CMATMOD Brownie (BSIAB II) Low Volume

Started by Slight Return, February 26, 2018, 07:22:53 PM

Previous topic - Next topic

Slight Return

Just tested the resistor that was getting no signal on one leg with the audio probe. The multimeter does nothing when testing it. When I put it into resistance mode the screen just reads "1," and when I touch that resistor, it stays at "1."

I tested every other resistor in the circuit. My results are in the picture. A total of 3 resistors on the entire circuit get a readout. The rest, the MM just stays at "1" and does nothing.

I'll let the pic speak for itself. I don't have a schematic yet and while BSIAB II seems close, my Brownie has more capacitors and resistors in the circuit than are listed in the BSIAB II schematic, so I have reason to believe it's different.

I don't know what any of this means, but not getting a signal or readout can't be good.



@ Slowpoke

Just read your message as I was about to post this one.

I did break off half the flange of one of the eyelet through-holes. That was on Q3, I think on the Drain leg connection. On the back of the circuit board. It broke when I was attempting to push the molten solder out of the hole with a guitar string. The central shaft going through the board looked to be in tact. I was going to replace the eyelet until I realized how difficult that might be. I found this website and I wonder if I should give it a try:

http://pinballrehab.com/1-articles/solid-state-repair/board-rework-test/230-repairing-plated-through-holes

NOTE: The issues started not when I tried J113's, and not even when I broke the flange on the through-hole on Q3, but when I attempted to solder in sockets for the NTE457 transistors in Q4 and Q5. That is when the issues started. At the time, I hadn't even removed Q3 yet. I got the sockets soldered in, but when I tested them with the transistors loaded, it didn't work. I removed the sockets, but ever since then, I have been having major issues with the Brownie.

Moving on...

I read that J113's are close to MPF102's, and since they're still being made I figured I'd have a better shot trying them than NOS. My Brownie came stock with MPF102's in Q1 and Q2, and then J201's for the rest. I'd read somewhere that the guy who designed it thought that when it's run at 18v, it sounds "best" with all MPF102's. So I went for as close an equivalent as I could find after researching online.

I should've just left it with the two NTE457's in Q4 and Q5. I tried that originally and had zero issues, and very much liked the sound...MPF102's in Q1 and Q2, NTE457's in Q4 and Q5, and a J201 in Q3. I simply thought I wouldn't have any issues experimenting with different transistors, and then this whole debacle happened. The pedal is crackling, squealing, cutting in and out and doing all kinds of weird stuff. I thought it was working fine, but I had it cranked today and noticed a myriad of problems, on top of sounding like garbage.

A true nightmare, especially as I don't have a lot of money and investing in a Brownie hurt my wallet pretty badly!

I was not happy with the sound out of the box, but now I wonder if I would've been just as happy with the stock transistors, and simply clipping one of the .0022uF capacitors to open up the treble a bit. That was the most immediate difference I noticed, even more so than the lower gain transistors. Even still, I found that if I only used 1 capacitor, it DID matter what position I had it in, so even if I end up getting another Brownie eventually, I'm scared that I'll ruin it by clipping one of the caps.

I do have some 2N5458's coming in the mail. I read that those are close enough to 2N5457 to be interchangeable. At this point I'm having a hard time imagining that I didn't completely butcher this thing.

Thank you for the advice. I'll try that tomorrow. I'll report back with my findings. I haven't tried testing with the trimpot yet. I had just set everything by ear initially. I'm assuming no damage can be done by adjusting the trimpot with reckless abandon? As that's what I did when trying to set it up, and I have no idea what I left it on for a while. 

I did find that Q2 Drain has no output. No audio signal coming through when probed. Q1 has output on Gate, Source, and Drain, although Source is almost inaudible on Q1. Audible but noticeably weaker and much quieter than Gate and Drain. Q2 has output on Gate and Source but *nothing* on Drain. And all the other transistors after that have no response to the audio probe either, on any of the legs.

That was right at the start of the circuit, after a couple resistors following the Input. One resistor went to Q1, the other resistor to a capacitor...the resistor going to Q1 got audio on both legs, but the other resistor, only on one. As far as the audio probe revealed, the problems start extremely early in the circuit after the Input, no audio signal coming from one of the resistors *directly after the Input* being Culprit #1.

duck_arse

I can see on your built board two caps look like they are tantalums, the two 1uF's. please check you have them in the right way, as they are usually marked for (+) leg whereas aluminium electros usually mark (-) leg.

looking on the ggg circuit dia, any part that connects to ground or +9V will have no signal on that leg. the other end of those components I think will all show signal, to some degree (obviously except for the LED string).
" I will say no more "

Slight Return

Is there any way to tell which way to orient the tantalum caps? They are oriented in opposite directions to each other. Only one has a + marking directly next to the circle it's mounted on (the tantalum cap next to the trim pot). The other + sign is not directly adjacent to the other tantalum cap, and it appears to be intended for another component.

