've never damaged a board.

Yes, it seems odd that five holes (three on the 700, two on the 490) 
would delaminate.  Not that I modded any modules in my first assembly 
batch, but if I recall correctly, I accidently soldered a TL072 
upside down while building the 320, and that didn't ruin any of the 
pads.  Maybe the tip on my WTCPT went bad.  I've been using the same 
one for a few years, but the surface of it looks fine (I tin 
regularly, and whenever I switch from organic too no clean).  I don't 
have any kind of thermometer that goes up to 700F!!

I tried this tonight, with a 320 in the audible range to my 700, no
other inputs connected.  I could not hear any bleedthrough, even when I
took my 490 filter down to minimum frequency, giving silence.  I'm
guessing your problem was from one of two causes:

1) You are overtaxing your power supply.
2) You power wiring setup is exacerbating the problem.

John Loffink
jloffink@... 

The Microtonal Synthesis Web Site
http://www.microtonal-synthesis.com/

The Wavemakers Modular and Integrated Synthesizer Web Site
http://www.wavemakers-synth.com/

> 
> My MOTM-700 was working fine except that it was effecting other
> modules through the power supply.
> 
> It caused an audible glitch in other modules.  This happened whenever
a
> router
> switched from one state to the other.  This happened even if the 700
> was only connected by its power connector.  The problem was easily
> heard  by using a 320 in the audible range.  Both routers within the
700

> were
> identical in effect, and the problem seemed twice as bad when both
> routers switched at the same time.
> 
> Has anyone reported this or a similar problem??  One could easily be
> fooled into thinking that a noise heard when switching audio is
> caused by the VCA circuitry.  My guess is that there is a transient
> from the LED's (perhaps both LED's are lit briefly exceeding the
> capacity of either the 12V or 5V regulator).
> 
> C1 is a capacitor going to ground between the 5V regulator and LED's.
> The one from my kit was a 10uF 50V cap. I didn't have any 50V caps
> that would fit, and the schematic shows a 10uF 25V.  So in order to
> provide more local storage, I put in in a 47uF 25V cap.  The problem
> went away, but after a couple of minutes the 700 started behaving
> erratically, and then stopped working altogether.  The cap burned
> out.  It's now a perfect DC conductor sitting in my trash bin.
> 
> So I put in a 47uF 35V cap.  I had it running all night, and it is
> still working fine.  No more PS glitch.
> 
> I still have no idea how 5 volts destroyed a 25V cap.
> 
> 
>

At 12:14 AM -0500 4/28/03, John Loffink wrote:
>I tried this tonight, with a 320 in the audible range to my 700, no
>other inputs connected.  I could not hear any bleedthrough, even when I
>took my 490 filter down to minimum frequency, giving silence.

Perhaps the problem effects some 700's and not others.  Also, the
ear is quite sensitive to the tiniest changes in pitch and the 320
doesn't have the voltage regulation of a deliberate VCO.

I'm not sure what you mean by taking your 490 down to minimum
frequency.  There were no patch cables between any modules.

>  I'm guessing your problem was from one of two causes:
>1) You are overtaxing your power supply.

Not possible -- I still had the problem with only the 320 and 700
connected to the supply.

>2) You power wiring setup is exacerbating the problem.

Don't forget, the only thing I changed was the cap, and changing the
cap solved the problem.  Changing the cap wouldn't increase the
output of my power supply or change how it was wired.

--- In motm@yahoogroups.com, media.nai wrote:
>
> While I'm on the subject of swapping resistors.  Does anyone know a
> good way to unsolder MOTM PCB's??  They seem to delaminate very
> easily.  Just removing R39 removed the foil around both pads.  You
> don't even want to know about the 700...

In my experience, the desoldering technique Moe suggests usually works 
best - sacrifice the component, not the PCB.  (In other words, first 
remove the component by clipping the leads on the component side as 
close to the the body of the component as you can get.  After the 
component is gone use any number of desoldering techniques to remove 
what remains of the leads from the PCB.)  This is especially true with 
ICs - the more leads on a part, the less likely it is you will get all 
of the solder out to free the part and this is where you start to have 
"accidents" ;)  So better to cut the part out and remove the remaining 
leads one at a time.

