Audiosan and Dazed2:
Yowch -- you're right! I'm back in BigE's camp (d'oh!). Thanks for smacking me upside the head.
Regards,
Scott
Yowch -- you're right! I'm back in BigE's camp (d'oh!). Thanks for smacking me upside the head.
Regards,
Scott
OK, so the parts are all back in place.
Since R6 was not burnt, the FE board was not affected.
BTW: Can I use 1/4 watt resistors for the bias and gate?
Since R6 was not burnt, the FE board was not affected.
BTW: Can I use 1/4 watt resistors for the bias and gate?
RE - 1/4 Bias resistor:
If there is 0.25 V across 0.5 ohms, that's 0.5 amps. P = I^2 * R => 0.5^2 * 0.5 = 0.25*0.5 = 0.125W.
The resistor would be operating at 50% capacity at quite low bias. I would recommend 0.5 watt resistors here, but the board spacing appears to be for 1/4 watt.
If there is 0.25 V across 0.5 ohms, that's 0.5 amps. P = I^2 * R => 0.5^2 * 0.5 = 0.25*0.5 = 0.125W.
The resistor would be operating at 50% capacity at quite low bias. I would recommend 0.5 watt resistors here, but the board spacing appears to be for 1/4 watt.
bias resistor?
do you mean source resistor?
The resistor that you can calculate the bias current through it by measuring the voltage drop and applying Ohm's Law?
That Source resistor passes all the output current.
That output current could approach 10Apk
Pmax = 10^2 * 0r5 /2 = 25W rms if a continuous sinewave were passing with a peak value of 10A.
But it is 50% duty cycle leaving an average maximum dissipation of 12.5W but an instantaneous peak of 50W.
What is the maximum rating of your Source resistor?
do you mean source resistor?
The resistor that you can calculate the bias current through it by measuring the voltage drop and applying Ohm's Law?
That Source resistor passes all the output current.
That output current could approach 10Apk
Pmax = 10^2 * 0r5 /2 = 25W rms if a continuous sinewave were passing with a peak value of 10A.
But it is 50% duty cycle leaving an average maximum dissipation of 12.5W but an instantaneous peak of 50W.
What is the maximum rating of your Source resistor?
Sorry, no. My post is all wrong. I was referring to the 2.2K ohm resistor in series with the thermistor , and then botched up the numbers badly.
So, it appears that something might be damaged on the Pside as well. When I try to adjust P1, the value across the test points on the FE board will go no higher than about 23.6 ohms in the Rchan -- the one that blew. ( Now wondering if the Pchan could've been damaged from the start with improper handling ( static ). It would explain the much larger distortion numbers..... ).
P1 was changed to no effect. P3 was changed as well, and both channels agree with measurements taken across the terminals of P3.
Ohmeter measurements of the P-channel transistors agree in their readings, except for one which is way off. It is possible that there is damage here, as P3 was turned first one way then the other. If this suspicious transistor were to be removed, then it is possible that the test points will read properly.
No power will be applied until that time. All aux circuits were tested with main transformer removed. They are ok.
P1 was changed to no effect. P3 was changed as well, and both channels agree with measurements taken across the terminals of P3.
Ohmeter measurements of the P-channel transistors agree in their readings, except for one which is way off. It is possible that there is damage here, as P3 was turned first one way then the other. If this suspicious transistor were to be removed, then it is possible that the test points will read properly.
No power will be applied until that time. All aux circuits were tested with main transformer removed. They are ok.
The pChannel transistors ohm out funny. Two have infinite resistance from leg to leg, and the other two have odd resistances, all quite low.
The quad will be changed, and the two infinite ones will be tested further. I suppose the resistors on the OP boards also ought to be changed.
The quad will be changed, and the two infinite ones will be tested further. I suppose the resistors on the OP boards also ought to be changed.
The pChannel transistors ohm out funny. Two have infinite resistance from leg to leg, and the other two have odd resistances, all quite low.
The quad will be changed, and the two infinite ones will be tested further. I suppose the resistors on the OP boards also ought to be changed.
The quad will be changed, and the two infinite ones will be tested further. I suppose the resistors on the OP boards also ought to be changed.
Symptom: P1 not moving to full scale ( approx 800 ohms ) in right channel as measured across test points T3 T4.
