Weird drive sound on a Bearhug v2.1

[Tuxedo3]

Hey guys, I'm building my second Bearhug Comp (the first one I built was the original version) and for some reason I'm getting this like drive/fuzz sound in the signal. It isn't a clean compression like the other one is. The build went really well and I biased the test pad to around 5.5v like the instructions said but I'm wondering if anyone else ran into this issue? I'll gladly include pictures.

[Tuxedo3]

Here's a pic if it's helpful.

[jimilee]

I'd reflow and and make sure you have component placement correct. Check for bridges  also.


Sent from my iPhone using Tapatalk

[midwayfair]

General troubleshooting:

Measure the resistance across Q3 (drain to SOURCE, not drain to gate) at idle. If it's not between 300 and 500 Ohms, the FET is either bad, out of spec, or incorrect in some other way (for example, a J201 will not fall in that range).

If the resistance at idle is correct, ensure that you are getting a rectified voltage by measuring DC at the gate of Q3 while either playing a chord on the guitar or running a test signal pulse. You should get anywhere between 1V and 5V with a guitar signal.

If you are getting a rectified voltage, measure the resistance across the FET again with a guitar or test signal. The resistance should rise to as high as several thousand ohms.

If all of these steps are functioning correctly, then your problem will be somewhere in the relatively simple amplifier section, which is just Q1 and its bias network.

Also, as always, POST VOLTAGES WITH YOUR INITIAL TROUBLESHOOTING POSTS. A huge number of problems can be itentified in all manner of projects simply from incorrect DC voltages.

[Tuxedo3]

Okay, thank you for all that. I will test it and post voltages later today.

[Tuxedo3]

Sorry about the delay on this.
Test pad voltage: 5.22v DC
Q3 resistance between drain and source: 168.7 ohms

After switching to a new Q3 (still a 2N5457) the resistance stayed the same.
I reflowed solder joints before doing this and the result was the same. I can post other voltages I'm just not sure what to post.

[midwayfair]

Follow the remainder of my steps. It's probable that both FETs are out of spec but you can test if the circuit's working before worrying about that.

Where did you get your FETs?

[Tuxedo3]

I will definitely follow the other steps.
I got the 2N5457's from ebay and used other ones in the same lot in a couple other pedals and it seemed to be fine, but those could definitely be weird. The BS170 and MPSA18 I bought from tayda.

[Tuxedo3]

Also, I took a video to show the issue a little better.

[Tuxedo3]

Okay, I checked the voltage at the gate of Q3 and it read anywhere from -0.2 to -0.5 v DC when I strummed.
Even though that was wrong I went to check the resistance at the gate of Q3 when I strummed and it would get up to about -12k ohms.
Something seems so wrong haha. Should I try replacing all the transistors?

[chuckd666]

Definitely get some legit transistors - if you got them from ebay for a 'cheap' price, they're likely to be fake. I bought a bunch of J201s from ebay which were all useless (or at least I don't know what they originally were!).

[Tuxedo3]

Definitely get some legit transistors - if you got them from ebay for a 'cheap' price, they're likely to be fake. I bought a bunch of J201s from ebay which were all useless (or at least I don't know what they originally were!).

I have never had issues with tayda transistors. I built upwards of 10 pedals and 5 buffers with exclusively their components (including these exact transistors on a Bearhug v1) so I would doubt they would suddenly  be bad.

[pickdropper]

Definitely get some legit transistors - if you got them from ebay for a 'cheap' price, they're likely to be fake. I bought a bunch of J201s from ebay which were all useless (or at least I don't know what they originally were!).

I have never had issues with tayda transistors. I built upwards of 10 pedals and 5 buffers with exclusively their components (including these exact transistors on a Bearhug v1) so I would doubt they would suddenly  be bad.

I think you've been luckier than most, then.  There have been known instances of non-legit parts coming from Tayda.  You can certainly save a bunch of money building pedals with Tayda parts, but it's not a bad idea to keep a couple examples of known good parts around for comparison.

[midwayfair]

Definitely get some legit transistors - if you got them from ebay for a 'cheap' price, they're likely to be fake. I bought a bunch of J201s from ebay which were all useless (or at least I don't know what they originally were!).

I have never had issues with tayda transistors. I built upwards of 10 pedals and 5 buffers with exclusively their components (including these exact transistors on a Bearhug v1) so I would doubt they would suddenly  be bad.

There are lots of times it doesn't matter. Unfortunate there are some uses where it does.

There is a way to hook up a FET so that it can be trimmed to have the proper resistance range regardless of the part that's used, but it requires a negative supply. You've measured your FET and its resistance was too low (about half the expected value) at idle compared to the ones I get from small bear, which in turn is amplifying the signal so much that it's clipping and then triggering s compression too hard. It's possible to clip the circuit with high output pickups even when it's working properly, but in this case there was something measurably wrong. It would be like grabbing a resistor marked 390R and getting a 168R one instead.

If you absolutely don't want to have to order a commonly faked or seconded part from a reputable source, replace the q1 drain and source resistors with something like 1k and it might be more like the expected gain range, but that will more than double the current draw of Q1.

EDIT just to clarify this. The source of the part is irrelevant. If you had gotten it from Smallbear and knew it was genuine and it measured that, you still would have had to buy another part. What is relevant is that it measures incorrectly and will not work with the circuit as designed. It's the same as if you had gotten a mismarked resistor or capacitor. By the same token, if you put a part in marked with one value and it's actually the other value and it causes a malfunction, it doesn't matter if you've gotten a million other resistors from the same supplier and never experienced a problem before. Doing the parts replacements I suggested might get a working effect with less of the problem but it is not the same as using a part that measures in-spec. Sometimes parts have to be selected. The easiest ways to avoid that are either to design the effect differently (which would have added significant complexity to this one and somewhat defeated the purpose) or buy batches of parts that you can generally trust.