Aleph J illustrated build guide

Member
Joined 2019
Paid Member
Well, congratulations on post #5000 :D Have to stay positive. :rofl:

It looks like at least two of them have given up their lives in the name of DIY. For peace of mind, I'd still replace all four. My reasoning is that the Mega-Testers don't put the parts through the same voltage / current they'll see in normal use. Even though two parts are identified properly as an N-Channel MOSFETs, I'd still be wary. Plus the piece of mind of an additional 5 dollars for new parts is worth it to me, you've already got the parts out, and the legs are already bent up a bit.

re: matching - I think you're describing the method for checking the Idss of the input JFETs with the 9V battery. To the best of my knowledge, which is pretty limited, you cannot match the output MOSFETs with your mega tester. There's an excellent article on the First Watt site about testing MOSFETs.

http://www.firstwatt.com/pdf/art_mos_test.pdf

That's a bit technical, but excellent reading. Personally, I found this one a bit easier for me to follow.

http://www.firstwatt.com/pdf/art_matching.pdf

For all practical purposes, many people use a bench supply capable of 22-24VDC @ up to 1A or so, a few 3W or higher rated resistors, and a DMM. Many people recommend mounting the parts to a heatsink for measurement.

Some answers to specific questions -

re: parts - See your PM. Mouser, Digi-Key, and Arrow, were showing out of stock in the US. I tried to buy some a few days ago. If you can get them, grab them. If you know you'll want to build more Pass FW amps in the future, and if budget allows, get 30 of each of the IRFP240 and IRFP9240. The only ones still made, I think, are branded Vishay. I had read a history of the parts awhile ago, but I don't remember the differences. Others that have been in the hobby longer and/or have better memory can chime in. Either way, the Vishay will be just fine. You only need the 4 for this board. No need to match between channels, the cool autobias circuit takes care of that (sort of).

re: checking other things while you've got them out. I checked back a bit, but could you upload, clear, well-lit photos of both the front and back of the board, please. There are some eagle-eyed members around that can spot a mis-stuffed part or anything else suspect very quickly.
 
Last edited:
Member
Joined 2016
Paid Member
I have built an Aleph J-variant with IRFP140. Unpaired, just one per channel in the lower position and one in the Aleph current source position.

The heatsinks I used are comparable to 4U/400.

I have run it with either 2.2A / 19V or 1.9A / 24V. Works fine, temperatures on the MOSFETs seem OK. Heatsinks were about "5 - 10s hand-hold time".

I had used thin Mica + good Goop that time. The "Aleph J single" is currently dismantled, because my M2 runs in that chassis, but when I rebuild it (hopefully later this year), I will use Keratherm.


Hope this helps,
best regards,
Claas
 
Almost ready for final assembly

I'm now just waiting for some hardware (bolts) to arrive in the post.#

The Aleph J is almost ready for assembly.

Since my last post and learning from some other posts, the 300VA 15-0-15V transformer has been replaced with a 500VA 18-0-18V.

The woodwork has received a piano black finish.

I've removed the fan so it is now just a chimney.

You might question why the Aleph J is on an aluminium plate instead of being mounted directly on the heatsink. The simple answer is that the heatsink appears to have become hardened and I was breaking 2.5mm drills trying to tap into it.
 

Attachments

  • 20200316_135510.jpg
    20200316_135510.jpg
    807.9 KB · Views: 640
  • 20200420_101723.jpg
    20200420_101723.jpg
    997.2 KB · Views: 632
Member
Joined 2003
Paid Member
I pulled the MOSFETs from the bad board and put them in the Mega 328 variant I have. The MOSFETs with the bubbling on the back tested as resistors, while the smooth ones tested as n-channel MOSFETs. I take this to mean that the two MOSFETs with the bubbling on back are toast. I checked with the DIYAudio store and they don't have any matched sets, nor do they expect to have any soon (3-4 months). Mouser does have the MOSFETs at about $2.32 each. How many do builders recommend I purchase to match them? 10? 20? Or does anyone have a matched quad they be interested in selling me?

