[BUILD] FET/500 Official Support Thread

[frenkonio]

Hi guys,
currently finished my first Fet 500 Rev. A. The -22 led is always on (just a bit) and the input reaches PK led (red) with just a small amount of input gain. Is it normal? I also don't feel good with metering, I would prefer something different from this mode (I mean only a couple of leds are lightning, I also have a Capi FC526 and it has all leds working when compressing so better for understanding what's happening).
Anyone could help?

 

[Echo North]

Hi,

Peak LED it set properly is set to +16dBu, generally the level you're getting close to clipping your converters.  It's measuring the peak rather than the RMS like the rest of the meter.  You could trim it to be higher if you like.

The -22 being dimly lit is an issue with the chip. Just put a 2K(ish) resistor across the -22 leads and it'll fix it.

Try putting the meters in BAR mode, see if you like that better.

Mike

 

[boneindian]

Hey all, just finished reading through the build thread to make sure no one else has had this problem and couldn't find anything quite like this.  Just finished two REV As that I built simultaneously and they're both displaying the same behavior at the calibration stage.  It is EXTREMELY difficult for me to get to .775 VAC across input pins 2 and 3 unless the unit is in bypass.  I can easily get it across pin 1 to either 2 or 3, but even cranking either my SSL console's osc output or one from a DAW the most I can get is about 22mv on the input.  At first I thought there might have been a "speako" in the calibration guide video and that the measurement was maybe supposed to be across the ground pin, but since it's the same in mnat's I've concluded I've done something screwy.

Both units behave exactly the same! Qbias trimmer does affect the output voltage, but since I'm starting with such an underpowered signal the most I can get on output from a .775VAC (measured across the wrong pins since reading across 2&3 will give me a max of 20mVAC I'm using this as my reference for now) input is .439VAC ...even with input and output knobs fully CCW. (Unless I blast the input to 5VAC)

For reference, a 5.5VAC reading across input pins 2&3 (on the XLR not the lunchbox) drops to 20mVAC when taken out of bypass.  In bypass unit shows same voltage on output as input.

At first, I thought I had done the input mod wrong, but even after removing the 100ohm resistor and reflowing the T-pad pins the behavior is the same.

Unit does pass audio and all LEDs appear to function, but I haven't really been able to get past the first calibration step because I can't get my INPUT voltages correct.

Both units are capable of gaining up the 20mVAC at input to 382mVAC at output and the meter can be set to display that level correctly. All LEDs appear to function properly although I clipped the BAR mode jumpers on one to see if that was part of the problem.

In short it seems many parts of the units are working correctly, but something is possibly screwy about my input stage.  It's a clean, neat build and I haven't been able to find any missed solder points or shorted ones.  Boards are still a tad sticky after their first isopropyl blast, but the fact that both units are behaving exactly the same makes me think it's not a flux issue.

Lemme know if you've got any ideas!

 

[Echo North]

Hey all, just finished reading through the build thread to make sure no one else has had this problem and couldn't find anything quite like this.  Just finished two REV As that I built simultaneously and they're both displaying the same behavior at the calibration stage.  It is EXTREMELY difficult for me to get to .775 VAC across input pins 2 and 3 unless the unit is in bypass.  I can easily get it across pin 1 to either 2 or 3, but even cranking either my SSL console's osc output or one from a DAW the most I can get is about 22mv on the input.  At first I thought there might have been a "speako" in the calibration guide video and that the measurement was maybe supposed to be across the ground pin, but since it's the same in mnat's I've concluded I've done something screwy.

Both units behave exactly the same! Qbias trimmer does affect the output voltage, but since I'm starting with such an underpowered signal the most I can get on output from a .775VAC (measured across the wrong pins since reading across 2&3 will give me a max of 20mVAC I'm using this as my reference for now) input is .439VAC ...even with input and output knobs fully CCW. (Unless I blast the input to 5VAC)

For reference, a 5.5VAC reading across input pins 2&3 (on the XLR not the lunchbox) drops to 20mVAC when taken out of bypass.  In bypass unit shows same voltage on output as input.

