bringing up aleph 3, no smoke, but no sound

[cowanrg]

well, thanks to some help from Magura, i have successfully powered on my aleph3. it doesnt smoke, everything looks ok, but there is no music. the speaker makes a hum (not really quiet, but not loud either. almost thought it was the transformer REALLY humming, but its the speaker). my rails are (under load) at around 22.5v or so. when they first switch on, its around 25-25, but then drops down a bit.

the thing that concerns me the most is that only one heatsink is getting warm... (i have two heatsinks, two transistors per sink). only one gets hot. the other gets warm, around 10F above room temp, but the other gets around 40F above room temp. not too hot to touch, but for sure warm. i didnt know if this is normal or not?

the transistors are matched, they are from my aleph2 project, so i have TONS of matched ones. i checked all connections and values for parts, they all appear fine. the light on the amp powers on, and nothing pops. (im using KRISTIJAN's boards, http://web.vip.hr/pcb-design.vip/aleph-3.html ).

hum, what else could i say to help out... OH, ive got my ground connected to the caps (using 2 26,000uf caps for testing), then to ground of rca input, and ground to speaker post. i tried connecting it to the ground pin on power cord, but didnt make a difference.

also, on the output of my source (its actually the headphone jack on a cd player with a volume knob), the volume does NOTHING to the level of the buzz. i can go from min to max and the buzz is the same. also, plugging and unplugging the source doesnt do anything, there isnt even a pop or a snap, so its like there isnt even an input.

any help is appreciated. i can take pics (albeit pretty low res), if it helps. its kinda a mess though, its all hooked up with alligator clips right now because the case is in the mail.

edit:

i am using a 250va transformer with dual 18v secondaries, into two 35a/600v bridge rectifiers, into two 26,000uf caps (for now), then into amp.

[Magura]

The easy solution to that is to start measuring the two sides and comparing them. Once you know where there is a difference, you will have some sort of guideline to get further.
A silly problem ive seen before is that the amp actually dont get any input signal. So to begin with id check the input side...does the amp actually get a signal?

Magura

[MikeW]

What is the DC offset? Do you have the proper voltage drops across your source resisters? Do you have the 4 volt drop across your 392 ohm resister?

[till]

its all hooked up with alligator clips right now

until today i do not really understand why, but when i built Zen V4 it did not work but smoked resisors and did HF oscillation wired with alligator clips. It did work ok with thick copper wires soldered instead of the clips.

[cowanrg]

hah, i could answer those questions if i knew what you were asking!

i know how to measure the DC offset, just measure the + and - terminals on the speaker posts with no input right?

the others im not 100% sure how to measure?

[cowanrg]

so, today do you understand why then?

i might have to try that out...

Originally posted by till


until today i do not really understand why, but when i built Zen V4 it did not work but smoked resisors and did HF oscillation wired with alligator clips. It did work ok with thick copper wires soldered instead of the clips.

[Brian Donaldson]

If the two sides are not the same temp, then they are not burning the same watts. That means you likly have a considerable offsett (dc voltage across the speaker terminals)
If the pos half is hotter, you likely have a problem in the signal side(neg half) like shorted output fet or input stage problem.
If the neg half is hot, the current source is likely current limiting the neg side and has a problem.

If you have a scope, a check of the outputs could tell if it's oscilating. I had a problem with my minialeph headphone amp where the input pair was oscillating, but the outputs didn't and you couldn't hear it (500,000 hz) but it made the thing sound muddy and undefined.

If it gets hot and the input has no effect on the output, I would first check for the correct voltages in the input current source and diff pair. Mosfets are quite fragile outside of their circuits.(static can kill one without you knowing)

[cowanrg]

Magura is helping me out online...

we have tested the current draw, check... tested the bias of the fets, not so good.

they are: 22.31V, 22.33V, 17.43V, 17.43V. however, we dont know how to fix it. i am of course useless in figuring it out

[MikeW]

DC offset, Take a volt meter attach one end to the output the other to ground.
Measure from one side to the other of the sourcre resisters. This will tell you the current flowing though your output MOSFETS.
The drop across the 392 ohm resister Lets you know if Q1 is biased properly.
It's a start.

[cowanrg]

MikeW,

well, measuring DC offset across the speaker terminals (output and ground), was 17V!!!!

i know enough to know thats not good.