Aleph J illustrated build guide

I knew you would beat me up about that. But you know me. I just had to pair them. :rofl:

To rephrase: does difference between MOSFETs affect need for range on pots? I guess the answer is yes, but not in such a degree that it is relevant :clown:

No singing. Heating right now. Still have half a psu left. Maybe run with just one? Naaah
 
Last edited:
The noise that will not go away, part 3

Back in post #6861, and again in post #7540, I discussed my issues with some noise in the right channel of my Aleph J.

This time, per responses and recommendations, I pulled the offending board from the chassis and reflowed most of the solder connections. I put it all back together, adjusted bias and offest, and once again had noise-free sound, this time for about another month.

And, like a horror movie villain, it's back. Same details all around. Not signal dependent at all - the noise can be heard over music being played, or with nothing connected to the amp at all. It's all DC noise coming from the right channel speaker outputs. I connected a multimeter and get a dc spike every time the noise occurs.

DC offset was at 0 - 1 mV when I powered the amp on, then spiked to around 22mV with the noise, then settled back to around 9mV and held.

Asking the community again for any additional insight into what might be going on here. I chopsticked all of the connections on the board again, and can't get the noise to occur. Also, when I hear it, if I poke around on the board, the noise doesn't change.

Maybe a bum DC offset pot? Bad cap?

It's driving me nuts, again. Any help would be really appreciated.

Thanks,
Brad
 
The noise that will not go away, part 3

Back in post #6861, and again in post #7540, I discussed my issues with some noise in the right channel of my Aleph J.

This time, per responses and recommendations, I pulled the offending board from the chassis and reflowed most of the solder connections. I put it all back together, adjusted bias and offest, and once again had noise-free sound, this time for about another month.

And, like a horror movie villain, it's back. Same details all around. Not signal dependent at all - the noise can be heard over music being played, or with nothing connected to the amp at all. It's all DC noise coming from the right channel speaker outputs. I connected a multimeter and get a dc spike every time the noise occurs.

DC offset was at 0 - 1 mV when I powered the amp on, then spiked to around 22mV with the noise, then settled back to around 9mV and held.

Asking the community again for any additional insight into what might be going on here. I chopsticked all of the connections on the board again, and can't get the noise to occur. Also, when I hear it, if I poke around on the board, the noise doesn't change.

Maybe a bum DC offset pot? Bad cap?

It's driving me nuts, again. Any help would be really appreciated.

Thanks,
Brad
 
Uh oh

Assembled my amp, all seemed well, PSU tested fine and bias and DC offset were set on the amp boards. Then I thought, wouldn't it be nice to have a front panel switch to power the rails and leave the back panel switch for standby mode. Exposing my ignorance, I got myself a nice anti-vandal ring illuminated (blue, of course) pushbutton switch, 5A 24V LED and temp wired it into the V+ circuit using wire nuts.

Tested first using the dim bulb limiter. V+ switched on and off and switch ring illuminated nicely and V- remained on regardless of switch state. Was wondering if that was a good idea, but decided to go full power. What could happen, maybe blow the switch?

Nope. Switch is fine, but smoked R2 on the right channel. Upon removing the switch and replacing R2, noticed it and R4 were both the wrong value - 221 instead of 221K. Replaced R2 & R4 on both boards and checked all other component values. They're good.

Powered up again with dim bulb, all looks good. Power up full. PSU 23.5V both V+ and V-. Left channel seems OK. LEDs light, heat sinks get hot, bias is stable at 0.4V after an hour, but DC offset is finicky. Can't get it stable below 1V. Sadly, right channel is badly broken. LEDs light, but heat sinks are cold, 0 voltage drop across the source resistors, DC offset is 20V and does not respond to R7 adjustment. Figure Q1A & B are done, but have not tested them yet. Was hoping someone might weigh in on what I should look for before I blindly start removing and replacing parts.
 
Official Court Jester
Joined 2003
Paid Member
Back in post #6861, and again in post #7540, I discussed my issues with some noise in the right channel of my Aleph J.

This time, per responses and recommendations, I pulled the offending board from the chassis and reflowed most of the solder connections. I put it all back together, adjusted bias and offest, and once again had noise-free sound, this time for about another month.

And, like a horror movie villain, it's back. Same details all around. Not signal dependent at all - the noise can be heard over music being played, or with nothing connected to the amp at all. It's all DC noise coming from the right channel speaker outputs. I connected a multimeter and get a dc spike every time the noise occurs.

DC offset was at 0 - 1 mV when I powered the amp on, then spiked to around 22mV with the noise, then settled back to around 9mV and held.

Asking the community again for any additional insight into what might be going on here. I chopsticked all of the connections on the board again, and can't get the noise to occur. Also, when I hear it, if I poke around on the board, the noise doesn't change.

Maybe a bum DC offset pot? Bad cap?

It's driving me nuts, again. Any help would be really appreciated.