At first, I thought I had done the input mod wrong, but even after removing the 100ohm resistor and reflowing the T-pad pins the behavior is the same.

Unit does pass audio and all LEDs appear to function, but I haven't really been able to get past the first calibration step because I can't get my INPUT voltages correct.

Both units are capable of gaining up the 20mVAC at input to 382mVAC at output and the meter can be set to display that level correctly. All LEDs appear to function properly although I clipped the BAR mode jumpers on one to see if that was part of the problem.

In short in seems many parts of the units are working correctly, but something is possibly screwy about my input stage.  It's a clean, neat build and I haven't been able to find any missed solder points or shorted ones.  Boards are still a tad sticky after their first isopropyl blast, but the fact that both units are behaving exactly the same makes me think it's not a flux issue.

Lemme know if you've got any ideas!

Your unit is just slamming hard into compression.  You need to figure out why that is.  Are you sure you have the Qbias all 20 turns to the one side that give you the most output (CCW if I recall)?  You need to turn it all the way CCW until you hear that click in the trimmer.

If that's not it, then there is another error in the build, possible you made the same error twice.

You can use the FET/RACK troubleshooting guide to find out the error.  I would guess something at bad 22 maybe.  But check pad 22,21, and the diode output.  You MUST use the anode (+) of D4 as your ground  reference (black probe).

https://www.hairballaudio.com/blog/resources/post/fetrack-troubleshooting-guide

Mike

 

[boneindian]

Qbias is fully CCW! Until output voltage stop increasing and it clicks.

Would slammed compression truly drop the voltage at the input?

I'll check the voltages and pads first thing tomorrow, but I'm having trouble understanding how the GR circuit would affect the voltages at the input...(just curious!)

The compression meter is fully illuminated though so that checks out I guess.

thanks for the help Mike!

Your unit is just slamming hard into compression.  You need to figure out why that is.  Are you sure you have the Qbias all 20 turns to the one side that give you the most output (CCW if I recall)?  You need to turn it all the way CCW until you hear that click in the trimmer.

If that's not it, then there is another error in the build, possible you made the same error twice.

You can use the FET/RACK troubleshooting guide to find out the error.  I would guess something at bad 22 maybe.  But check pad 22,21, and the diode output.  You MUST use the anode (+) of D4 as your ground  reference (black probe).

https://www.hairballaudio.com/blog/resources/post/fetrack-troubleshooting-guide

Mike

 

[Echo North]

Qbias is fully CCW! Until output voltage stop increasing and it clicks.

Would slammed compression truly drop the voltage at the input?

I'll check the voltages and pads first thing tomorrow, but I'm having trouble understanding how the GR circuit would affect the voltages at the input...(just curious!)

The compression meter is fully illuminated though so that checks out I guess.

thanks for the help Mike!

The GR element Q1 is a variable resistor to ground. It's the first thing after the input transformer. Basically your input is grounded right at the input.

Come to think of it. Your Qbias is fine all is good in GR mode.  Your control voltage in GR ON must be way off. That would be a side-chain issue (pad 22/21/CR2).

Oh you'll need to look at the schematic to sort where Pad 22/21 is.  It's not on the 500 PCB.  If you have trouble with it, just post here.

Mike

 

[boneindian]

The GR element Q1 is a variable resistor to ground. It's the first thing after the input transformer. Basically your input is grounded right at the input.

Come to think of it. Your Qbias is fine all is good in GR mode.  Your control voltage in GR ON must be way off. That would be a side-chain issue (pad 22/21/CR2).

Oh you'll need to look at the schematic to sort where Pad 22/21 is.  It's not on the 500 PCB.  If you have trouble with it, just post here.

Mike

Just be clear, I haven't gotten to testing with GR ON.  The BYPASS I was talking about is the full bypass and meant to be an indicator that the associated relays work. All other testing has been done in GR OFF mode as per the calibration instructions

 

[Echo North]

Oh then I would suspect your issue might be that you’re getting no qbias to your Q1 gate.  That testing is also in the guide.

