I made a little movie of the light bulb test to show you what's happening.
This is a 150w bulb.
This is a 150w bulb.
In fact the light bulb stay lit as soon as I connect the amp board to ground (Cgnd to Gnd output of the PSU). When I disconnect the Gnd on the amp board, the bulb lights, then dim to extinction and both fan and speaker relays works as expected.
I repeat - light bulb tester is not useful , as in case with F5 testing
every Aleph is from start already biased too heavily , even if not set to desired Iq and 0 offset
every Aleph is from start already biased too heavily , even if not set to desired Iq and 0 offset
so shall I try without the bulb tester? Is it the bulb tester witch prevent fan controller and relays to start when amp ground is connected?
I'm quite scared to fry something without it
I'm quite scared to fry something without it
Your connections appear correct. Nothing happens when ground isn't connected because there is nothing to turn on Q3, since Z5 has no current.
The problem you are having with connecting C-ground to the PSU ground seems like you have a fet or two shorted, or they are being turned on hard. Check for continuity between source and drain pins of any output fets, remembering that there is a diode in the body of the FETs.
Recheck continuity from Q8-12 Drains to chassis ground. I think it's time to remove the output boards and check for foreign matter.
One test option is to disconnect the output boards, install R19 and R22 along with one of your "spare" IRFP240s in Q6/7. Sort of a high voltage mini A. That will allow you to isolate the problem to the front end or output boards.
Re: Bulb tester, with the lower net voltage on the rails with a bulb in series operating voltages are thown off. Once you are absolutely sure that there is no short from FET to ground, go ahead and skip the bulb. If it makes you feel safer, install 10A fuses in the mains rails. They will blow if there is a catastrophic failure, although they may not save the FETs.
See my comment in post 203 about low voltage preventing fan and relay operation.
The problem you are having with connecting C-ground to the PSU ground seems like you have a fet or two shorted, or they are being turned on hard. Check for continuity between source and drain pins of any output fets, remembering that there is a diode in the body of the FETs.
Recheck continuity from Q8-12 Drains to chassis ground. I think it's time to remove the output boards and check for foreign matter.
One test option is to disconnect the output boards, install R19 and R22 along with one of your "spare" IRFP240s in Q6/7. Sort of a high voltage mini A. That will allow you to isolate the problem to the front end or output boards.
Re: Bulb tester, with the lower net voltage on the rails with a bulb in series operating voltages are thown off. Once you are absolutely sure that there is no short from FET to ground, go ahead and skip the bulb. If it makes you feel safer, install 10A fuses in the mains rails. They will blow if there is a catastrophic failure, although they may not save the FETs.
See my comment in post 203 about low voltage preventing fan and relay operation.
Last edited:
I rechecked continuity between Source and drain pins of every fets (Q8 - Q13) as well as continuity between drain and chassis: no continuity anywhere.
Double check the values of R13 and R14, and the proper orientation of Q4/5 before firing up. Gate connection (G+) to V+ should be 3K.
I read 1.45K for R14 and for R13 (witch is a 39k res) the resistance is slowly increasing during the test (quickly increase to 30K then slowly increasing after that, I stopped at 33.6k after 2min)
What do you mean by Gate connection should be 3K, how can I test this?
Again thanks a lot for helping me like this
Well, I have unmounted the amp and header boards to check the insulators and fets for foreign materials.... nothing to signal here. I've also replaced the 39K res by a new one, the measuring has the same behaviour as before (quickly rise to 28-29K then slows down and stabilize to 38K (maybe a little more if I waited longer), all the other res on the board return exact and instantaneous results.
I'll go to sleep and go on tomorrow, I think i'll give a try without the bulb tester, cross fingers 😀
thanks a lot folks and wish you a good evening.
I'll go to sleep and go on tomorrow, I think i'll give a try without the bulb tester, cross fingers 😀
thanks a lot folks and wish you a good evening.
The changing apparent resistance at R13 is due to capacitors in circuit, normal. R14 has nothing across it so doesn't change.
I meant to say check R14 and R15. Both are supposed to be 1.5K, in series from +rail to the +G. While you're at it, check the value of R20. I'm trying to ensure that the current source isn't upside down and turned on hard.