Fix: Replaced all Pchannel transistors in the right channel. P1 operates properly.
Fix: Replaced all Pchannel transistors in the right channel. P1 operates properly.
Turn the bias setting resistors right down to zero ohms. Check with an ohm-meter.
Start up the amp via a mains bulb tester.
Measure the current through the jFETs at the source resistors.
Measure the Vds of each jFET
Start up the amp via a mains bulb tester.
Measure the current through the jFETs at the source resistors.
Measure the Vds of each jFET
It lives! But not before burning a ground trace off an output board with some microscopic swarf pressing it's way between a diode and the heatsink. Ohm it out, no sweat. Give it an extra little twist to hold it down a touch better and it shorts...... So, I jumpered the ground trace and cleaned the heatsinks with nail polish remover.
Now being biased.
OMG are the bourns trim pots awful.
Now being biased.
OMG are the bourns trim pots awful.
It lives! But not before burning a ground trace off an output board with some microscopic swarf pressing it's way between a diode and the heatsink. Ohm it out, no sweat. Give it an extra little twist to hold it down a touch better and it shorts...... So, I jumpered the ground trace and cleaned the heatsinks with nail polish remover.
Now being biased.
OMG are the bourns trim pots awful. The technical term is "Sticktion".
Now being biased.
OMG are the bourns trim pots awful. The technical term is "Sticktion".
Last edited:
In my F5TV3 I want to incorporate chokes in the PSU, eigther a Hammond ZJ159 in a CLC configuration, or a 159ZC in a LC configuration. Which one to go for and why?
With a choke in the PSU I hve been warned not to use a soft start for the transformer, but only NTC on the primair. Comments on that to, please.
Eivind Stillingen
With a choke in the PSU I hve been warned not to use a soft start for the transformer, but only NTC on the primair. Comments on that to, please.
Eivind Stillingen
I can't see the reason for that warning.
Particularly since an NTC in the Primary route is a soft start that is recognised as among the best ways of soft starting a transformer.
AndrewT - I think he was advised to not use a relay/resistor style soft start, and instead use an NTC in the primary.
Liliya - use CLC. Remember that your capacitors must be rated at least 50% higher than the steady-state rail voltage due to the choke acting as a boost converter on startup. Also, cascode the input, so that voltage doesn't fry the Jfets.
Liliya - use CLC. Remember that your capacitors must be rated at least 50% higher than the steady-state rail voltage due to the choke acting as a boost converter on startup. Also, cascode the input, so that voltage doesn't fry the Jfets.
It is a relay soft start but with NTC instead of resistors. About the capasitors rating 50% higher than the steady state rail voltage was a new and very important imformation. Rail voltage is 25V AC (34V DC). Capacitors I use are 33000uF50V. Is this safe? Input is cacoded.
Eivind Stillingen
Eivind Stillingen
The amp works, and is DEAD quiet. The THD measures as 0.055% L channel, and 0.035% Rchannel. Biased at 250 mV.
Something is wrong with the SNR measurement, as it shows just 28 dB, and THD+ noise around 2-6%. I suspect that using a cheap bnc to rca adapter is a problem, and will get a bnc cable with clips on the DUT side.
THANKS to everyone for helping and putting up with my drama.
Special thanks to dazed2 who came over last night and watched meters over my shoulder while I worked. I swear I would still be cursing today.
And thanks to AndrewT. It might be a good idea to install rail fuses....
The sonic signature is FAR softer sounding that the amps that dazed2 has produced. It is not even as aggressive as dazed2's F5!
Is this what P3 controls or can it be the inclusion of 330 uF Nichicon bipolar caps on the output boards?
Something is wrong with the SNR measurement, as it shows just 28 dB, and THD+ noise around 2-6%. I suspect that using a cheap bnc to rca adapter is a problem, and will get a bnc cable with clips on the DUT side.
THANKS to everyone for helping and putting up with my drama.
Special thanks to dazed2 who came over last night and watched meters over my shoulder while I worked. I swear I would still be cursing today.
And thanks to AndrewT. It might be a good idea to install rail fuses....
The sonic signature is FAR softer sounding that the amps that dazed2 has produced. It is not even as aggressive as dazed2's F5!
Is this what P3 controls or can it be the inclusion of 330 uF Nichicon bipolar caps on the output boards?