Just be clear, I haven't gotten to testing with GR ON.  The BYPASS I was talking about is the full bypass and meant to be an indicator that the associated relays work.

 

[boneindian]

Oh then I would suspect your issue might be that you’re getting no qbias to your Q1 gate.  That testing is also in the guide.

Ok, here are the results of today's testing along with a couple questions...I'll try and keep this succinct but please bear with me!:

First I went ahead and tested a couple of your initial recommendations using the anode of D4 as ground, mainly I wanted to look at Q1 since it's right at the front of the circuit and is where TP1 is in the FET/RACK troubleshooting guide... all I could figure out is that I had -2.1VDC on the relevant pin . I took this to mean I was getting Qbias at the gate even though I was finding it a little tough to decipher which pad was the gate between the two slightly different schematics and my own squinting at the board.

I also tried to measure the "output of the diode" as per your instructions Mike, which I took to mean D4 and got a reading of -.778VDC at the cathode of D4

Here's where things get a little funky.

I moved over to just the Trouble Shooting guide, but initially forgot to switch my ground from the D4 anode to the chassis!  While testing TP18 as instructed in the guide (the schematics of the 500 and the Rack are different but I am fairly confident I was able to find the correct point to measure) I got readings of -2.63VDC with the Qbias fully CCW and -1VDC with it fully CW...Perfectly within the range specified in the guide! (for a Rev D at least, I'm doing Rev A but it was close enough to give me hope).

Additionally, rotating the Qbias CW caused the GR lights to go out one by one, indicating more reduction and in  line with the circuit's expected behavior. (I also realized Mike, that when responding to your first analysis I made the same mistake many had and assumed all GR lights on meant total slammed compression as opposed to the opposite so, no, it does not appear my unit is over compressing)

However, upon switching my ground to the chassis suddenly all voltage readings  the Qbias became pretty crazy.  I read -18VDC at TP18 and also -18VDC and -16VDC on the other two pins of the Qbias trimmer resistor.  I had been ready to move on to checking the amplification stages, but those readings are so out of whack that I followed the instructions in the guide to go back to the forum. 

The thing is, the Troubleshooting guide is for the Rack version so it's unclear to me whether the chassis of my Radial Workhorse Powerstation is an appropriate place to ground my DMM for these troubleshooting tests or whether the "anode of D4" should replace the chassis ground in the troubleshooting guide when working on a 500 series module. Especially since they were very close to the expected range when using the anode of D4.

So, except for the crazy readings I got when using the chassis as ground the compressor seems to be behaving as expected, the fact that the input voltage across pins 2&3 of the XLR drops suddenly when the circuit is engaged makes me wonder if there's something screwy going on with the grounding of my rack.  Just a guess (and maybe a dumb one) but that's all I've got for tonight.

 

[Echo North]

Ok, here are the results of today's testing along with a couple questions...I'll try and keep this succinct but please bear with me!:

First I went ahead and tested a couple of your initial recommendations using the anode of D4 as ground, mainly I wanted to look at Q1 since it's right at the front of the circuit and is where TP1 is in the FET/RACK troubleshooting guide... all I could figure out is that I had -2.1VDC on the relevant pin . I took this to mean I was getting Qbias at the gate even though I was finding it a little tough to decipher which pad was the gate between the two slightly different schematics and my own squinting at the board.

If you have -2.1VDC at your Gate, that is good.

I also tried to measure the "output of the diode" as per your instructions Mike, which I took to mean D4 and got a reading of -.778VDC at the cathode of D4

You just measured the voltage drop across the diode.

Here's where things get a little funky.

I moved over to just the Trouble Shooting guide, but initially forgot to switch my ground from the D4 anode to the chassis! 

Look what I wrote above, ALWAYS use the anode of D4 as your audio ground in the FET/500.  It is electronically isolated from the chassis. Using the chassis as ground doesn't work and will give you worthless readings.

While testing TP18 as instructed in the guide (the schematics of the 500 and the Rack are different but I am fairly confident I was able to find the correct point to measure) I got readings of -2.63VDC with the Qbias fully CCW and -1VDC with it fully CW...Perfectly within the range specified in the guide! (for a Rev D at least, I'm doing Rev A but it was close enough to give me hope).