I would just try reversing the other one to the same orientation as the tantalum cap next to the trim pot, but I'm unsure of this. Is it safe to experiment with that or if I get it wrong could it short something out?

Slowpoke101

#23
If you look carefully at the solder pads that the tantalums are solder to you will see that one pad is square and the other is round. The square pad is the positive connection. If you look at the tantalums closely you will see a value and polarity marking. This polarity marking marks the positive lead which is the opposite of the markings on an electrolytic capacitor where the negative lead is usually the one that is marked.....Usually.

***Edit: I am aware that I just basically repeated D_A's advice (apologies) but I have found lately that people are having trouble with determining what the polarity of a tantalum capacitor is. It is not obvious and can be difficult to see. If you are building something or repairing something that needs a tantalum installed, the positive leg on a new tantalum is usually longer than the negative leg....assuming that the legs haven't already been trimmed.

Anyway, any progress on checking the trimpot and where the voltage for Q5's Drain has gotten to?
  • SUPPORTER
..

Slight Return

Quote from: Slowpoke101 on March 02, 2018, 04:18:09 PM
If you look carefully at the solder pads that the tantalums are solder to you will see that one pad is square and the other is round. The square pad is the positive connection. If you look at the tantalums closely you will see a value and polarity marking. This polarity marking marks the positive lead which is the opposite of the markings on an electrolytic capacitor where the negative lead is usually the one that is marked.....Usually.

***Edit: I am aware that I just basically repeated D_A's advice (apologies) but I have found lately that people are having trouble with determining what the polarity of a tantalum capacitor is. It is not obvious and can be difficult to see. If you are building something or repairing something that needs a tantalum installed, the positive leg on a new tantalum is usually longer than the negative leg....assuming that the legs haven't already been trimmed.

Anyway, any progress on checking the trimpot and where the voltage for Q5's Drain has gotten to?

Thanks so much for the advice. I really appreciate it.

I plugged the pedal into my board with the 18v power supply powered on. Black leg of multimeter to ground, red leg to Drain of Q5. The reading seemed to be correct. I was able to adjust the trimpot so the Drain leg of Q5 read at 9.14, while it was hooked up to the 18v power supply. Half of the supply voltage is the goal, correct?

All I did different from yesterday was this: The set of 2N5458 transistors I ordered came in the mail today. I just put them in. I left one of the original MPF102's in Q1...then from Q2 to the end, 2N5458. I got rid of all the J113's. Only reason I didn't use MPF102 in Q2 as well as Q1 is because I stupidly clipped the legs off *very short* to try out a J113, as when I probed it the other night and it was getting no sound, I thought maybe the transistor was bad. Kind of upset now, as the MPF102's in Q1 and Q2 are what I originally wanted to try.

When my replacement MPF102's come in I suppose I could give those a whirl. I'm a bit gun-shy after this whole fiasco although perhaps it was as simple as not using the right transistors. 

J113's are listed as "This device is designed for low level analog switching, sample and hold circuits and chopper stabilized amplifiers." I am not sure what that means but perhaps they're simply incompatible with this circuit.

Maybe the J113's were causing a problem? I'm not saying my pedal is fixed. I sure hope so, but I am only in the cursory stages of testing it with the new transistors.

But I am getting more signals now with the audio probe, and the Drain leg voltage seems to be perfectly in spec. Running at 18v, the trimpot seemed perfectly adjustable and being able to get the Drain to ~9v sure is a good sign. To be fair, I forgot about the trimpot advice before I swapped the transistors...so I did not test the Drain with the trimpot with the J113's in there.

All I did was get rid of the J113's, put in 2N5458's, and now the pedal seems to be working without issue. 

I will keep in mind that things seemed to be OK with the NTE457 transistors, but then things went wrong while I was still using those transistors. I don't trust I'm out of the woods yet. But these definitely seem to fit better.

I'm feeling tempted, if this pedal continues to work, to just buy one of the BSIAB II kits and try to make one from scratch if I would like to further tweak the tone. I have some MPF102 transistors coming in the mail as well that I planned on experimenting with.

