[BUILD] New FET/RACK Official Help Thread - Please read first post!

[Harpo]

Q12: C= 26.23 ; B= -9.14 ; E= -1.59

doesn't look to be working correctly...

 

[YoDudeRock]

Hi Mike (and friends),
      I am having a weird problem with a Rev A.  It was operating perfectly normal for about a month...
now the VU needle has started meandering around from zero  (above and below). 
There's no noticeable audible effect, although it's probably subtly affecting the compression. 
I did some checking with my multi-meter and found voltages in every component in the GR Control Amp
section meandering around too.
For instance on Q10, the Collector is a rock solid: 30.78, but the Base and Emitter are fluctuating from 13-17v DC. 
Pretty much everything below that in the GR Control Amp section is fluctuating by about the same ratios too.
What would be the most probable cause?    Loose connection?  Bad capacitor, transistor, resistor?

I've changed both Q1 and Q11 (had a spare matched set), Q10, R49, R50 and C18. 
I've also re-heated some solder joints too (R50 particularly), which has greatly reduced the amount that the VU meter is moving, but it's still "moving" back and forth a little bit.
Any ideas?

Any help is much appreciated,
Chris Trentham

 

[Echo North]

Could really be  any of those.  I'd suspect a leaky cap or damaged semiconductor in that section.  Also, look for uncleaned flux.

Mike

 

[YoDudeRock]

Mike,
    Thanks so much.  Just wanted to follow up.    I finally, reluctantly replaced the rest of the 2n3707's at Q9, Q8 and Q7.    And success!  Voltages are all rock solid now and the same with the VU.  Before, voltages in the GR Control Amp section were meandering all over the place like a stock ticker during the trading day.    It was obviously one of those 2n3707's causing voltages to fluctuate in that manner.    Hopefully this helps someone else in the future. 

Thanks again,
Chris Trentham

 

[Echo North]

Mike,
    Thanks so much.  Just wanted to follow up.    I finally, reluctantly replaced the rest of the 2n3707's at Q9, Q8 and Q7.    And success!  Voltages are all rock solid now and the same with the VU.  Before, voltages in the GR Control Amp section were meandering all over the place like a stock ticker during the trading day.    It was obviously one of those 2n3707's causing voltages to fluctuate in that manner.    Hopefully this helps someone else in the future. 

Thanks again,
Chris Trentham

Nice!

 

[YoDudeRock]

Mike (and friends) 
    I am working on two Rev A's and I ordered the matched 2n5457's and two stereo link cards.
Question...Is it important that the hFE of Q12 and Q13 in both units (all 4 transistors) be matched as closely as possible? 
Is there any advantage to matching all four transistors to the same value?     
The reason I ask is, right now I don't have 4 that match, I have 3 at 405 hFE and 3 others at 410 hFE (and various others).

Also, any other tips to "matching" units?    The guy I'm making these for says he wants them "matched as closely as possible".


Thanks,
Chris

 

[Echo North]

Mike (and friends) 
    I am working on two Rev A's and I ordered the matched 2n5457's and two stereo link cards.
Question...Is it important that the hFE of Q12 and Q13 in both units (all 4 transistors) be matched as closely as possible? 
Is there any advantage to matching all four transistors to the same value?     
The reason I ask is, right now I don't have 4 that match, I have 3 at 405 hFE and 3 others at 410 hFE (and various others).

Also, any other tips to "matching" units?    The guy I'm making these for says he wants them "matched as closely as possible".


Thanks,
Chris

Q12 and Q13 being matched just allows you to calibrate the comparator meter circuit on the individual unit.  Don't worry about matching those.

Mike

 

[YoDudeRock]

Mike,
    I'm testing my 2 Rev A's with Active Link cards.  I noticed there is a little bleed from one unit to the other, when Linked.  If I only send a signal only to unit 1, I'll still see about -50db coming from Unit 2 on the other side (in my DAW, Reaper) and vice versa.  Is this normal?  I assume it is since this is living, breathing analog.    I tried recording it and it's basically inaudible.

This is for a customer of mine, so I didn't want to send it to them if this isn't normal.

Thanks,
Chris Trentham

 

[More_Gain]

Hello everybody,

I've just completed the assembly of  a Rev A kit I bought about a year ago (it's a 2017 v1.14 pcb) . I have +29.7vdc and - 9.44vdc at the power rails.

During the q bias I was getting only 0.132vac at output. I started going through the troubleshooting list and got  -1.337vdc to -0.776vdc at TP18 a little off but seems usable.

I measured 0.088vac at TP1 (little off), and then 0.017vac at TP15 (way off). I then grabbed my scope to try to figure out where the signal drop off occurs, and found it at the gate of Q1.

After reading the beginning of this thread I tried removing Q1 and the gain is vastly improved. I measure 3.63vac at TP15 and 16.45vac across output (ultra hot). I measured +12vdc at the gate of Q1 which also seems WAY off.

I've tried replacing Q1 and Q11 with spare 2n5457s I had with no improvement.

Let me know if you have any ideas of what to try next.

 

[Echo North]

Mike,
    I'm testing my 2 Rev A's with Active Link cards.  I noticed there is a little bleed from one unit to the other, when Linked.  If I only send a signal only to unit 1, I'll still see about -50db coming from Unit 2 on the other side (in my DAW, Reaper) and vice versa.  Is this normal?  I assume it is since this is living, breathing analog.    I tried recording it and it's basically inaudible.