Additionally, rotating the Qbias CW caused the GR lights to go out one by one, indicating more reduction and in  line with the circuit's expected behavior. (I also realized Mike, that when responding to your first analysis I made the same mistake many had and assumed all GR lights on meant total slammed compression as opposed to the opposite so, no, it does not appear my unit is over compressing)

However, upon switching my ground to the chassis suddenly all voltage readings  the Qbias became pretty crazy.  I read -18VDC at TP18 and also -18VDC and -16VDC on the other two pins of the Qbias trimmer resistor.  I had been ready to move on to checking the amplification stages, but those readings are so out of whack that I followed the instructions in the guide to go back to the forum. 

ya never use the chassis as ground in the FET/500.  It's isolated and meaningless.

The thing is, the Troubleshooting guide is for the Rack version so it's unclear to me whether the chassis of my Radial Workhorse Powerstation is an appropriate place to ground my DMM for these troubleshooting tests or whether the "anode of D4" should replace the chassis ground in the troubleshooting guide when working on a 500 series module. Especially since they were very close to the expected range when using the anode of D4.

Always D4.

Still can't get any meaningful level at your input?  Are you sure you have the 2N5457 oriented correctly at Q1?

Mike

 

[boneindian]

If you have -2.1VDC at your Gate, that is good.

You just measured the voltage drop across the diode.


Look what I wrote above, ALWAYS use the anode of D4 as your audio ground in the FET/500.  It is electronically isolated from the chassis. Using the chassis as ground doesn't work and will give you worthless readings.

ya never use the chassis as ground in the FET/500.  It's isolated and meaningless.

Always D4.

Still can't get any meaningful level at your input?  Are you sure you have the 2N5457 oriented correctly at Q1?

Mike

Gotcha on D4,  I just got thrown because of how adamant the troubleshooting guide is about the chassis ground

Still getting the crazy voltage drop on the input.  Q1 is definitely oriented correctly as long as proper orientation follows the footprint on the PCB.

I’m gonna go ahead and proceed to the next section of the troubleshooting guide and see what I get since the Qbias readings were fine.

Is it worth it if I can’t get to .775VAC with my oscillator input to the comp under load?

I’ll report back with my findings. Let me know if you have any other ideas about what might be causing the input voltage drop!

 

[Echo North]

Ya you could at least see if you can get a decent level to TP15 which is the CW pin of the output transformer.  Maybe there is just some error in your process I can't figure out.  Trying to think why your input is dying.

Thoughts:

1. Relays not fully soldered/shorted
2. Full Bypass switch not fully soldered/shorted
3. Tpad not fully soldered
4. Input Tx not fully soldered

Here are the TP locations.

“AGND” at the anode of D4
“TP1” at secondary + of input transformer
"TP15" at CW lug of output pot
“TP17” at wiper of output pot
“TP22” at - side of C17
"TP 21” at - side of C21
"TP 19” at - side of CR2
“TP18” at wiper of R59

You could also send them in or 1 in for flat rate repair.

https://www.hairballaudio.com/build-support-services

Though in my experience when a customer has 2 units with the same issue, 50% of the time the errors are different.  Also keep in mind starting next Tuesday I will not be online or available for a week.

Mike

 

[boneindian]

Ya you could at least see if you can get a decent level to TP15 which is the CW pin of the output transformer.  Maybe there is just some error in your process I can't figure out.  Trying to think why your input is dying.

Thoughts:

1. Relays not fully soldered/shorted
2. Full Bypass switch not fully soldered/shorted
3. Tpad not fully soldered
4. Input Tx not fully soldered

Here are the TP locations.

“AGND” at the anode of D4
“TP1” at secondary + of input transformer
"TP15" at CW lug of output pot
“TP17” at wiper of output pot
“TP22” at - side of C17
"TP 21” at - side of C21
"TP 19” at - side of CR2
“TP18” at wiper of R59

You could also send them in or 1 in for flat rate repair.

https://www.hairballaudio.com/build-support-services

Though in my experience when a customer has 2 units with the same issue, 50% of the time the errors are different.  Also keep in mind starting next Tuesday I will not be online or available for a week.

