Aleph-X builder's thread.

Jump to Latest
M

muhy3

Member
Joined 2004
Eric said:
My first thought is to check your Q2 mosfet(s) and make sure they are COMPLETELY isolated from your heat sink. Use your DMM, there should be no connectivity between any of the legs and the sink. You may need to remove the device from the board first and depending on your implementation, this could be a lot of work.

Also, check out the voltages around your input differential and the output mosfets between the two channels. This might provide some clues about where the errant voltage is appearing.
Click to expand...

Thanks for the feedback.

An externally hosted image should be here but it was not working when we last tested it.


Fortunately I have set everything up as per the photo for a "test" run.

I don't get enough time to check the input diff or the volts between the output mosfets before the windings inside the resistor start to glow red.

I will also need to order more 0.22ohm resistors now :) although the current ones haven't burned all the way, I was quick enough to switch everything off.

I have replaced the 9610's, so the next thing is perhaps I have fried the MOSFETS? I can't think of anything else really, both boards are identical and both have different issues...
 
M

muhy3

Member
Joined 2004
Eric said:
Before you replace too many parts, you need to positively identify the problem - otherwise the new parts will cook as well... Check your mosfet mounting for shorts to the sink first.
Click to expand...

One of the wires to the 044 was the wrong way around. I double checked all four mosfets now, all wired OK. Oh, and there was a short, no idea how since I used insulated pads and compound, but I fixed that too - good idea guys!

However, perhaps during this fiasco I did some other damage because now R6 and R41 start to warm up a fair bit (within 5min getting too hot to touch).

Any suggestions? Thank you :)
 
M

muhy3

Member
Joined 2004
Dennis Hui said:
Looks like you're making progress. Congratulations.

Are you getting about 0.5V across R6 and R41? With them at 0.22 ohm, you're
getting over 1W . Depending on the resistors they could get fairly hot.

Cheers,
Dennis
Click to expand...

Hi Dennis,

Thanks :) and also the email support hehehe

I'm able to get 0.5v across R6 and R41 now no problem, they are 0.22ohm resistors 3W.

They no longer heat up, but now R4 and R45 heat up, because I am getting too much absolute DC offset, >12v and VR2 has no impact (i.e. there is no voltage change at all from changing VR2).

I also notice that Q2 and Q11 do not heat up, but Q1 and Q10 do.

-Vlad
 
Eric

Eric

Pass DIY Addict
Joined 2000
sites.google.com/bucknell.edu/ericshometheater/erics-home-theater
Paid Member
Excellent- you're making progress now.

First things first: google search on dim bulb tester. Build one and use it to power up your amp. A variac would help, but the dim-bulb tester will be cheaper and still allow you to see if things work without frying parts so quickly. For testing, you might want to measure as Dennis suggests: hook up your volt meters, flip the power on, make a quick reading, then power down right away.

You said you had a short in your 044, but you didn't say which transistor it was. I presume it was one of Q2 or Q11...

Q2 and Q11 are not heating because of your off-balance bias to Q5 and Q7 as measured by voltages across R23 and R25. I would double check R10, R13, R35, R39, and replace Q4 and Q9 as a starting point. It may be necessary to replace Q5/Q7, but start with resistor vales and then Q4/Q9.
 
Last edited:
M

muhy3

Member
Joined 2004
Eric said:
Excellent- you're making progress now.

First things first: google search on dim bulb tester. Build one and use it to power up your amp. A variac would help, but the dim-bulb tester will be cheaper and still allow you to see if things work without frying parts so quickly. For testing, you might want to measure as Dennis suggests: hook up your volt meters, flip the power on, make a quick reading, then power down right away.

You said you had a short in your 044, but you didn't say which transistor it was. I presume it was one of Q2 or Q11...

Q2 and Q11 are not heating because of your off-balance bias to Q5 and Q7 as measured by voltages across R23 and R25. I would double check R10, R13, R35, R39, and replace Q4 and Q9 as a starting point. It may be necessary to replace Q5/Q7, but start with resistor vales and then Q4/Q9.
Click to expand...

Thanks Eric, it's Saturday morning here in Sydney and I'm at work until 1pm, but will be home tonight and give your suggestions a try. I've double checked all the component values, they are as per the design and R10,R39 = 221 ohms, and R13,R35 = 100 ohms. I am suspect of the MPSA18's at Q4 and Q9 too, but didn't want to replace too many parts (I have spares, since I bought enough parts to build 4 x AX's).

Will let you know how I go :) thank you!
 
M

muhy3

Member
Joined 2004
Dennis Hui said:
Hi Vlad,

Do you recall how the matching for the diff pair Q5/Q7 was done?

Another thing I noticed is that the total current through R23/R25 seem a bit low,
even with V2 at its lowest value. Can you confirm your part values for D1 and R17?

Cheers,
Dennis
Click to expand...

I purchased matched Q5/Q7 from Jim's Audio. I have a few spares, so I'll set up tonight and match them myself and replace once matched.

D1 = 1N5239B and R17 = 1.5k ohm

Thanks!
 
M

muhy3

Member
Joined 2004
So I bought a bench top PSU and went through to try and match Vgs for the 9610 from Jim's Audio... turns out none of them matched! I had a stack of others, and matched a couple exactly, to like within 0.001v. It didn't seem to make a difference to the circuit. To me it seems like the overcurrent protection on Q9 is being triggered with the volts reading 0.65 across R39... but 0.15 across R13 :(


Sent from my iPhone using Tapatalk
 
Eric

Eric

Pass DIY Addict
Joined 2000
sites.google.com/bucknell.edu/ericshometheater/erics-home-theater
Paid Member
Yes, D1 should provide forward voltage of about 9v (give or take a few tenths). It is critical that Q5 and Q7 be as closely matched as possible. 0.001v is about as good as it gets. If you were using poorly matched transistors the first time around, this could be part of the problem of large offset that won't move with V2 and overheading output resistors.

I'm still suspect of Q4 and Q9 given the short you found.
 
Last edited:
Rodeodave

Rodeodave

Member
Joined 2004
Paid Member
I have a question regarding the output loading resistors (R1/R4 and R44/R45 on the Aleph-X rev 1.0 pcb). As I'm building my Aleph X specifically for driving planar headphones (HE6), the load impedance seen by the amp will be a flat 50 Ohm. How should I size the loading resistors? From what I've gathered, the typical value used is usually about four times the impedance of the speaker's impedance, so about 32 Ohm for an 8 Ohm speaker. In the case of a 50 Ohm impedance that would amount to 200 Ohms, which would practically barely load the output... So, what do? 100 Ohm maybe?

Also, about the McMillan resistors (R46/R47), is there any sense in trying to omit them? Or are they most definitely needed to insure low and steady DC offset at the outputs?
 
Eric

Eric

Pass DIY Addict
Joined 2000
sites.google.com/bucknell.edu/ericshometheater/erics-home-theater
Paid Member
I don't have definitive answers for your questions. My R1/R4 and R44/R45 are set to about 78R with five 390R resistors in parallel. Guess this puts me somewhere above 8x my speaker impedance. As for the McMillan resistors, there was a great deal of discussion earlier in this thread or the other huge Aleph-X thread. The discussion that I remember is that they solve a problem related to offset stability (though I don't remember the details at this point). You can try building without them, but my guess is that you'll end up putting them in anyhow.
 
You must log in or register to reply here.
Top Bottom