Help with Tortello's headphone amp.

[DRC]

Frequency of oscillation

Kram,
What was the frequency of oscillation (or the sweep speed for the scope traces so I can work it out) ?

If it's low (Hz) then it's something called 'motor-boating' (that's what it would sound like !). This is due to an interaction between the PSU and amp.

If it is much higher (MHz) then due to the O/P devices.

Dave

 

[DRC]

Correct value for R18 ?

Kram,
Check the value of R18. If this is too high it could explain the big supply variation on Q1D.

Dave

 

[Kram]

hello again.

R18 is 5.8R according to my meter (it is supposed to be 5.6R though), voltage directly accross r18 is 1.07V. Voltage from R18 to gnd is 24.4V.

V accross R10 is 0.433V

V accross R12 is 0.476V

Q1D and Q1S were wrongly labled, they should be swaped around.

scope pic labled Q1S (really Q1D) is 714KHz (each horizontal square is 200ns amd there are 7 squares between the 2 biggest peaks. 1/T = frequency)

scope pic labled Q1D (really Q1S) is 666KHz and the amplitude is roughly 13V p-p.

I am about to try that 0.001uF cap you mentioned. I will let you know if it has any effect.

Thanks
Kram.

 

[Kram]

Hello

I tried BDPs suggestion of putting a 0.001uF cap accross Q3C and B but it hasnt really worked. It did stop the oscillations as far as i could tell and there was no dc on the output any more.

Here are some readings i took
Q1G =10.28V
Q1D = 23.04V
Q1S =5.64V

Q2G = 10.14V
Q2D = 5.17V
Q2S = 0.457V

Q3C = 10.14V
Q3B = 5.81V
Q3E = 5.16V

Something is obviously still wrong.

Oh yeah, Q1 gets really hot now while Q2 is barely warm.

Help
Kram

 

[DRC]

Maybe I see it ?

Kram,
[1] I think Q1, Q3 etc are operating correctly as a constant
current source (ignoring load sensing feedback). Because :-

The voltage accross R10 is :-
Q1S - Q2D = 5.64 - 5.17 = 0.47V -> 213mA through R10

near all of this must be passing through Q1 as R3 only has :-
Q3B - Q1S = 5.81 - 5.64 = 0.17V - > 0.17mA through R3

also Q3 is regulating this current as it is held in conduction by :-
Q3B - Q3E = 5.81 - 5.16 = 0.65V
(ie correct forward bias)

roughly, the current through R1 / R2 is :-
Q1D - Q3C = 23.04 - 10.14 = 12.90V / 2K -> 6.45mA

caveat:
The puzzle is R5, as the voltage across R5 :-
Q3C - Q1G = 10.28 - 10.14 = 0.14V -> 0.63mA ???????
I don't think it should be this high, but I really don't know for sure !

[2] The circuit around Q2 is not right ! Because :-

R19 / R14 (should) form a potential devider, so that :-
V@Q2G = V@Q2D * ( R14 / (R9 + R14) )
V@Q2G = V@Q2D * "less than 1"
except that it can't be because V@Q2G > V@Q2D
(I thought you had transposed the measurments on Q2
but these as consistant with other measurments)

If you can figure out why Q2G is at a higher voltage than
Q2D then problem solved. Junction of R14 / R7 / R9 / R6 / Q2G shorted ?
(maybe to Q3C ??)

DID YOU BREAK THE TRACK BETWEEN TOP R8 AND TOP R6

Dave

 

[Kram]

DID YOU BREAK THE TRACK BETWEEN TOP R8 AND TOP R6

Bugger

I feel like a right plonker. I even built this circuit twice and made the same mistake both times.

I now have a working amp, my first one

Thanks for your help DRC and everyone else too.

Kram.

 

[DRC]

I now have a working amp, my first one

Well done !