Mike

Thanks for all this Mike, I seem to be getting 2.1VAC at TP 1. It’s the same pin with the dot on the circuit board to line up the transformer, correct?

Only getting 56mVAC at TP15. (Although in your first sentence you said the CW pin of the output TX—brown wire?—where I’m actually getting  250-900mv depending on the position of the Output knob)

I can reflow the solder on the Input TX and the T-pad but they look perfect, filled, and shiny at the moment.

What’s the best way to test the full bypass switch (which seems to at least be bypassing properly) and relays for shorts?

Not that this would explain the behavior I’m experiencing, but what’s the value of the T-pads currently used in this build? Mine is marked 500ohm-1k-500ohm. It’s one of the only parts in the BOM not listed with a value so I couldn’t check it against that.

I’d consider repair eventually, but I’ve built a stereo FET rack unit successfully and every stage of these two seems to be functioning properly (in terms of gaining up signals and reducing them) aside from this weird voltage drop at the top so I’m gonna keep at it for awhile longer. I was pretty methodical and careful with this build, I tested every component before stuffing and everything looks clean and complete. Looking back at component values on the stuffed board hasn’t revealed anything either.

 

[Echo North]

Thanks for all this Mike, I seem to be getting 2.1VAC at TP 1. It’s the same pin with the dot on the circuit board to line up the transformer, correct?

Your T-Pad is fine.  It's our custom part.

No that is the primary.  You want the pin across from it.

2.1 VAC is an insane amount of level, even for the primary.  You must be measuring something incorrectly at the input.  These reading are super weird.

What is your level at TP1 (across from the one you measured)?

Mike

 

[Echo North]

Only getting 56mVAC at TP15. (Although in your first sentence you said the CW pin of the output TX—brown wire?—where I’m actually getting  250-900mv depending on the position of the Output knob)

TP 15 is the CW lead on the output pot. Measure that again as well.  It should not change with rotation of the output.

Again, all measurements with respect to D4.

 

[boneindian]

Your T-Pad is fine.  It's our custom part.

No that is the primary.  You want the pin across from it.

2.1 VAC is an insane amount of level, even for the primary.  You must be measuring something incorrectly at the input.  These reading are super weird.

What is your level at TP1 (across from the one you measured)?

Mike

My level at TP1 (pin across from the one marked with dot, in input TX  pin row closest to the edge connector) is only 1.9mVAC. The pin right next to it on the secondary(one pin closer to the bottom of the chassis) is 184mVAC (only including that info since it's closer to what I should be seeing at TP1 and I have no idea what's going on at this point).

TP15 is still about 56mVAC

The troubleshooting guide indicated that my problem is somewhere at the input (T-Pad/Input TX)

My ground probe is firmly clipped with an appropriately sized clip to the anode of D4.  I tested for continuity on the pad on the other side of the board to be sure.

 

[Echo North]

My level at TP1 (pin across from the one marked with dot, in input TX  pin row closest to the edge connector) is only 1.9mVAC. The pin right next to it on the secondary(one pin closer to the bottom of the chassis) is 184mVAC (only including that info since it's closer to what I should be seeing at TP1 and I have no idea what's going on at this point).

TP15 is still about 56mVAC

The troubleshooting guide indicated that my problem is somewhere at the input (T-Pad/Input TX)

That pin is unconnected.  Probably just reading EMI/RFI.

No idea what the issue is.  The odds getting a bad input TX or bad T-pad are like 1/500, the odds of having 2 of them bad seems astronomical.

You have a signal generator?  Maybe try inserting a 100 mVAC, 1K signal at TP1, then measure at TP15.  Basically you're by passing the relays, tpad, and input and putting the level you should have at TP1 directly into TP1.

Mike

 

[Echo North]

You have the spacer under the input tx right?

 

[boneindian]

You have the spacer under the input tx right?

yep

 

[Echo North]

yep

Try the signal generator.  That will tell us a lot.