Thanks,
Brad

time for dumbest and in many cases , most effective method - start replacing part by part, starting with semis

input JFets - replace both of them at once, with matched pair, of course

if you want to go with hope of having bad cheap part, start with small and cheap semis


Assembled my amp, all seemed well, PSU tested fine and bias and DC offset were set on the amp boards. Then I thought, wouldn't it be nice to have a front panel switch to power the rails and leave the back panel switch for standby mode. Exposing my ignorance, I got myself a nice anti-vandal ring illuminated (blue, of course) pushbutton switch, 5A 24V LED and temp wired it into the V+ circuit using wire nuts.

Tested first using the dim bulb limiter. V+ switched on and off and switch ring illuminated nicely and V- remained on regardless of switch state. Was wondering if that was a good idea, but decided to go full power. What could happen, maybe blow the switch?

Nope. Switch is fine, but smoked R2 on the right channel. Upon removing the switch and replacing R2, noticed it and R4 were both the wrong value - 221 instead of 221K. Replaced R2 & R4 on both boards and checked all other component values. They're good.

Powered up again with dim bulb, all looks good. Power up full. PSU 23.5V both V+ and V-. Left channel seems OK. LEDs light, heat sinks get hot, bias is stable at 0.4V after an hour, but DC offset is finicky. Can't get it stable below 1V. Sadly, right channel is badly broken. LEDs light, but heat sinks are cold, 0 voltage drop across the source resistors, DC offset is 20V and does not respond to R7 adjustment. Figure Q1A & B are done, but have not tested them yet. Was hoping someone might weigh in on what I should look for before I blindly start removing and replacing parts.

switching DC rails is job for heavy duty relays

besides that, that's no-no

BIG NO-NO ........ you're not playing with bulbs, that's amp

point us to your reference schm, so we can talk more
 
Member
Joined 2016
Paid Member
To bpr, re: noise

I had been chasing some noise on my SissySIT R.3 a few weeks back ... started exchanging parts until the noise went away.

If you are starting replacing parts, Zeners are also probable candidates. Sometimes you just get a noisy one.

As Zen Mod said, small transistors are candidates. And I would look at the pots as well.


Regards, Claas
 
The noise that will not go away, part 3

I have some thoughts on chasing that trouble. After chasing noise on telephone lines for longer than I care to admit. There is one thing I have learned. Divide and conquer. I would lift the legs on C5 and R4 facing J2 and C1. This will open the feedback loop. The free leg of R4 will need to be grounded to the same ground as R2. You should be able to trace the foil on the back side of the board. Both the - in and the + in need to be grounded. With an AC coupled oscilloscope or a signal tracer I would look at the Drain of Q1a and the base of Q2 for noise , if noise is found the trouble is most likely in the front end ( Q1a , Q1b , D1 or Q2 ) you will of course need to heatsink the output FETs Q5,Q6,Q7,Q8 while running this test. If no noise is found then I would go to the output stage. Pulling Q7 and Q8 would leave the current source Q5 and Q6 with no load. A large resistor, and I mean large could load the current source in place of Q7 and Q8 . If the noise is still there than the current source is the problem. Replacing Q7 and Q8 would be easier than putting a resistor to load the current source. Hope this helps .
 
Bpr: Dennis once advised me to use a scope for such issues: Following the signal path from start to finish, trying to find out where in the circuit the noise occurs, and hence likely bad parts. I didn’t try it then, ended up building a new one (Wayne’s linestage), hehe. but maybe that’s worth a shot before pulling it all apart.

Edit: I’d follow ZM and ESL’s advice. If you have the equip for the latter.
 
Last edited:
This is how I connected the switch.
 

Attachments

  • Aleph J Switch.jpg
    Aleph J Switch.jpg
    591.9 KB · Views: 191
time for dumbest and in many cases , most effective method - start replacing part by part, starting with semis

input JFets - replace both of them at once, with matched pair, of course

if you want to go with hope of having bad cheap part, start with small and cheap semis

Replaced Q2 - Q4 last night, and the channel stopped working entirely - no voltage at all across the resistors at the MOSFETS, zero heat on the MOSFETs and the heatsinks.

Tonight, replaced the JFets with the single matched pair I still had on hand, and also replaced all of the MOSFETs. And, nothing. Channel still doesn't work at all. At least with the noisy channel, I could still listen to music. The board is getting power properly, but that's about it at this point.

So, should I get a new set of JFets and try again? Oh, and to answer an earlier question, I don't have a scope, and at least for this project I'm not planning on picking one up, so if there's help to be given, keep that in mind.

Not sure what next step to take. Any help is greatly appreciated.

Thanks again,
Brad
 
Member
Joined 2012
Paid Member
If it was working prior to replacing Q2 - Q4 and stopped working after replacement, then most likely one or more of the replacement parts are faulty. At that point the thing to do would have been to remove those replacement parts and check to see if they were faulty.

The replacement of the other parts before addressing the Q2 - Q4 issue would not have fixed the problem. That actually complicates the issue as the subsequent parts replacement may inadvertently cause more problems if one or more of those replacement parts are also faulty.

So at this point you should remove Q2 - Q4 and check to see if they are faulty, and replace if necessary.