Assembly, Problems, and Murphy's Law

[Mr. Man]

I assembled a T15 recently and after checking my work, successful testing with a voltmeter, I decided it deserved giving it a go. And plugging it into several sound sources I got total harmonic distortion.

So the next day I pulled it out of the chassis gave it a look over and found R5 and R33 were in each others place. Right, that's not a problem to remedy. Out comes the soldering iron, and with it murphy's law.

I had placed the resistors in their correct place and under the assumption that I may have missed something else I had another look over it to see that everything was in the right place. It was. Pulling the resistors out of tight spaces was bad, but putting them back was murder. I started burning the PCB and making a hash of everything. Ouch. And having it back together, I tested it with a meter.

The reading from PGND and the +20 was a hair under 19volts. It's well within the 1volt tolerance that the instructions specify, so I check the negative side and it read all of -1.9volts. No, that's not a typo.

Puzzled, I checked my work quite carefully (for the third time). Nothing is wrong with the order of the components on the board yet it refuses to read something on the meter that warrants plugging an XLR into it. I have a suspicion that this is a result of a faulty component on the PCB.

Anyone have any ideas on why I would, with incorrect resistor configuration at first, read correct voltages and why the correct component configuration would read only -1.9volts?

[tpryan]

You say the voltages were correct before you swapped R5 and R33? If the voltages were OK, why mess with those resistors?

The local regulator voltages are set by the ratio of R30/R31 and R33/R34. I'd start by making sure those resistors are the correct values. Make sure you have continuity between R33, R34, and U6, pin 1. You may have broken a trace removing R33. Let us know what you find.

[Mr. Man]

Yes the voltages were correct before I swapped the resistors. And I swapped them because I was getting only total harmonic distortion and then when I re-inspected it noticed R5 and R33 were in each other's place.

I'll give it another look over. Do you have any ideas on voltages or resistance values I should read in those areas?

[Mr. Man]

There is good continuity between R33 R34 and U6. The trace is starting to separate at R33 but it still holds good on the ohmmeter. Which brings me back to my theory on a faulty component. 16 volts is still disappearing off to somewhere, but where?

It's starting to become bittersweet for me. I also have the A12 and was absolutely gobsmacked at how good they sounded. On the other hand I have more time and money invested in this that I'm wondering just how much it's worth to my non-expendable money if I'm to buy and assemble these with a so-far 50% failure rate?

[tpryan]

What resistance do you measure across R30, R31, R33, and R34?

[Mr. Man]

Well it doesn't matter any more because after reading the "Be Careful w/ R30, R33, & R37!" thread I went to test the output mA at the phantom circuit under the assumption i could have R37 and R33 switched and when I put my meter probes across pin 1 & 3 of SW2, R35 burned up, along with my remaining patience on this piece.

[tpryan]

Suit yourself. You can proceed systematically and help us figure out what's actually wrong, or try things at random and hope to get lucky. The systematic approach will likely get you a working module sooner.

[Mr. Man]

Hang on, then.

I bother myself to read the other threads and go back and test the circuit to see that "aha!" R30, R33, and R37 have the same color codes. So I power it up and check the phantom circuit to find whether the numbers are in a discrepancy with the instructions. And guess what, it IS incorrect -along with the readings at R33. Meaning that I soldered the 15k resistor where a 120ohm resistor should be (and vice versa).

It just so happens that while testing, R35 burns itself out. So, please don't be too quick to write me off as some random whiny customer who can't be bothered to follow the instructions or simple common sense. It's very easy for people to whine; If I came across that way, I'm sorry. As a possible solution to you with having to deal with this particular problem, I'd recommend that you add an addendum at step 4 of the instructions to check R30, R33, and R37 with a meter when placing on the PCB and not only rely on the bill of materials as they appear to have the same color code.

So I'm going to order myself a replacement resistor for R35, switch R33 and R37 and it should be perfect.

I'm sorry you have to deal with idiots like me and trying your patience. I'll try to be less of one in the future. Thanks for your help.

[dkatz42]

The instructions do recommend testing all of the resistors with an ohmmeter, but it might be good to add a warning in the same spot about color codes that are valid when read in either direction.

These were never an issue in the days of 5% and 10% resistors, but I can only imagine that there was a fight between the standards folks wanting a unique stripe color for 1% resistors on human factors grounds, up against a pushback from the manufacturers on how much it would cost to be able to apply a 13th paint color. Or something...