For desoldering, solder suckers and solder wick both work fine if 
used correctly.  With the solder sucker you should be able to suck the 
solder and the remains of the lead out all in one operation.  With 
solder wick, you might need to grab what remains of the component lead 
with some hemostats, heat up the pad, and pull the lead out - then use 
the solder wick to clean the remaining solder out of the pad.  (With 
solder wick, it sometimes helps to add a little bit of solder to the 
pad before desoldering.  Most solder wick has flux so be sure and get 
the no clean type if you use it.)


As far as tools, the Edsyn Soldapullt is a pretty good solder sucker.  
(I'm pretty sure Paul recommends these!)  I have tried some others 
(including the blue aluminum one that Radio Shack sells) that didnt 
work as well...


Seth

--- In motm@yahoogroups.com, "strohs56k" <strohs@e...> wrote:
> 
> In my experience, the desoldering technique Moe suggests usually
> works best - sacrifice the component, not the PCB. (In other words,
> first remove the component by clipping the leads on the component
> side [...]

Just one more thing - this technique (obviously) doesn't work with 
components where you can't get at the leads on the component side of 
the PCB.  (Things like radial electrolytic capacitors, PCB mounted 
pots, etc. where the leads are hidden under the component.)  With such 
components, you are stuck with trying to get the solder out with the 
component in place.

If you end up with a component where just one lead refuses to break 
loose, often you can reapply solder to just that one pad and pull the 
component out while that pad is still hot.  (Often the difficult lead 
is the one with a heavy ground or power connection where the extra 
copper is sucking away heat from the joint.)  It should be easy to 
clean out that last pad once the component is gone.


Seth

At 5:06 AM +0000 4/29/03, strohs56k wrote:
>--- In motm@yahoogroups.com, "strohs56k" <strohs@e...> wrote:
>>
>  > In my experience, the desoldering technique Moe suggests usually
>>  works best - sacrifice the component, not the PCB. (In other words,
>>  first remove the component by clipping the leads on the component
>>  side [...]
>
>Just one more thing - this technique (obviously) doesn't work with
>components where you can't get at the leads on the component side of
>the PCB.  (Things like radial electrolytic capacitors, PCB mounted
>pots, etc. where the leads are hidden under the component.)  With such
>components, you are stuck with trying to get the solder out with the
>component in place.

That was the case with the 700 -- radial can and a pot.

>If you end up with a component where just one lead refuses to break
>loose, often you can reapply solder to just that one pad and pull the
>component out while that pad is still hot.  (Often the difficult lead
>is the one with a heavy ground or power connection where the extra
>copper is sucking away heat from the joint.)  It should be easy to
>clean out that last pad once the component is gone.

Thanks for all your advice :)

I agree 100% with Moe.  Minimize your dwell time at all costs!  This is what
gets the pads lifting off of the PCB.  When doing adjacent holes, allow time
for the first repair to totally cool down before moving to the next hole.

Although I usually don't have to use them, I have seen a couple of techniques
used with good success.

Once everything is trimmed off nicely.  Heat up the lead that is remaining in
the hole.  When the solder has melted, thwack the PCB against a piece of wood
perdendicular to the lead and the lead will come flying out.  Take the
appropriate precautions.

The second method is to leave a little of the lead sticking out of the hole. 
Place the tip of your soldering iron on one side of the lead.  When the solder
has melted, use a knife (X-actos are excellent for this to) and lift the lead
out.  The trick hear is to minimize the amount of thermal transfer between the
lead and the knife blade.

When do this, make sure there isn't a lot of air movement arount the solder. 
This means turn off fans and block any draft.

Ideally you want to get the lead out on the first try.  I said ideally.  But
doing it right the first time, means that there is a lot bigger chance the pad
will stay attached to the board.

Happy (De-)Soldering!
Shemp

> > While I'm on the subject of swapping resistors.  Does anyone know a
> > good way to unsolder MOTM PCB's??  They seem to delaminate very
> > easily. 
> 
> The first thing you should ALWAYS do is to cut the component out. 
> Clip all the legs with a sharp pair of dykes. It's not worth saving 
> that component worth a few cents (worst case a few dollars) to ruin 
> your board. Then it becomes much simpler to work on one pad at a 
> time. I like to heat from one side and use a vacuum sucker from the 
> other. Properly timed, you can heat the pad just until the solder 
> liquifies again, then suck the leg out the other side.


__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com