USAmps 2000x

[Dr Zeus]

Ive got this amp infrom a customer which I'd like to try and fix. From the looks of it, theres a bunch wrong with it, starting with the way in which the heatsink is designed with a 3/16 inch groove running the entire length behind all it's transistors. This may have caused this amp to fail since there was possibly areas which the transistors make poor or even no contact with the heatsink.

This amp was also modified to where the proteciton circutry was disabled: D119, D120, and D121 were clearly clipped in half with a pair of dikes. This amp was probably over-loaded with subwoofers which eventually burnt out and back-fried most of the amplifier's 24 power supply fets I75344P. I believe IRF3205 should work as a replacement. This amp already has 47 ohm gate resistors which need replacement. All of the output transistors are either MJE4353 or MJE4343 and measure 11.5 ohms across leg 1 & 3.

During its final run, this amp likely melted it's USAmps equiped flexiglass cover and therefore this amp was brought to me without it. I beleive Lowes sells plexi in bulk.

This is going to be fun if the repair is feasible. Looks like the parts are going to cost over 100$.

 

[Dr Zeus]

Q114 which is one out of the 2 pairs of PS driver/power transistors marked MJE15030 shorted across all legs. Removed. Also removed all PS fets which measure shorted.

I applied power just ground and power to the amp and my power supply started to drag down to 10 volts. After ~5 seconds Q417 an MJE4343 output went POW and now measures ~0ohms between legs 2&3. Seems this amp wants to power without REM activated.

Q420-Q425 are MJE4353 - These are now measuring all bad on the PCB
Q320-Q325 are MJE4353
Q414-Q419 are MJE4343 - These are now measuring all bad on the PCB
Q314-Q319 are MJE4343

 

[Dr Zeus]

What the heck was USAmps thinking? It is not like this amp has really that much technology goin on in it, but; the way everything is soldered on the PCB makes me really think twice about fixing it, until I really figure out the best way how not to ruin the vias. Its like every leg and lead of every component is sharply bent in alternating directions and soldered against the pcb underneith. Since the components are very closely mounted to the PCB, this leaves poor chance at straightening the leads before de-solder to save the vias. I've not seen this before in any other amp.

I better put this amp down until I have more time to figure out how to work with it.

 

[Dr Zeus]

 

[Perry Babin]

Someone emailed me about an amp like this. I told him that I didn't know anyone that would work on US Amps amplifiers other than US Amps (and they wouldn't repair this one). I bet I'll know one more person that won't work on them when you finish this one.

Cut the legs off flush with the top of the board and remove the remaining leg from the bottom.

 

[Dr Zeus]

Its sure going to be a challenge if it is even possible. I'll pull the rectifiers and try getting the PS section working first.

 

[Dr Zeus]

Q411 is an audio driver, MJE15031 which shorted across all legs. Removed.
Found a bad 0.10 ohm 5% source resistor at R437 reading open. Removed.

 

[Dr Zeus]

The two 25v 220uf caps in the center of this picture between the diodes and the reglators do like like having ~45vDC across their leads and tend to want to smoke when the amp is powered on a 5A of 13vDC.

I got the amp to 1/2 power up on one power supply. its cranking some major un-regulated voltage to the now-removed rectifiers. I believe since the second power supply is currently toast theres a good possibility I'll have to either fix that supply or just remove the caps for now.

I dont know what these three diodes are supposed to be. Supposedly this is the protection circuit which commonly gets modified on these amps so that they can run .5ohm and blow up like this one did. The owner said he had it disabled, LOL and now wants to protection circuitry back so the amp will run 2~4 ohms. Any idea what these would be? They are all connected to Pin 9 of the IC next to them: a Motorola brand PIC16F84A / 0142HBR.

 

[Perry Babin]

US Amps should be able to give you the part numbers for the diodes.

 

[Cr@sh]

I have one of these if you need me to test anything for you. I bought it new in 1999 and it has performed flawlessly since. Best amp I have ever owned.

 

[Dr Zeus]

Hey Crash thanks for helping this out.

Any chance you can see what kind of diodes I'm missing in the above photo? There are three small diodes in close proximity to the 18pin IC which on my amp someone clipped as a 'modification'.

 

[Cr@sh]

Sure can, I have to pull it out from the car. I planned on doing this today so give me a day or two.

 

[Dr Zeus]

If you are willing, otherwise I'll just send an email to USAmp. I dont want you going out of your way partner.

 

[Cr@sh]

You'll have to contact US Amps. I see no numbers on them. Mine is the green board as well. Shouldn't be any difference but anyway they'll help ya. I just contacted them a month ago about my USA50x and they got back to me within days.

 

[Dr Zeus]

I forgot to update, but I heard back from USAmps. They said that once those diodes are clipped, the SAT control unit needs to be re-programmed and is usually not worth it. Beats me why clipping diodes would re-program nvram. Ah well.

I already spoke with this amp's owner and this repair is going to be an as-is but a working repair, if I can find the parts.

I'm trying to source a good place to buy MJE4353 outputs. The complementary MJE4343 Arrow stocks which I also need, but no-where I can find through OctoPart any good non-NTE brand MJE4353.

 

[Perry Babin]

If you can't find them, the FJA4313 and the FJA4213 may be good subs. I have never tried them as a sub for the motorolas.

 

[Dr Zeus]

Into re-building the damaged PS section. I replaced a damaged PS driver; MJE15030 and also the two 100ohm 1/2w resistors just next door. Removed all the bad fets and some of the bad 47ohm gate resistors. Replaced some of the gate resistors with 47ohm brand new units, and installed 4 IRF3205 fets; one in each corner of PS section of the board.

I still have all the primary/rail rectifiers removed from circuit.

Power-on heated the fets instantly. Upon inspection, I notice the fets also heat instantly with only Bat+ and ground connected to my power supply. Looknig further, leg 2 of the fets are directly connected to bat+, while leg 3 f the fets are directly connected to ground. Does this sound like there might be a shorted transformer?

 

[Perry Babin]

When you say 'directly connected' do you mean it reads 0 ohms to the B+? That's normal. It's connected via the primary transformer windings.

What's the DC voltage on the gate leg of each of the FETs?

 

[Dr Zeus]

I found the other PS driver, an MJE15031 reading OL off the board so I replaced it with a new MJE15031G. I am now able to get this amp to partially power up with the rectifiers removed, but there are two small 25v, 220uf electrolytic caps next to the LM7915CT and LM7815CT regulators which still want to explode as mentioned before. Installed, these caps do not like having 55vDC across them. Once removed, the voltage across them is only 22vDC.

With them both removed, the LM7915 gets hot very quickly, but I am able to measure decent regulated voltage on the opamps.

 

[Perry Babin]

It may be a regulated power supply that's producing too much output (to the rectifiers for the regulated power supply) with the rectifiers out of the circuit.