With 220uF as C7, Thd is much higher in the sim, so by replacing it, this amp is loosing all its quality... There must be a better way..
But I'll try..
But I'll try..
With C7=200pF, it's oscillating, but voltages on drivers' bases are more or less correct, and negative rail is stuck to -9V.
With C7=100uF (also tried 470uF), it's not oscillating, but voltages on drivers' bases are wrong (the upper one at -38V), and -38V at the output.
With C7=100uF (also tried 470uF), it's not oscillating, but voltages on drivers' bases are wrong (the upper one at -38V), and -38V at the output.
I tried C7=300pF and 500pF. Same behavior.
C7=1nF and up, oscillates heavily in the sim.
But will try 1nF.
C7=1nF and up, oscillates heavily in the sim.
But will try 1nF.
Tried C7=1n5. Same behavior. Oscillations.
By the way - I checked the correctness of the PCB, and values of all resistors and capacitors. Everything seems to be correct.
So far I was playing with R20 (3 Ohm - 10 Ohm),
and with C7 (200p, 300p, 500p, 100u, 470u).
D3/D4 were removed.
By the way - I checked the correctness of the PCB, and values of all resistors and capacitors. Everything seems to be correct.
So far I was playing with R20 (3 Ohm - 10 Ohm),
and with C7 (200p, 300p, 500p, 100u, 470u).
D3/D4 were removed.
I'm testing with TLE 2081. Better TL 071.
Will try different op-amps, but usually - if it works with one op-amp, but not the other, this is just a sign of general instability or at least sensitivity...
From my experience if something oscillates with TL 071, it will get worse with faster op-amps...
Very large loop gain 140db20khz , and this affects the design of the PCB .
Your file in lts spice give . I'll take a look
Your file in lts spice give . I'll take a look
Last edited:
>very large loop gain , and this affects the design of the PCB
That's true, but it's the same true for all these amps from this thread: 4 of them worked great, with similar pcb.
Also amps with op-amps built previously, including composite amps with 2 op-amps, worked ok with my PCBs..
Tried TL071 - same results as TLE2081.
Tried LT356 - slightly different - oscillates for 0.5s, and then stops, negative rail DOES NOT stick to -9V, but stays at -45V, and the output is at -38V.
So similar behavior to the case when C7=100uF.
From all the amps I built with op-amps (7-8 working amps), changing op-amp was never a solution.. so far 🙂
That's true, but it's the same true for all these amps from this thread: 4 of them worked great, with similar pcb.
Also amps with op-amps built previously, including composite amps with 2 op-amps, worked ok with my PCBs..
Tried TL071 - same results as TLE2081.
Tried LT356 - slightly different - oscillates for 0.5s, and then stops, negative rail DOES NOT stick to -9V, but stays at -45V, and the output is at -38V.
So similar behavior to the case when C7=100uF.
From all the amps I built with op-amps (7-8 working amps), changing op-amp was never a solution.. so far 🙂
Will continue tomorrow. Making 2n3055 version of the previous LMK amp was also a torture, but after 2 weeks of thinking,simming and re-trying, it eventually worked like a champ..
The truth is, that this is 'sensitive' amp - really small changes in few caps or resistors here and there, makes it unstable..
The truth is, that this is 'sensitive' amp - really small changes in few caps or resistors here and there, makes it unstable..
>oscillates for 0.5s, and then stops, negative rail DOES NOT stick to -9V, but stays at -45V, and the output is at -38V.
This is equivalent of shorting R14 (50 Ohm in the emiiter or Q6) to the negative rail.
Upper driver Base voltage at -35V, Lower driver's Base at -44V, Output at -38V
That's exactly was what I saw.
Lower Q6 is fully conducting, and upper Q5 not conducting at all.
All the voltages fit perfectly..
This is equivalent of shorting R14 (50 Ohm in the emiiter or Q6) to the negative rail.
Upper driver Base voltage at -35V, Lower driver's Base at -44V, Output at -38V
That's exactly was what I saw.
Lower Q6 is fully conducting, and upper Q5 not conducting at all.
All the voltages fit perfectly..
Last edited:
In the sim, I also see that exactly this situation will happen, if Q1 is shorted (C-->E).
Upper driver Base voltage at -35V, Lower driver's Base at -44V, Output at -38V
Tomorrow will replace it, maybe Q1 is bad...
In the sim, 50/100 Ohm seems to be perfect.
Q1 is my suspect now 🙂
Try in your sim - short E and C of Q1.
Upper driver Base voltage at -35V, Lower driver's Base at -44V, Output at -38V
Tomorrow will replace it, maybe Q1 is bad...
In the sim, 50/100 Ohm seems to be perfect.
Q1 is my suspect now 🙂
Try in your sim - short E and C of Q1.
Bc546B has max Vce = 65V, according to the sheet.
In this amp, his Vce = 35V, that should be good enough..
But maybe some oscillations killed it...
Maybe will remove Zobel, so it doesn't burn, and will try to measure all voltages.
With Zobel, it was difficult, because I didn't want to keep amp running for longer than 5s.
In this amp, his Vce = 35V, that should be good enough..
But maybe some oscillations killed it...
Maybe will remove Zobel, so it doesn't burn, and will try to measure all voltages.
With Zobel, it was difficult, because I didn't want to keep amp running for longer than 5s.