This is for a customer of mine, so I didn't want to send it to them if this isn't normal.

Thanks,
Chris Trentham

Probably, I've never tested it in that way.  It's a summing amp basically, it's not perfect but it's better than the stock link for sure.

Mike

 

[Echo North]

Hello everybody,

I've just completed the assembly of  a Rev A kit I bought about a year ago (it's a 2017 v1.14 pcb) . I have +29.7vdc and - 9.44vdc at the power rails.

During the q bias I was getting only 0.132vac at output. I started going through the troubleshooting list and got  -1.337vdc to -0.776vdc at TP18 a little off but seems usable.

I measured 0.088vac at TP1 (little off), and then 0.017vac at TP15 (way off). I then grabbed my scope to try to figure out where the signal drop off occurs, and found it at the gate of Q1.

After reading the beginning of this thread I tried removing Q1 and the gain is vastly improved. I measure 3.63vac at TP15 and 16.45vac across output (ultra hot). I measured +12vdc at the gate of Q1 which also seems WAY off.

I've tried replacing Q1 and Q11 with spare 2n5457s I had with no improvement.

Let me know if you have any ideas of what to try next.

Signal at the Gate of Q1?  There should be no signal there, just a negative DC bias voltage. Q1 isn't set up and a traditional transistor, more of a variable resistor to ground.

The 5457's particularity on the Rev A are very tightly spec'd.  Tossing any 2 in there, will probably not work out.  If you have +12VDC at the gate of Q1 that is your problem, not the FET.  You'll never pass signal with positive DC voltage there.

Do you have the CN2/CN3 heading in backwards or something? I check all your wired connectors for orientation.

Mike

 

[Deejsirois]

Those GR amp outputs look fine to me.  I would definitely get proper FETs, I'm guessing you have knock-offs.

Mike

Well, tried a set of matched FETs I got from you and still the same distorted signal when the gain reduction is turned on. I also replaced all the 2N3707 transistors in the GR control amp section just to make sure that wasn't the issue. I once again checked all the resistors in the GR control amp section as well as all the resistors on the ratio pcb....everything is correct.  Need a new thought/direction to head in here because I am at a loss.

 

[Echo North]

Well, tried a set of matched FETs I got from you and still the same distorted signal when the gain reduction is turned on. I also replaced all the 2N3707 transistors in the GR control amp section just to make sure that wasn't the issue. I once again checked all the resistors in the GR control amp section as well as all the resistors on the ratio pcb....everything is correct.  Need a new thought/direction to head in here because I am at a loss.

So it works fine with no FET? Other than the lack of compression?

 

[Deejsirois]

So it works fine with no FET? Other than the lack of compression?

No, when I remove Q1 AND turn the attack on, I get the nasty distortion when 20:1 is selected. It's less of an issue with 12:1 until 4:1, which sounds almost normal. If I put the Q1 FET in and keep the attack off, audio sounds beautiful. If I remove Q1 and keep the attack off, the audio sounds beautiful.

 

[Echo North]

No, when I remove Q1 AND turn the attack on, I get the nasty distortion. If I put the Q1 FET in and keep the attack off, audio sounds beautiful. If I remove Q1 and keep the attack off, the audio sounds beautiful.

That's nuts.  So nothing sounds good with the GR ON, FET in or out.  And you still cant get a voltage drop at your Qbias but you get a -1VDC to -1.7 (ish) reading at the gate?

 

[Deejsirois]

That's nuts.  So nothing sounds good with the GR ON, FET in or out.  And you still cant get a voltage drop at your Qbias but you get a -1VDC to -1.7 (ish) reading at the gate?

Correct, nothing sounds good with GR ON. However, I can fully calibrate the unit...so we are at least good on that! Calibration went without a hitch, and I was so pumped thinking everything was golden....until I turned the GR on 

 

[Echo North]

Correct, nothing sounds good with GR ON. However, I can fully calibrate the unit...so we are at least good on that! Calibration went without a hitch, and I was so pumped thinking everything was golden....until I turned the GR on 

Oh ok.  So this is different than originally when you couldn't even calibrate the bias?

So everything calibrated normal, but in GR mode you have a lot of distortion?  Re-check your Pad 22/21/CR2 voltages under calibration.

Mike

 

[Deejsirois]

Oh ok.  So this is different than originally when you couldn't even calibrate the bias?

So everything calibrated normal, but in GR mode you have a lot of distortion?  Re-check your Pad 22/21/CR2 voltages under calibration.

Mike

Input: 12 O’Clock
Output: 12 O’Clock
Attack: GR ON (fully CW)
Release: Full CW
Ratio: Check All
Meter: GR
Signal: Sine 1kHz @ 0dBu (0.775 VAC across input + and -)


Pad 22                                          Pad 21                                                      CR2  Anode

20:1 =    0.460    VAC            20:1 = -5.80      VDC                      20:1=  -5.86  VDC
12:1 =    0.231                        12:1 =  -3.34                                        12:1=  -3.34
8:1 =      0.153                          8:1 =    -2.55                                          8:1=  -2.55
4:1 =      0.76.8                        4:1 =    -1.76                                          4:1 =  -1.76

 

[Echo North]

Anode CR2?

 

[Deejsirois]

Anode CR2?

Just modified my answer above