Mine is blown. Mosfet in the power supply blown open, blown gate resistors on the psu mosfets, blown capacitor in the RC network on the secondary side. blown output mosfet. etc.
Interesting that they use 3x IRF9640 mosfets and 2x IRF640 mosfets in the amp output section and of course one of the 640's blew. why the asymmetrical number of output devices? did they just run out of room in the chassis LOL!
The Class-D controller IC, I am trying to figure out what it is. I removed the label and of course they scratched off the number under it but not all the way. it's a 24 pin SMD device and I can clearly read the first letter of the part number as a "D" the second letter looks like an "L" or maybe an "I" and i can see several numbers that follow but i can't quite read them.
So anyone know of any class-D control IC's that start with DI-xxxxx or DL-xxxxx???
it's a 9v relay and nope, no voltage across it. the relay works. I tested it with a bench supply and the surrounding parts are good. I'm sure it burnt up what ever parts that sense the output for DC. But I got tired and will look at it again tomorrow. I haven't figured out how they sense the output yet. or...I am assuming it has a protection circuit of some sort to sense DC at the output. It may just be a time delay or may be some function from the class-D controller IC. But I can view the signal ahead of the relay and it looks fine.
poking around with it today I managed to get the protect relay to kick in...have no idea how, but I noticed that there is a 4.7uf cap in the power supply section that gets really hot! it appears to be a bipolar EL cap...sure wish i had a schematic....
ANYONE have ANY Eclipse schematics?? maybe one has a similar TL494 based power supply??
It seems i forgot to check the driver transistors for the power supply mosfets and one was shorted. so the amp would run, but the TL494 was driving a 10 ohm load on one side so it was drawing excessive current from its power supply drawing it down to the point where there was only 6v to try and pull in a 9V relay! once I fixed the near shorted driver transistor the amp ran fine and pulled in the speaker relay just fine etc. BUT. I was just running it with one IRF640 and one IRF9640 output mosfet as that's all i had on hand until i could order some. so after getting it running. I pulled out the singles and put in matched sets and well now it doesn't run. I pulled the board back out and I don't see any shorts. But I was tired and i just called it a night.
The only other thing I can think of...and I have NO idea why this would be a factor. is that the singles were genuine IR parts. the replacement 640's are IRF640N and the 9640s are Vishay/Siliconix parts all sourced from Digikey. so no idea why the N vs non N parts would make a difference or why the Vishay parts would make any difference...but...I must have missed something...tomorrow...but it's progress at least....
GOT IT FIXED!!!! turns out it way my error after all sort of... I Managed to find some genuine IR 640 and 9640 Non "N" Parts and i started off by putting in one 640 and one 9640 and it ran. so I tried 2x 640's and noticed that the turn on surge was a bit higher but it again ran. So 2x 640 and 2x 9640's and it again ran but this time it had a much bigger turn on surge!
So 3x 9640's and 2x 640's (this is the way the amp is built) and the amp won't run....But seeing as how the turn on surge got bigger and bigger each time I added a fet I started to suspect that my little bench supply was really the problem. so I dig out the big supply and low and behold the amp fired right up and ran just fine! Walla! done!
All back together and managed to squeeze a solid 400 watts at 4 ohms out of it at 100hz. I would call that good..
I got mine fixed too. I would just like to do on record and post the symptoms for future benefit. I finally got mine to turn on about every other power-up. Half the time it would power into protection and the other half it would work. I finally thought I would replace all the outputs. That did the trick. I think one of the bunch could have been leaky. The IRF640s (2) didn't even have the same date code. I think that might do it for Eclipse for me.