I could build a brand new circuit board from scratch, put sockets in for the transistors, and all that good stuff. And try some of the other mods listed on this community that sound very cool, but I would not want to risk doing to my Brownie given what's already happened so far.

Slowpoke101

If the pedal is now working that's great :icon_biggrin:
Adjusting the trimpot to half (ish) the supply voltage is the initial setting. As long as the pedal is working you now adjust the trimpot setting "by ear"...Adjust it to where you find it sounding the best according to your tastes.

Getting a kit would be a good idea just for the fun value and being able to leave this pedal alone until you feel more confident with modifying it.

Sometimes when you buy components (particularly FETs) the component can be faulty or below specifications. This can make things very difficult to work out what is going on. For a newcomer to the craft this can be a very bad experience. You may have received some below spec' FETs which won't work in this circuit but may work in others.

Testing components is not difficult but can be confusing. There are many helpful articles on this site (and elsewhere) that can help with this. A case in point is your own resistance measurements. You were able to measure only three resistors. All those resistors were under 1K (1000 Ohms) which means that your multimeter was not set to the correct range(s) to be able to measure the other resistors. Spend some time with your multimeter and learn its ways because a multimeter can be one of the most useful pieces of test equipment that you can have.

Hopefully you have repaired your pedal and now you can get on with playing without any problems.
  • SUPPORTER
..

Slight Return

I definitely need to learn how to use the multimeter. So it wasn't that they had no reading...I was just using the meter incorrectly.

I can understand DCV at "20" for testing voltages that will be around 0-20. Simple enough. When I tested the Drain leg of Q5, I was able to understand that I needed to set the meter to DCV and in the numerical range of what I expected to find.

For pickups, I understand to use 20k as I'm looking for the resistance of the wire, and up to 20k range, ex. 6.2k, 12k, etc. That's about as deep as my understanding of multimeters goes at the moment.

Fortunately, my friend left a busted DS-1 at my house months ago. I replaced the battery clip and blown diode but the pedal still doesn't work. I'm excited to use the audio probe I made to test it out. I do think it's a convenient method too, using one of the spare pickups I made and my cell phone to play music into it while I probe...seems to work very well.

And I was pleased to find that I am getting a signal from the pedal. It was overloaded with more voltage than it could handle, the diode blew, and the battery clip was also bulging out. So it seems the pedal has some major issues, and I also noticed a solder bridge on a couple components on the back.

But it could be a great project to help get me up to speed using the multimeter and learning how to find faults in circuit boards on effects.

For now, the Brownie is working. Maybe it really was just shoddy transistors, or inappropriate transistors for the circuit, period...NTE457 and J113's are transistors that are "allegedly" substitutes, but I have never heard of anyone using them in a Brownie. Sticking to more standard transistors that have been proven to work in the circuit might have just been the ticket.

Slight Return

#27
Oh my god.

Okay...sounded perfect, but it's not working again. After a while of playing, it starts making a crackling noise that keeps going whether the volume is on or off. Then the pedal starts cutting in and out intermittently. Then it just stops working. Blatty, terrible sounding, with almost zero volume and almost no gain to speak of.

It works great...then stops working. This has been the pattern several times now. Just when I think it's fixed, and working perfect....it cuts out.

Any ideas what measurements I should try now? I'm not sure what it could be, but again, if it's relevant...sounds great, works great, then just stops after about 15 minutes or so of being on, and dies.

Just plugged it back in, and it sounds great...then after several minutes, it dies out again. Definitely a pattern here, but I have no idea what could be causing this. It seems if I leave it off for a while, and then try it again, it works, but under continuous use it just craps out.

Slowpoke101

Frustrating.
Power it down for a while then power back on. Confirm that it is working and quickly measure the voltages on the transistors.
Wait for it to fail then re-measure the transistor voltages again. Post them here so we can have a look.
The other thing to try is to run the pedal at 9 volts and see if it fails at that voltage.
  • SUPPORTER
..

Slight Return

#29
Made a video demoing what it sounds like when it starts going out. I did find that I could cause this kind of failure by putting the gain knob all the way up and playing for a little bit. It maintained this issue even after bringing the Gain knob down.

https://www.youtube.com/watch?v=W4d5Qkjlrhk&feature=youtu.be

@ Slowpoke

I will try this. For right now, what about polarity? I just noticed that my Fulltone OCD has a Positive Polarity symbol next to the 9-18v adapter, and the Brownie has a Negative Polarity symbol next to the 9-18v adapter. I've never done anything with my Voodoo Lab Pedal Power 2 Plus to accommodate anything, just used the power adapter cords that came with it.