Are Q4/5 inserted correctly? It looks like Q5 is following the silkscreen. I remember you saying you were using BCxxx, which should go the other way around.
When you are ready to fire it up, have a voltmeter across R40 or R39. If you have multiple voltmeters, do both and across the output, looking for offset. If you have loads of meters, check the voltage from the negative rail to the -G connection. Be careful not to short anything if you are brave enough to move the voltmeter while the amp is on.
I meant to say check R14 and R15. Both are supposed to be 1.5K, in series from +rail to the +G. While you're at it, check the value of R20. I'm trying to ensure that the current source isn't upside down and turned on hard.
Are Q4/5 inserted correctly? It looks like Q5 is following the silkscreen. I remember you saying you were using BCxxx, which should go the other way around.
When you are ready to fire it up, have a voltmeter across R40 or R39. If you have multiple voltmeters, do both and across the output, looking for offset. If you have loads of meters, check the voltage from the negative rail to the -G connection. Be careful not to short anything if you are brave enough to move the voltmeter while the amp is on.
Last edited:
R20 = 1K
R14 =1.5K
R15 = 100K (if I look at schematics, it's supposed to be 100K, not 1.5K), In fact R16 is supposed to be 1.5K,
Q4/Q5 = In fact I put Ztx450 here, not the BC, I checked the orientation and it seems good, for exemple Q4 emitter goes directly to R17/R18 while Collector goes from Q2 drain.
R14 =1.5K
R15 = 100K (if I look at schematics, it's supposed to be 100K, not 1.5K), In fact R16 is supposed to be 1.5K,
Q4/Q5 = In fact I put Ztx450 here, not the BC, I checked the orientation and it seems good, for exemple Q4 emitter goes directly to R17/R18 while Collector goes from Q2 drain.
Last edited:
YEAH!!!!! it works!!!
Everything start flawlessly, fan rotating, relays clicking and fets warming!!! 😀😀😀
I have done some measurements:
Multilmeter on DC voltage mode
Output to gnd (200mv caliber) = (after 5min) 14mv DC slowly increasing
R40 = 0.202V DC
R39 = -0.200V DC
During the time taken to write this post, the sink is homogenously warm to the touch 😀 BURN BABY BURN!!! yeah!😎
Everything start flawlessly, fan rotating, relays clicking and fets warming!!! 😀😀😀
I have done some measurements:
Multilmeter on DC voltage mode
Output to gnd (200mv caliber) = (after 5min) 14mv DC slowly increasing
R40 = 0.202V DC
R39 = -0.200V DC
During the time taken to write this post, the sink is homogenously warm to the touch 😀 BURN BABY BURN!!! yeah!😎
Last edited:
Sounds like whatever was in your isolators disappeared when you disassembled it, or the bulb was keeping it from starting. Great.
Your bias seems very low. What is the voltage between the negative rail and the g- ?
Your bias seems very low. What is the voltage between the negative rail and the g- ?
Well, as I began to build the second board, I just came across a problem, Digikey sent 10K instead of 100K res (for R2 and R15) I don't know how I could have missed it on the first board, that should explain the low bias?
I'll check voltage between V- and -G.
I try to find some 100k 1% res, I can only find 1/4w in miniature format, will this be enough or shall I go for larger ones (1/2w)?
I try to find some 100k 1% res, I can only find 1/4w in miniature format, will this be enough or shall I go for larger ones (1/2w)?
those tiny ones are OK for the task
recheck all other resistors
if Iq is still low - increase value of R13 (ref. to http://www.diyaudio.com/forums/pass...d-aleph30-needs-some-help-11.html#post3829852 ) and chase 0 at output varying R18
recheck all other resistors
if Iq is still low - increase value of R13 (ref. to http://www.diyaudio.com/forums/pass...d-aleph30-needs-some-help-11.html#post3829852 ) and chase 0 at output varying R18
If I want to decrease the DC offset on the output, I have to rise the value of R18 right (witch is 392R now)? I plan to order 470 560 680 and
802R res
802R res
- Status
- Not open for further replies.