Zen-like headphone amp trouble

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
P

Peppe

Member
Joined 2006
Paid Member
Hallo,

I am working on my second DIY-project and exactly on 'tortello' (Marcello) project:

http://www.diyaudio.com/forums/showthread.php?postid=19396

http://www.headwize.com/projects/showfile.php?file=pellerano_prj.htm

I used the second version of 'digi01' (Zang) boards:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=29784&perpage=10&pagenumber=1

The first time the amp worked perfect; bias set at 175 mA. After a couple of hours the bias of one channel dropped to 30 mA. The second day I monitored both channels and the bias of the broken channel started perfectly and slowly dropped to the minimum.

I checked the transistor and both the mosfets and they seem to work, also the trimmer that set the bias works ok.

In the weekend I hope to have time to fix the problem, can somebody have idea on what I have to looking for?

Thanks in advance for the answers and thanks to Marcello and Mr. Nelson to give me the possibility to have a fun with this project.

Ciao
Peppe
 
P

Peppe

Member
Joined 2006
Paid Member
Update:

I sostituited both Q3: no changing!

I inverted Q1 and Q2 from one to the other channel and now both channels have the same problem: when the temp rise the bias current goes down! :bawling:

Never saw the temp raise more than 40° C.

What's strange is I made two amps: the first one works incredible well! The only difference is the imput caps, on the second one are electrolit instead non polar MKT.

Nobady have an idea to help a newebe? :)
 

Attachments

  • img_2870.jpg
    img_2870.jpg
    92.9 KB · Views: 700
tortello

tortello

Member
Joined 2001
Hi Peppe,
check the DC voltage values as pubblished in the Headwize page: you should find similar values.
As soon as the amp goes warm and bad, re-check them all and probably you will find the problem.
Even a passive component (R2 for example...) can drift from the nominal value, and so to change the bias point of the circuit.
These problems are quite easy to discover.
Feel free to contact me, the email address is in the HW page.
Good luck!
:smash:

Marcello
 
tortello

tortello

Member
Joined 2001
Yes,
it will work correctly, but you should re-calculate the polarization network to optimize performances.
Unfortunately I can't do without YOUR transformer...
Or you might decide to sacrifice the excess of voltage increasing the PI filter resistor (R18): you will tame down the noise, in this way, and will be able to set the correct bias point for the circuit.
Best regards

Marcello
 
luvdunhill

luvdunhill

Member
Joined 2006
hm, I'm not sure how to go about the recalculation. There really isn't much other than the output specs 2 X 22VAC @ 2.73A and the regulation, which is 11%. What would I need to measure? Also, it seems that I would have to increase R18, but I will calculate it and see what I come up with. Perhaps this is easiest.

Thanks!
 
tortello

tortello

Member
Joined 2001
Let's build the circuit(s) and set R18 to have the same drain voltage for Q1 as the "original" one.
R18 will be higher in value.
Of course the time costants of the power supply will be different...
It would be better to re-bias the whole circuit, but this is a less easy way.

Marcello
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom