Blown Amp/Mixer

[Agent24]

Re: Blown Amp/Mixer

I had a feeling that would probably be the case...

[Trinite]

Re: Blown Amp/Mixer

Those crooked heatsinks need to be set so they cannot touch...those are NOT isolated TO220 devices on those heatsinks. You have a short circuit hazard.

I checked. Both heatsinks are grounded, so them touching is not an issue. The 7815 and 7812 regs each are grounded through the center leg, so their contact with the heatsink again poses no issue. Otherwise, I would have straightened them out; but in this case, the work was unnecessary and a posed a risk of PCB damage, so I left them as is.

[hardwareguy]

Re: Blown Amp/Mixer

Those crooked heatsinks need to be set so they cannot touch...those are NOT isolated TO220 devices on those heatsinks. You have a short circuit hazard.

[Trinite]

Re: Blown Amp/Mixer

Also replaced all the non-name caps in the amp section, which got rid of a lot of the AC hum, along with grounding it. Actually sounds quite decent now. In any case, it's a much nicer unit than when I first started.

Just for laughs, I uploaded a pic of the warning label to show the quality of the "Engrish" used, which also implies the original build quality of the unit.

[Agent24]

Re: Blown Amp/Mixer

Nice job!

[Trinite]

Re: Blown Amp/Mixer

Been busy lately, but finally got around to fixing this unit. Replaced 2 op-amps, 2 burned out resistors, added a bleed resistor on the high voltage section and added a ground-lift switch.

Ended up learning quite a bit about audio systems in the process as well.

[Trinite]

Re: Blown Amp/Mixer

Here is a diagram of the circuit that provides the +60V to the phantom power reg.

Note: Ground reference is provided by the transformer center tap and is not referenced to the -20VDC output.

[Trinite]

Re: Blown Amp/Mixer

Seems to me The 15V reg is being propped up by the 33v zener for 48V Phantom power So 40v input is wrong.

Waste of time and money upgrading the op amps on this thing, replace with any cheap dual device NE5532 or whatever is in your parts bin.
Those burned resistors would have been fuseable to save the main regs.

The input power is 40V (-20V/+20V). However, the phantom power uses two capacitors coupled with the AC inputs and is supplied with a +20VDC bias. So the reg is actually getting 60V.

Also, I do not know if the previous resistors were fusible, since they were burned beyond recognition, but I am replacing them with fusible ones.

[Th3_uN1Qu3]

Re: Blown Amp/Mixer

replace with any cheap dual device NE5532 or whatever is in your parts bin.

NE5532 will oscillate if you drop it directly where a 4558 used to be, it's happened to me. Better just keep it safe and get 4558s.

[Hotiron]

Re: Blown Amp/Mixer

i'm starting to wonder how the hell this thing lasted as long as it did...?

40V on that 33V zener as a half-assed attempt at limiting the voltage going to the 7815...
zener took a crap, 7815 got the full 40-ish volts and decided to 'splode a little..

Seems to me The 15V reg is being propped up by the 33v zener for 48V Phantom power So 40v input is wrong.

Waste of time and money upgrading the op amps on this thing, replace with any cheap dual device NE5532 or whatever is in your parts bin.
Those burned resistors would have been fuseable to save the main regs.

[Trinite]

Re: Blown Amp/Mixer

The unit consists of a total of 20 4558s. 8 on the inputs, 4 on the output, 7 on the EQ and one for the headphones. I could replace them all, but knowing this unit the circuitry is probably quite poorly designed. Seems like I should just get some 4558s then.

[Th3_uN1Qu3]

Re: Blown Amp/Mixer

4558s? Aw man... they sound like ass.

Now, the problem is that they are STUPID stable (because of their low bandwidth) so if you upgrade them to higher quality ones and the circuit is designed poorly, the "better" opamps may oscillate. Been thru it myself.

The 4580 is *slightly* better, but since they're both dirt cheap i'd just get how many you need of both of them. If the 4580s cause trouble you can go back to the 4558. If all the mixer is full of 4558s just buy 4558s - coz it's still gonna sound like a 4558 even if it passes thru only one of them.

[Trinite]

Re: Blown Amp/Mixer

Looks like this is quite the trainwreck. I couldn't get sound out of the speakers, so I took the front panel off the unit and found two toasted resistors. These supply a positive and negative voltage to an array of opamps, one of which had had failed shorted. The opamps are PJ4558CD - couldn't find them on Digikey so replacing them with NJM4558D. The resistor codes were unreadable; fortunately, the value (47ohm) was printed on the board.

Thankfully parts are cheap, otherwise this would be quite the moneypit...

The PCB does indicate a 4580 chip should be placed where the 4558 chips are. Would the more expensive 4580 be a better choice in this application? The 4558 has a GBW of 3MHz and the 4580 has a GBW of 12MHz, while the original chip only has a GBW of 2.5MHz.

NJM4558: http://search.digikey.com/scripts/Dk...NJM4558D%23-ND
RC4580:http://search.digikey.com/scripts/Dk...296-15368-5-ND

[Trinite]

Re: Blown Amp/Mixer

It looks like it's been cut with a hacksaw.

You know, it's quite the "hack" job. Definitely "cutting" corners

Anyway, I got another hack job shown in the uploads. In this one it looks like they made a mistake in the PCB layout and went about cutting traces and running wires to reroute the connections. I also did my own hack job to try and fix the zener diode-VR hack job. I installed a 3A fuse in series with the zener, so if something does go wrong again, it will only knock out the 48V phantom power to the mics, instead of taking out the whole unit. Also replaced the 40V filter caps with some nice Nicichon audio caps; also replaced some of the other cheapo brands.

Got it up and running. Seems to be working alright, although the VR heatsinks are quite warm. Next up will be to test the audio functions.

[Trinite]

Re: Blown Amp/Mixer

Might just put a fuse in series with the zener. Also might add a bleed resistor, as the capacitors were still holding charge after 48 hours, so I decided to discharge them with a pencil lead

http://www.youtube.com/watch?v=bmTtuj9WKKM

[Th3_uN1Qu3]

Re: Blown Amp/Mixer

highest i've found is 35V

digikey: LM341T-15-ND

Which is exactly what the 7915 supports in the first place so you still need that pesky zener.

[Agent24]

Re: Blown Amp/Mixer

High quality, made in China!

You know it's el-cheapo when you see those multi-patterned ribbon cables

[Scenic]

Re: Blown Amp/Mixer

highest i've found is 35V

digikey: LM341T-15-ND

[Badcappy]

Re: Blown Amp/Mixer

You can replace the 7815 with a 15V regulator that can withstand a higher input voltage.

[Th3_uN1Qu3]

Re: Blown Amp/Mixer

Use a 2W zener. Or two of them 1Nwhatevertheywere in parallel.