If I do need to match them myself can I use the Mega or do I use the 9v battery/DMM method Jim demonstrated in several places (M2x forum, Pearl2 forum), or something else? Also, I don't know which variant I have now (ON Semi or Vishay - I purchased them from the DIYAudio store). The spec sheets of each is a little different, particularly the voltage of the part. I'm not sure if this matters.Do all eight need to be matched or can each board have a matched quad that may not match the other board?

I'm worried that the toast MOSFETs are a symptom of something else, but I don't know what else i can test while the parts are out.

Any suggestions are greatly appreciated.

If you get entire tubes of them, I believe 25? (call and ask) they should be from same batch and are usually fairly close matched as is.

If you plan on other Pass builds, you will use them or others here will buy.

Russellc
 
I pulled the MOSFETs from the bad board and put them in the Mega 328 variant I have. The MOSFETs with the bubbling on the back tested as resistors, while the smooth ones tested as n-channel MOSFETs. I take this to mean that the two MOSFETs with the bubbling on back are toast. I checked with the DIYAudio store and they don't have any matched sets, nor do they expect to have any soon (3-4 months). Mouser does have the MOSFETs at about $2.32 each. How many do builders recommend I purchase to match them? 10? 20? Or does anyone have a matched quad they be interested in selling me?

If I do need to match them myself can I use the Mega or do I use the 9v battery/DMM method Jim demonstrated in several places (M2x forum, Pearl2 forum), or something else? Also, I don't know which variant I have now (ON Semi or Vishay - I purchased them from the DIYAudio store). The spec sheets of each is a little different, particularly the voltage of the part. I'm not sure if this matters.Do all eight need to be matched or can each board have a matched quad that may not match the other board?

I'm worried that the toast MOSFETs are a symptom of something else, but I don't know what else i can test while the parts are out.

Any suggestions are greatly appreciated.

I have to check my stash of matched devices. But I should be able to help you.

PM me which what you need.

Cheers,

Greg
 
You might question why the Aleph J is on an aluminium plate instead of being mounted directly on the heatsink. The simple answer is that the heatsink appears to have become hardened and I was breaking 2.5mm drills trying to tap into it.

You may already have taken care of this, but if that mounting arrangement is going to be permanent, I'd put a good layer of thermal grease between the main chassis and the plate that the board is mounted to. You want to make sure you have good thermal contact between the 2 pieces. :)
 
Member
Joined 2003
Paid Member
Just noticed a little noise in one channel

As header says, was putting Iron pre in system with Aleph J where BA3 FE did reside. All this trying out of different pre had me listening closer than normal and while paused listened carefully at speaker, way close up, ear next to drivers. This amp has previously been quite, totally quiet.

Left side, dead silent. Right side, a little noise. I say "noise" because it seemed a little higher in frequency than 60 hz, and could also hear a little fizz from tweeter area.

Now this is very low level nothing unless ear right up close, nothing anywhere near listening position. Still yet, it used to be dead quiet.

I first blamed it on Iron Pre, not even in box yet, and ground consisting of wire from Iron pre pad to alligator on Aleph J ground.

Trading the BA3 FE back in, same result. Now for best sleep favorite amp must be restored to dead quiet. Also about to do the half and half mono block thing with this amp and M2 and I want to go into that in healthy shape.

Doesn't interfere with listening pleasure perse, but now that I know it's there it will bug me silly.

Any ideas before I start everything from checking wire routes, which didn't change to rotating transformer and if necessary, building new boards!

Oddly, I just ordered a quad of matched J74 in BL from Punkydawg...don't think I will need them here.

Thanks,

Russellc
 
Last edited:
Member
Joined 2013
Paid Member
I'm neither good at this, or smart enough to explain things, however, I've had good success fixing things by replacing the capacitors in some circuits.

Since the noise is such a low level, hard to believe it is from the input since it would be amplified. C3 and C6 are on the output of the circuit,therefore I guess them.
 
Member
Joined 2003
Paid Member
Well, the first test clears the amp. With shorting plugs in RCA Jack's, amp is dead silent on both sides.

As both preamps had the same deal, I don't think it is preamp related. That leaves the source. Laptop streaming WASAPI, into Audioquest D/A converter. Now some cable swapping channels to narrow it down.

Oh well, glad amp is still quiet!

Thanks for tip, Dennis

Russellc