Is this something I should be concerned about?

EDIT: Nevermind...I didn't realize the small "half circle" in the polarity symbol was facing towards positive in both pedals, with the "open" side of the circle facing toward negative, meaning that they're the same. I guess one of them just wrote it upside down, depending on how you look at it.

Slowpoke101

QuoteI didn't realize the small "half circle" in the polarity symbol was facing towards positive in both pedals, with the "open" side of the circle facing toward negative, meaning that they're the same. I guess one of them just wrote it upside down, depending on how you look at it.

From that description it seems that the polarity would be correct. Make sure that you have the manual for the power supply as some units are rather complex as to the options they can provide.

Your BSIAB2 pedal has a polarity protection diode (in series with positive line) in it so the pedal would not work at all if you had the power reversed.

After listening to your video I noticed that as the played note or chord decays you hear a distinct fizzle then it seems to gate off. This is a sign of poor biasing in these FET based distortion pedals. Something is obviously upset and letting the team down. A crook FET can do this but get the voltages first and we will see.


  • SUPPORTER
..

Slight Return

With the effect powered on at 18v, and sounding like it did in that YouTube clip, these are the readings I got. For simplification, the numbers will all be ordered Gate first, Source second, and Drain last. I did wait for a while and powered it up but it failed to work properly...so for now I'll put these voltages down, hopefully I can get the others down soon.

Q1:

0
6
9.6

Q2:

6
9.6
18

Q3:

0
0.7
9.2

Q4:

5.8
9.1
18

Q5:

1.7
3.0
3.3

Here's the crackling sound, too. It was doing this almost the entire time I was testing it.

(Not *while* I was testing it, specifically...it was doing this noise whether I was probing it with the MM or not)

https://youtu.be/AhxciU-VrCM

Slowpoke101

I still recommend trying the pedal with 9 volts, just to see if it calms down.
The voltages seem to be close to what I would expect to see at 18 volts. All but for Q5.
Try a different FET just to see if it improves.

  • SUPPORTER
..

Slight Return

Tried running the pedal at 9v, but no change.

Also changed the transistor in Q5, also no change. I can record those voltages if you think it would help. But I haven't been able to get the pedal sounding good for any length of time since it fizzled out the other day. Just continues to do the fizzling thing and hasn't gone away since.

Is the poor biasing simply a result of bad FET transistors? If so, how common is this? Is it necessary to spend a lot of money on a ton of transistors and audition dozens upon dozens of different ones before finding a set that works?

Is how it sounds also consistent with a cold solder joint? Upon the advice here I did look the joints over, and they look fine. I re-flowed them just in case, but that didn't seem to work either. To check again, I *very* lightly moved each transistor, just to see if the legs could flex a little bit, without breaking free from the solder joint...which should indicate a bad joint. They all were able to flex a bit without the legs breaking free.

I would be more than happy to realize I'm at fault here. But I just don't understand why it isn't working! Very confusing.

Either way, I'm so frustrated with this I decided to buy one of the BSIAB II kits. I'm not *completely* giving up on the Brownie yet but it is looking pretty bleak. I just can't wrap my head around what the issue is.

https://www.amazon.com/Socket-Transistor-pitch-inch-2-54/dp/B075Z1GH1K/ref=sr_1_1?ie=UTF8&qid=1520142336&sr=8-1&keywords=transistor+sockets

I got those sockets originally for the Brownie. I've been planning on using these for all the transistors when I build my BSIAB II from the kit.

Any thoughts on that? I've heard some people say that sockets are not as reliable as soldering directly to the board. However, for all the issues I'm having, being able to swap them out quickly seems like an advantage. It would also guarantee that I wouldn't accidentally damage them by overheating them.

I hesitated a bit, as the CMATMODS Brownie is made by a reputable builder, and I highly, highly doubt there is anything wrong with anything else on the board...like in the first comment in this thread, it was working before I tinkered....then I tinkered and it stopped working.

The only thing I can imagine is that I somehow made bad solder joints, or I have bad FETs. I am more inclined to think that they're just bad joints. I used flux when I soldered them all on and it seemed to flow well and touch everywhere it should and nowhere it shouldn't. But if they are bad, I wonder if there's any way to tell? I visually inspected them to the best of my ability, and while they're not gonna win any beauty contests they seem pretty solid.

