Official M2 schematic

Member
Joined 2006
Paid Member
Please take a look at post 1008 or link http://www.diyaudio.com/forums/pass-labs/281520-official-m2-schematic-101.html#post4664760

This is the signal i get on speakers. The input is dead queit also when shorting it the signal remains on speakers. Only difference is when I move the toroids away from the circuit I get different modulation and volume.

Thanks

On the psu do you have your GND return to chassis with CL60 ?

The signal is certainly high frequency but aperiodic which is odd.
 
Yes I tried also to rotate it and the modulation of the signal is a little different.
Today I played whit different position of toroids and its clear that I need to move them 10cm out of the enclosure to get almost no hum but still... This toroids hum also so I put them on aluminium plate and I get almost the same sound from it as is in speakers...I tried a DC block but no change, they just hum...:) Odd experience so far! It looks like the quality of toroids is the most important thing! This were made by toroidy.pl.
 
Yes I tried also to rotate it and the modulation of the signal is a little different.
Today I played whit different position of toroids and its clear that I need to move them 10cm out of the enclosure to get almost no hum but still...Odd experience so far! It looks like the quality of toroids is the most important thing! This were made by toroidy.pl.

That's a great point! I have a Piltron toroid in my F6, it's about 25% smaller in size than the Antek. I think I'll try swapping them and squeezing it another 1/2" further forward.
 
Member
Joined 2006
Paid Member
change that 10uF for start

though , if crackling is nonexistent with input shorted ........ recheck all solder joints

most important is that you excluded input JFtes and xformers from list of suspects , rest is peanuts


Thanks everyone for the help. It's fixed. I used my meager oscilloscope skills and found the noise everywhere including at the input with no source. It had to be the Jfets. I replaced them again with a pair from a working F5 board. The bad ones test fine with my Peak analyzer so now I'll have to go through my stash of sk170/j74 and see if I have been sold fakes. They were matched for IDSS so I'm not sure...
 
Member
Joined 2002
Paid Member
Hi ZUM911,

Good to see you've found the problem.

When you said, the bad jfets tested fine, what information did the Peak analyzer
give you? Are you able to do curve traces of the jfets.

If I read your posts correctly, both the first and second jfet sets were bad. Did
they come from the same source as your good channel ones?

Dennis
 
New Power supply and Layout

I spent a day getting this sorted but It was well worth the change. Now the HUM is present at 2' vs 20'. The 300VA Piltron doesn't have a shield wire but is much quieter in this configuration. Now it's cooking time.:note::note::note:

First photo is Antek 400VA second 300VA Piltron.
 

Attachments

  • M2-1.JPG
    M2-1.JPG
    184 KB · Views: 891
  • M2-2.jpg
    M2-2.jpg
    778.5 KB · Views: 890
Member
Joined 2006
Paid Member
I spent a day getting this sorted but It was well worth the change. Now the HUM is present at 2' vs 20'. The 300VA Piltron doesn't have a shield wire but is much quieter in this configuration. Now it's cooking time.:note::note::note:

First photo is Antek 400VA second 300VA Piltron.

Congrats...you will note that is the configuration NP uses...i use it as well.

Because transformer is furthest away from input !!!
:cool:
 
Member
Joined 2006
Paid Member
Hi ZUM911,

Good to see you've found the problem.

When you said, the bad jfets tested fine, what information did the Peak analyzer
give you? Are you able to do curve traces of the jfets.

If I read your posts correctly, both the first and second jfet sets were bad. Did
they come from the same source as your good channel ones?

Dennis


The analyzer basically says its good or bad and identifies the pins. I started gathering jfets when the sk170 and j74 were discontinued and got them from a few different sources. Unfortunately they are mixed now after I tested them.

I don't have a curve tracer and only did IDSS matching.

In a few days I'll test the bad ones again and have a closer look to see if there are anything funny about them.