ProtoDAC TDA1387 X8 project

[kLOkUIS]

I soldered a second Protodac together but wasn’t successful.
After measuring the resistors I found out that the 2 output resistors had not the correct value.130k Ohms instead of the 430 Ohms. Mistake from the shop, but I should have checked them before soldering.. I replaced them with the correct value, but without success.

Could I have damaged some components powering it up with the wrong resistors in place?

[hifinet]

I am not sure. The 130k resistors would generate a much higher than normal voltage at the analog outputs. I wouldn't think this would damage the module. Play a 50Hz 0dB sine wave file and measure the DC and AC voltages at pins 6 and 25 of the module with a DMM. Also, be sure the I2S resistors (R1-R3) are correct.

[MarkC568]

Finally received another 1387 module, so rebuilt my number two DAC, which was having distortion in one channel. (The number one build, a kit from @hifinet has always worked flawlessly)

The existing 1387 module was being uncooperative to removal, so I decided to salvage the polycarbonate caps and z-foil resistors to start fresh with the new 1387 on a blank board.

I was able to salvage the RCA jacks by breaking the old board out from around the large solder posts. This resulted in only having to unsolder the through-hole plating left attached to the posts.

Both channels are distortion-free on this new build.

[Tim Curtis]

Nice :-)

Is that a DIY Amp?

[hifinet]

Very nice looking system.

For the kits, I test all of the TDA1387 X8 modules, and reject about 7%. That's a high failure rate due to the fact that the chips are used and recycled. I would recommend buying extras if you are building on your own.

Also, prepare for failure. Solder only the active pins on the module (1,2,3,5,6,25,28). Test the setup. If everything sounds good, then you can solder the rest if you like.

I think the easiest way to remove a bad module is to clip the soldered pins with a flush cutter. Then unsolder the individual pins from the PCB and clean up the pads.

Using a desoldering tool (some, like the Hakko, are very expensive) or heat gun can also work.

Try to work with the existing PCB as much as possible, because removing components like the GPIO, RCA jacks and module is difficult. Fortunately, the most expensive parts are the Z-foil resistors, which are easy to remove, but also very easy to damage.

Another thing to mention (and it probably should be in the Builder Guide) is to test the connections after soldering the GPIO header. Using a DMM, be sure you don't have shorts between 5V power (pins 2 and 4) and ground (pin 6) and also 3.3V (pin 1). Consider not soldering pin 1 (or just remove pin 1 from the header), since it is unused. A short between 3.3V pin 1 and 5V pin 2 can kill the RPi. 

It is hard to get shorts from solder bridges of the GPIO pins due to the solder mask of the PCB, but possible if you use too much solder.

[Tim Curtis]

I'll update the guide later today.

[hifinet]

Thank you.

[MarkC568]

Nice :-)

Is that a DIY Amp?

It's a kit from AliExpress that I took a chance on not being junk. Chose this one in particular in that it was simple, no transistors, had no bluetooth/screens/meters, and was a push-pull type so it has reasonable output power for the price.

https://www.aliexpress.us/item/2255800100204524.html

Didn't come with any instructions, just components, chassis, wire, parts, tubes, and a schematic. All was well packed and arrived undamaged. PCB was nice quality, with markings for all mounted components.

Was a fun build, and I'm surprised at how good it sounds. Only a few minor issues:

- Power transformer primary is wound for 110v, causing the heater voltage run too high on 122v wall power. Added a couple of resistors to bring that down.
- Output transformer secondary wires were labeled incorrectly, had to swap ground/8R connections. 
- Came with weird off-brand electrolytics, so I built with Nichicons instead
- RCA inputs are not great. I have a better pair to swap in at some point.

Bonus additions:
- Bleed resistors on the filter caps for safety
- Added pilot on the front panel

[hifinet]

Off topic, but from reading diyA posts on instability problems with Ali tube amps, incorrectly labeled output transformer leads seems to be very common. Usually the primary leads are reversed, which causes positive feedback and oscillation. Best to have a dummy load on first power up.

[Nutul]

Usually the primary leads are reversed

I am confused here... how can a transformer have its leads "reversed"?!? We're not talking DC here, guys...