Slowpoke101

Let's see. Your pedal was working and then you decided to try different transistors in it to make it sound 'better' when running at 18 volts. Not a silly idea at all.

Here are the variables; There is no guarantee that your new transistors are able to work in the circuit or work at all.
                                De-soldering can damage the circuit board (mainly the through plating in the component hole pads).
                                Soldering can also introduce problems - Shorts, dry joints, overheated components....

So really there is very little that can be wrong. The most likely problem would be unsuitable (or unusable) transistors or a solder connection point is faulty. The fact that the pedal sort of works and then fails points the finger of suspicion towards the transistors. Installing those transistor sockets may be a good idea. Some are really good and hold the transistor leads very well and some are terrible, bordering on useless. I generally solder the transistors into the socket (if I bothered to use one) when I've got everything working properly.

I think that Smallbear has suitable transistors available that are known to work correctly. This would save you from buying a fortune's worth of transistors and auditioning them - which is best done in a working pedal. Your new kit will also have the correct transistors (assuming that you bought the full kit and not just the board).

Good idea to get the kit. Always fun to build things. But get it going on 9 volts first and then try 18 volts. Beware of the voltage ratings of any electrolytic or tantalum capacitor that may be used. Make sure that 18 volts will not exceed it.

  • SUPPORTER
..

duck_arse

can I request a few numbers? with the power off, and the meter to Ohms, can you give us the resistance reading from the gate of Q5 to ground, and from the gate of Q5 to the supply line? and just for fun, the resistance from the gate of Q3 to ground, with the gain pot at each extreme of rotation.
" I will say no more "

Slight Return

How do I do this exactly?

I went through all the Ohm settings on my multimeter, which one would be appropriate for this?

I didn't get any readings at all unless I had the red probe on ground, and the black on the Gate of the transistor.

In which case I got:

Q5 Gate to Ground: -598

I get ~631 if I put the black probe on the Gate of Q5 and the red probe on the positive lug of the 9v adapter. Is this what is meant by the supply line? It started around 644 but with continual testing the number gets lower and lower, like a minute later now it's reading at 621.

Doing my best to test what I can but I'm pretty sure I'm doing this wrong. Any clarification? Is black probe on Gate of Q5 wrong or is that what I'm supposed to do?

I've been using the tab on the input jack that connects to the negative terminal on the 9v power supply as Ground. Is that incorrect?

Going that way, I found -589 with the Gain all the way up, and -601 with the Gain all the way down on Q3.

PRR

> bad solder joints....any way to tell? I visually inspected them to the best of my ability, and while they're not gonna win any beauty contests they seem pretty solid.

Get good light. Get magnifier. If you are over 40, use your readers too.

Joints should at least pass the Beauty Contest Swimsuit Test. You should be able to see the parts through the suit/solder. With blobbed solder you can be SURE the solder has "wet" the joint.

A bad FET is *possible*. I recently saw a case where a transistor was dropping in and out of conduction (from a tech who knows bad solder all too well). But it is far less likely than bad joints.
  • SUPPORTER

Slight Return

Just re-flowed all the joints, especially on Q5.

Plugged it in, ran it, and worked great for about 3 minutes. I checked the Drain leg on Q5 while it was working great and it was reading out right at 4.5v after adjusting the trimpot.

When it started to cut out, I quickly put the meter back on. It Read around 1v. Then it jumped all around from 2 to 3 and back to 1 and then finally settled somewhere around 2.5v to 3v. Very quiet, very low gain, very fizzly sounding.

Re-flowed it again....started out at 4.3v.

1 minute later, as I'm sitting here writing this, I hear the static hum/buzz of the pedal being on disappear. I go over and check the Drain leg of Q5 again and it's at 3.0v. Starts crackling, check it again, it's at 3.37v.

This is all without doing anything at all other than just leaving the pedal powered on and plugged in. Not moving it, pressing the switch or changing any knobs or anything except intermittently checking it with the voltmeter.

I ordered a good Solder Sucker that should be coming in soon. I noticed that the more I tried to re-flow, the more solder built up because I forgot I'd tinned the iron beforehand.

Slowpoke101

Have a quick look at this picture.



Q3 Source leg seems to be missing solder on the top layer pad. I cannot see if there is a track connecting this pad to anything else (on the top layer) but there may be. If I recall you did say that you had trouble with damaging the plate through on Q3's connection pads. You may need to see if you can get some solder into the connection point.

Another thing to see if it makes any difference is to re-install the 0.0022uF capacitors and see if the pedal returns to the original condition.

  • SUPPORTER
..