Help With Aleph

[Jay]

Due to experimentation with IRFZ44, my Aleph was broken (+Vcc at output). After fixing and replacing the transistors (I switched to IRFP250s on another working heatsink), alas, I found a distorted sound coming out from the Aleph. It seems like the bias current is too low or something. Many of the component values are different from standard Aleph, but they worked well before.

What is the minimum current expected flowing through the Source resistor in the signal section (not ccs)? I found 2.8V on 1 Ohm.

Thanks...

[Jay]

No No No! It's not 2.8V! Damm, what was that value?

[Jay]

It was 0.38V...

[Jay]

It was 0.38V on each 1 ohm resistor in 2 paralleld MOSFET.

[grataku]

When you get done talking to yourself, please put down some info we can use to help you, if that's what you want.

Please describe in detail the mods you made, what kind of distortion you see etc.

[UrSv]

One useful piece of info would definitely be which Aleph you are talking about...

[Jay]

In fact I’ve never built an exact Aleph as Aleph 3 or 5 or 2. All materials are installed on test bed. I can use 1 or 2 or 3 or 4 or 5 MOSFET in parallel simply by wiring. I have several working power supplies, 40V, 33V, 30V and 27V.

I had a working Aleph with 3 (or 2) IRFP250, whichever the power supply. I assembled IRFZ44 on another heatsink, and I moved the PCB to the IRFZ. The sound was more detail in low volume and low power supply voltage. I then accidentally cut the feedback resistor (10K) off while the amp was on. This one created the +Vcc on the output.

I moved the broken circuit to the previous heatsink (of IRP250) and
1) Replaced all the IRF9610s (to measure Vgs of my other MOSFETs)
2) Replaced MPSA42 with MPSA18 (To conform with standard Aleph
3) Replaced feedback cap from ROE 470/40 to PUREISM 470/16 (more slim).
4) Replaced ccs resistor (the one connected to the Vcc) from 4K7 to 2K2!
5) Paralleled differential stage ccs ground resistor to change from 10K to 5K.
6) Disconnect a 470uF connecting the Aleph ccs (after the 2K2) with output (before the output resistors) and connected it instead to output (after the output resistors)

That’s all. I don’t think that such changes should create such a problem. I will be more comfortable if there were no sound at all!

I assumed that it’s about current. I haven’t checked the amount of current drawn from the power supply, but the main heatsink is only slightly hotter than usual, but the IRF9610 do run very hot! (This one I didn’t expect!). The IRF9610s run very hot with only 10mA flowing to the input IRF9610 and 10mA+ flowing to the feedback IRFP9610.

[Jay]

I remembered now why I used those 1 Ohm resistors. I didn't have sufficient heatsink. I biased the output MOSFETs with slightly greater than 1A.

With 0.38V across 1 Ohm resistor it's not reaching 1A. No wonder I got that sound. In order to do the Math, I disconnected the current source feedback (in fact both the big capacitors). To my surprise, the sound got back to normal (and louder). I measured the voltage across 1 Ohm resistor, about 0.48V, almost 1A bias current. The heatsink was getting hotter than before... (Weird, why this didn't happend before??#!@$??)

I'm still replacing the 1 Ohm resistors in the other channel with 0.5 Ohm PAINTON resistors. Without those feedback capacitor, I assume that the sound characteristics must be better. I think I will temporarily leave those capacitor out while concentrating on designing my tube preamp...