Bear Hug Compressor...paging midwayfair! (SOLVED)

[pryde]

Not sure when you say measure resistance "across" Q3 but with probes on G and S (or S and D), I am getting about ~16k resistance.

With Q3 out and D/S jumped, I trace loud signal up to D3 anode. At D3 anode the volume drops pretty significantly and there is clipping as well.

Also, voltage jumps from 0 to ~2.5v at D3 anode when striking heavy chord.

D3 bad?

[midwayfair]

Not sure when you say measure resistance "across" Q3 but with probes on G and S (or S and D), I am getting about ~16k resistance.

With Q3 out and D/S jumped, I trace loud signal up to D3 anode. At D3 anode the volume drops pretty significantly and there is clipping as well.

Also, voltage jumps from 0 to ~2.5v at D3 anode when striking heavy chord.

D3 bad?

The voltage change should be negative. I just want to confirm that.

So far everything's working fine except for some reason your FET has too much resistance between the drain and the source. In-circuit, it should only be ~300 ohms. When you create the voltage at D3 (which is right but does seem a little low), it applies voltage at the gate of the FET and increases the resistance between the gate and source (so you ought to measure ~300 ohms from ground to the drain when Q3 is plugged in, then play a chord and see it jump up to 1-2K).

But at this point, I might have to wait until you can get photos, and wait till I can get home to give you precise voltages off a working model, because electronically I'm fairly stumped. Your FETs just don't seem to be linearizing. I'll give it some more thought and pull out mine when I get home, unless someone else has theirs handy that they can test.

[pryde]

Yes Q3 is wonky concerning voltages. I will get some good pics posted this evening and we can go from there. It might help if I knew anything about compressor circuits but not so much 

[midwayfair]

Yes Q3 is wonky concerning voltages. I will get some good pics posted this evening and we can go from there. It might help if I knew anything about compressor circuits but not so much  

Check the original project thread -- I explain the inner workings of the Bearhug. I don't know if it'll help with your build, but it might.

But you've confirmed that resistance to ground at the drain of Q3 does not change with a guitar signal? edit: Resistance at Q3, not voltages!

[pryde]

OK for clarification: vol on full, compression on full...

Q3 drain settles at 8.8k resistance, then jumps up to ~14k with hard strumming. So yes the resistance does change.

thoughts?

[midwayfair]

OK for clarification: vol on full, compression on full...

Q3 drain settles at 8.8k resistance, then jumps up to ~14k with hard strumming. So yes the resistance does change.

thoughts?

Man, that is *crazy* out of spec. Flip it around, and try another FET (if you have a 2N5952, that will work, too). Same-ish readings?

Just out of curiosity, did you get the 2N5457 from Tayda? I had a batch recently that didn't work in the Blue Warbler and I'm not certain they didn't get a bunch of bad trannies.

[pryde]

DING DING DING 

Bad 2N5457's it was!

I never ordered from tayda but I have a mixed bag between mammoth and mouser. I tried 4 more with the same result then pulled one out of an old orange squeezer vero I have in the pile and bingo, it works!

The drain resistance is 430R then jumps to about 1k with hard strumming. I assume this is within a reasonable spec?

Man it is kinda hard to trouble-shoot when you have a batch of bad transistors 
Thanks for all your help Jon

[midwayfair]

DING DING DING 

Bad 2N5457's it was!

I never ordered from tayda but I have a mixed bag between mammoth and mouser. I tried 4 more with the same result then pulled one out of an old orange squeezer vero I have in the pile and bingo, it works!

The drain resistance is 430R then jumps to about 1k with hard strumming. I assume this is within a reasonable spec?

Man it is kinda hard to trouble-shoot when you have a batch of bad transistors 
Thanks for all your help Jon

sounds like that's a good FET!

I hear ya, it SUCKS troubleshooting around a bad part ... my favorite was some mis-marked resistors I got recently. Big difference between 8.2R and 8.2K ...

Glad it's working, and thanks for sticking with the troubleshooting through to the end. Jumping to 1K sounds about right for single coils. My DQ caster hits 1.3K in my test unit, well within several tolerances of Q1 and Q2.

[pryde]

thank you again for everything. Kudos to a really nice compressor. It gets that subtle-sweet OS compression and then can really clamp down from there if you need it to.

As a bonus, it has a great clean boost with the comp turned all the way down.

Looks like I got a small batch of 2n5457's to sort through to see if any are usable 

[midwayfair]

thank you again for everything. Kudos to a really nice compressor. It gets that subtle-sweet OS compression and then can really clamp down from there if you need it to.

As a bonus, it has a great clean boost with the comp turned all the way down.

Looks like I got a small batch of 2n5457's to sort through to see if any are usable 

Even if they aren't usable in a linearized FET application, they will still make good amplifiers and/or buffers. You could test them in a Fatpants for that.

There are few parts that are useless just for failing on some weird spec.