SOLVED Cattle Driver - low output volume

[MattG]

I just built the Cattle Driver. It works, except the output volume seems way too low to me. Unity gain is just past 3:00 on the level control (i.e. a little past 75%). I'm thinking it's probably not supposed to be like that. I think it's working as-expected otherwise, though I can't be 100% sure, since I'm just going off online videos of the Cattle Driver and the Buffalo TD-X (of which the CD is supposed to be a clone). Other than the volume issue, there's no other obvious problems.

Does anyone else happen to have a Cattle Driver that's willing to take voltage readings for me to compare with, particularly around the op-amps? Otherwise, what's the most likely culprit to start with for debugging this?

Thanks!

 

[twebb6778]

Take a close look to see if you've got any cold solder joints and verify all resistor values are correct. Can you post some pics of our board?

 

[MattG]

Thank you for the quick reply! I'll go through and double-check the resistor values.

Here are my voltages:

At power input: 9.37 V

TL072:
pin 1: 4.24
pin 2: 4.24
pin 3: 3.85
pin 4: 0
pin 5: 3.86
pin 6: 4.24
pin 7: 4.24
pin 8: 8.46

LM301:
pin 1: 1.44
pin 2: 4.21
pin 3: 3.83
pin 4: 0
pin 5: 1.36
pin 6: 4.21
pin 7: 8.46
pin 8: 3.61

Edit: Resistor values check out. I also swapped in different op-amps, and re-measured pin voltages - only negligible differences compared to those I listed above. Looks like I'll have to pull it out of the case.

Here's a pic of one side of the board, in the enclosure. If all the resistor values check out, and I don't see any other obvious problems, I'll remove it from the case and post more pics of the other side of the PCB.

Thanks again!

 

[twebb6778]

Not that it should matter too much, but is there any reason you used poly box caps over electrolytics?

 

[MattG]

No particularly compelling reason for using film caps over electrolytics. But I spent a long time over at diyAudio doing audio electronics (though with a focus on music reproduction/playback, as opposed to guitar-specific), and developed a neurosis for avoiding electrolytics in the signal path. In theory, film caps ought to last longer than electrolytics (though I know that's not really a legitimate concern in this application). A minor practical benefit: I don't have to worry about polarity!

If nothing else, they look cool.

 

[okstateblues]

If you can, get good lighting on the underside of the board when taking pictures. Cold joints recently bit me on my recent build.

 

[MattG]

I removed the board from the case. I won't have time until tomorrow to un-solder the two lower pots that block the view on some of the board. But from what I can see, I don't see any obvious issues. Before I took the pictures, I thought there may have been some tiny solder bridges, but I cleaned all those up, yet the problem remains. I'm using this fancy Cardas Soldering Wire Quad Eutectic Silver Solder, which is supposed to greatly reduce the chances of a cold solder joint.

Let me know if you see anything I missed. If no one here notices anything, when I remove those pots, I'll probably re-touch all the existing solder joints.

Sorry, these pics aren't the greatest - it was hard to get the camera to focus properly and have everything well lit. If the problem persists, I'll have to get out the DSLR and tripod.

Thanks again!

 

[okstateblues]

Do you have an audio probe? I recently made one for my simulcast build and it helped me super fast. After you have it built you will have to follow the schematic through the audio path to track down where the signal dramatically decreases. If you need more info, check out my simulcast build in the troubleshooting section.

 

[MattG]

Do you have an audio probe?

I do now! I built one, and tracked the problem down to what appears to be the output (level) pot. The solder joints all looked fine, so I went ahead and replaced the pot.

Unfortunately, the problem remains. It's maybe slightly better now, unity appears to be just before 3:00pm, whereas before it was just after 3:00pm. But that could be explained by pot tolerance.

Any thoughts?

Thanks again!

 

[okstateblues]

Have you checked all of your values for your resistors/caps? Glad the probe helped but it still sounds like there may be a value that is off. When you probed along the audio path, did you get a drastic change in volume?

 

[MattG]

I checked all the resistor values, but can't check all the cap values unless I un-solder them. For now, let's assume they are correct, since I really try to triple-check component values before soldering (not that I've never made a mistake!).

Anyway, I spent some more time with the audio probe, as yesterday, I basically only probed around the opamp inputs/outputs. So, again with the probe, this is what I'm seeing:

The output of the LM301AN (pin 6) is noticeably louder than the input signal. So taking that as my starting point:

• There is a definite volume drop across C10 (330p). This is where I'm not sure what to expect, as it's part of the tone circuit, which I assume should result in some volume drop.
• I hear the same volume drop across R7 (22k).
• Both sides of R8 (1k) have the same volume to my ears. This is the "reduced" volume, i.e., similar to the "quiet side" of C10/R7. This is the series resistor just after the tone stack.
• On the TL072, pins 5, 6 and 7 all have the same volume. Pin 7 is the output, and it looks like that opamp is configured to have some gain (presumably tone stack recovery). So should I expect pin 7 to be louder than pins 5 and 6?

