no sound from my aleph!!

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
greetings all,

i just finished an aleph5 and am getting no sound from it. i can hear the left channel hum a bit as soon as the amp turns on, then it fades a bit ( i installed the inrush limiters). nothing out of the right. everything is as the schematics say, however, i added a current mirror to each channel. it is a little cramped in the amp to poke around with a multi-meter. are there any crucial checks i could perform to minimize my poking around? or should i open the sucka up. my concerns are that i made the case following lottus blossum's schematic, so it is rather difficult to open the case. here's a little check list of where i am currently:

-amp built as per schematics, with the addition of current mirror.
-can hear slight hum from left channel, nothing from right
-slight smell (new electronic parts?)
-how long should it take for the mosfets to heat up?
- should i turn it on and let it stand for a while before poking with multimeter?

any help would be MUCH appreciated,
Generally if a project doesn't go you get the multi meter and start poking around which is going to mean opening it up. I normally check the operation of the power supply before connecting it up. Start with the power supply rails and work from there. Check voltage drops across resistors, that sort of thing. Its bound to be a bad connection - something simple. Just don't electrocute yourself.
amp blues

i checked some voltages and everything is fine upstream from Q5. i wonder if it's the current mirror, is anyone familiar with that circuit? if so, maybe we can debug that.
here's how i have it installed:

the collectors from the CM's are connected to the drains of Q1 and Q2.

the emitters are connected to the negative voltage rail via a 33 ohm resistor each.

the bases are connected to each other, then that is soldered to the source of Q2. (this was a little fuzzy for me)

does that sound right?
I'm assuming that the current mirror is on the front end differential. That alone could give you about thirty-seven different kinds of nightmares. Why? Gain went through the roof. Now you have stability issues to worry about.
1) First things first--get that amp off of your speakers and onto some load resistors. Do not pass go, do not collect $200. Do it now. (If the load resistors commence to getting hot without any input signal, you've got a problem.)
2) I'd start with a scope, looking at the output to make sure that there aren't any multi-megahertz oscillations running out the back door.
3) While you're back there, check for DC at the output. There probably isn't any, but why not check to be sure...
4) Check the internal voltages as per the schematic. Make sure that the gates to the output devices (the bottom set) are somewhere near where they need to be. Check the voltage across the Source resistors (R64-66) to make sure that you've got a reasonable amount of quiescent bias current going (the amp will run with a fairly wide variation here, so don't panic if you're a bit high or low). Check the front end current source to make sure it's working properly.
And, as Dave said, check connections, hoping that it turns out to be something simple.
--The MOSFETs on my Aleph 2s get palpably warm within a couple of minutes. Since you're almost certainly using air-cooling, yours will heat up more quickly than mine.
--For voltage testing purposes, you can start as soon as the amp is turned on. The voltages will settle in somewhat as it warms up, but you're not looking for tenths of a volt, you're looking for something much larger.

i rewired the current mirrors and somehow i now get DC out of the right output. and still no sound at all.
i need to do some serious circuit testing. the problem is that i don't have a scope. only a multi-meter.

i checked the power supply and all is well.
should i take out the mirror and install the original resistor?

i should tell you that i have a pair of old crappy drivers that i use to debug stuff with. they can tell me whether it's working or not. is that okay instead of a load resistor?

any suggestions on what to do next?

thanks all,
I would say yes, remove the current mirrors and put the origional resitor back in. You will then be able to check the voltages across each channel as per the voltages listed in the service manual.

Then once the amp is up and running, maybe look at putting the current mirror back in, but as Grey has pointed out it looks like they maybe more troble than their worth.

Lest you guys think I was a bit off the beam in my earlier post (more than usual, that is), I'd like to note that the network at work fell to its knees. I wrote that just after Dave's post, and spent the rest of the shift trying to get it out. I see that a copy got through, but was a little behind the times. Oh, well.
--Current mirrors are nothing but two current sources in tandem, and I'm sure you guys know by now how I feel about current sources...I vote for the resistor. (Jam will show up in a minute to put in a word or two in defense of current's a long running thing he and I tease each other about.)
--There was a thread about modifying Alephs a while back wherein someone figured they'd come up with a New and Original thought--Golly, why not use current mirrors in the front end of an Aleph, since Nelson was silly enough to leave them out. Nelson weighed in and said that he'd tried them. Yes, they measured better. No, they didn't sound as good.
--Speakers, crappy or otherwise, tend to have a bad hair day when faced with DC. I'd toss a resistor on the output. They're cheaper and can take more punishment.
--You don't mention voltages in the circuit. I'd start checking voltages. I'm real curious as to what the drive is going into the outputs.

let's see, here's what i've measured so far. i removed the mirrors and have the resistor back in.
these measurements are at referenced points from the service manual.

my supply voltages are ~32 volts +/- at the rails

voltage across zeners ~9.03 vdc.

voltage across Q5 ~ 5 vdc.

voltage measured at the outputs....... ~30 vdc!!! ?????

virtually no VDC at the ouput mosfet resistors (1 ohm)???

i'm thinking that bad ground scheme might be at least part of the problem. here's how i have it:

earth-thermistor-input (neg)-chassis

the filter caps are ganged with the tranny to another spot on the chassis.

that's it for grounds.


What voltage do you have across the drain resistor of the differential?
I strongly suspect that you might have a wiring error or defecive mosfets.

Sorry I am late but I will have to agree with you on this one. The current mirror hurts the sound of the amplifer (I have tried it) much like Nelson says. Now on the other hand if you look at the current source on the tail of the differential .............
as for the mis- wiring, i got the artwork for the boards from mark finnis and was really careful when populating. i also checked the solder conxns. and was really careful about static discharge around my hexfets. i really hope that it's something other than those possible problems.

cheers (hopefully),
Stop the presses!
Hang on a've got both (Q1 & Q2) Drains tied together?
(Shuffling sounds of Grey grubbing for a schematic so as to get part numbers straight...)
The input MOSFET (Q1) shares a node with R14, the collector of Q4, C6, and R52-54.
The 'backside' MOSFET (Q2) Drain goes to the negative rail. Only. It does not go through R14, nor does it connect to Q4, etc. It doesn't need a load of any kind, as you're not trying to take an output from its Drain, so it doesn't need anything to work against.

let me try this again. this is my second project and i'm not as precise with my descriptions as i need to be. sorry about the confusion.

i have the drain of Q1 indeed sharing a node with r14 and then q4, c6 and the mosfets.

Q2 drains directly to the negative rail.

i hope this gets us in the right direction.

Scott, I experienced a similar issue with an Aleph 30 monoblock I built - output at +ve voltage rail, almost 0 current through the resistors in the output stage. I wonder if the problem is the same?

In my case, I tracked down the problem to the drains of Q1 and Q2 being shorted via their cases. I'd cleverly used mica washers to insulate the cases of the 9610s from the heatsink, and then has used metal bolts to attach them to the heatsink. Duh!

This shorted the drains of the 9610s together via the heatsink. So the lower half of the output stage was effectively off (drive voltage was at the -ve rail), which is why there was no current flowing through the output stage resistors (I had no output load, remember).

What voltage do you have at the gate of the MOSFETs in the lower half of the output stage, relative to the -ve rail? Is it more than 3.8-4v?

Another test would be to check if the current through your load resistor is the same as the current though the upper half of output stage. If so, the lower half is not being driven on.

Hope this helps.
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.