I guess I need a little more hand-holding, let me know what else I can probe that might give a clue.

Thanks!

 

[Nostradoomus]

Are you sure those are 47pf caps at the top? And 150p in with the 150n at the bottom right?

 

[Barry]

Are you sure those are 47pf caps at the top? And 150p in with the 150n at the bottom right?

Those 47pf look huge for picofarads

 

[MattG]

Are you sure those are 47pf caps at the top? And 150p in with the 150n at the bottom right?

Dang, I think you caught the problem!

I just reviewed my Mouser order, the 47pF caps at the top (C12, C13) are correct, unless Mouser sent me the wrong part. I ordered Wima FKP2O100471D00JF00. I don't have a C meter to verify, but they are labeled "47/1000-", so I'm going to assume those are right. Those in particular I did extra checks on, since I expected them to be smaller, being in the pF range. I'm guessing it has to do with the dielectric: all the smaller ones (2.5mm width) are MKS2 (Polyethylene Terephthalate), but I couldn't find that series in a 47pF. So C12/13 at the top are FKP2 series (Polypropylene).

But C3, the 150p, I mis-read that as another 150n. So C3 is almost certainly off by a factor of 1000. I'm assuming the function of that cap is to roll off some highs, and a too-big value probably rolls off way too much of the desired frequencies?

I checked all the resistor values, but can't check all the cap values unless I un-solder them. For now, let's assume they are correct, since I really try to triple-check component values before soldering (not that I've never made a mistake!).

Well, now I look like a complete fool! Got a fix for my injured pride?

I'll address C3 and report back. (Not sure if I have a 150p on hand, might have to order it.)

 

[Nostradoomus]

Ah they’re huge because they’re 1000v, those should be fine. If you don’t have a 150p cap put something close like 100 or 220p and see what happens.

 

[MattG]

I found this 120pF cap in my parts bin. Now unity gain is around noon - is that expected? That seems like an intuitive spot for unity, so I'm thinking I'm probably good now?

120 (versus 150) is probably close enough to call it a day, right? Is the following correct for the purpose of C3: it acts as a high-pass filter, so frequencies above the cutoff are kept at unity gain, and below are forced through the voltage divider circuit, which results in them getting amplified? In other words, this cap ultimately results in some high frequency rolloff for a last bit of tone shaping of the final output signal? How do I calculate the cutoff frequency in this arrangement? I'm guessing 120 vs 150 doesn't make a huge difference. Now I'm struggling with perfectionist neurosis versus laziness of wanting to call it done.

It's also a 1kV cap, keeping with my deliciously overkill cap theme.

Thank you so much for catching that! I managed to misread that 150pF countless times... I even made my own BOM spreadsheet to work off of. Where's the facepalm emoji?

Anyway - again, thank you!

 

[Nostradoomus]

I’d call it a day if you’re happy with it! And yeah it’s a roll off cap.

 

[MattG]

And yeah it’s a roll off cap.

So the other thing I noticed when debugging (and didn't mention): the bass pot seemed to have a more significant impact on the volume than I would expect. At the time, I chalked it up to a particularity of the circuit. But now I realize, with that incorrect 150nF in the output opamp's feedback loop, only bass frequencies were actually getting amplified. That should have been a clue!

Anyway, thanks again for your help, much appreciated!

Edit: Also, that cap shouldn't affect opamp pin voltages, right? So the voltages I initially posted should be valid for comparison, if anyone else has to debug their Cattle Driver?

 

[Nostradoomus]

What are your voltages now? Pin 1 and 5 of the LM301 look a bit low originally.

EDIT: Pin 5 doesn’t look connected in the schematic...it should be 0v

 

[MattG]

Here are the voltages now:

Supply voltage: 9.36

TL072:
pin 1: 4.23
pin 2: 4.23
pin 3: 3.84
pin 4: 0
pin 5: 3.84
pin 6: 4.22
pin 7: 4.22
pin 8: 8.43

LM301:
pin 1: 1.45
pin 2: 4.12
pin 3: 3.77
pin 4: 0
pin 5: 1.35
pin 6: 4.12
pin 7: 8.43
pin 8: 3.52

Essentially the same as before.

For the LM301, either my build has another bug, or maybe that opamp naturally puts some voltage on pin5?