You can zoom by picking a spot sufficiently far along that the supply has settled (say, 15 seconds out), then zoom on it by dragging the cursor to "box" a small section. Or you can start the sim at (say) 15 seconds and run for one or two seconds. That will fill the window pretty well, and you can look at the min and max voltages to get the ripple.
If you start at zero, the rise from 0 to 300V (or whatever) will swamp the ripple.
If you start at zero, the rise from 0 to 300V (or whatever) will swamp the ripple.
Fired it up yesterday, after only two puffs of smoke it actually worked! Kind of...had to take the 285V and 320V rails down a bit as the Hammond power transformer ended up a little underrated.
The puff of smoke was one of the 47R resistors on the UL screen connection to one of the output valves. A small light show accompanied the burning resistor each time (pretty sure that's not good). I swapped valves and everything seems fine now in that area.
I'm noticing some disparate measurements between the output stages in each channel. I should preface this by saying I don't have the chassis work done and so do not yet have a signal input or negative feedback hooked up. The cathodes of the output valves in the left channel are running at 7V and 9V while the right channel is at 14.5V and 24.5V.
On the left channel (7V and 9V output valve cathode voltages) the input stage anode reads 65V and the phase splitter anode reads 215V. Both of these readings are higher than normal.
What else should I be looking at next? Are these readings indicative of a misplaced part? Should i wait to take these measurements until I've wired the secondaries of the OPT and have negative feedback hooked up?
Thanks for the help!
The puff of smoke was one of the 47R resistors on the UL screen connection to one of the output valves. A small light show accompanied the burning resistor each time (pretty sure that's not good). I swapped valves and everything seems fine now in that area.
I'm noticing some disparate measurements between the output stages in each channel. I should preface this by saying I don't have the chassis work done and so do not yet have a signal input or negative feedback hooked up. The cathodes of the output valves in the left channel are running at 7V and 9V while the right channel is at 14.5V and 24.5V.
On the left channel (7V and 9V output valve cathode voltages) the input stage anode reads 65V and the phase splitter anode reads 215V. Both of these readings are higher than normal.
What else should I be looking at next? Are these readings indicative of a misplaced part? Should i wait to take these measurements until I've wired the secondaries of the OPT and have negative feedback hooked up?
Thanks for the help!
Hmmm. The only output valve I'd be happy about is the one with 14.5V on its cathode. With the valves out (and having waited for everything to discharge) measure the resistance from each output cathode to ground to check that it really is 270 Ohm. Who made the EL84?
The 65V and 215V sound about right.
The 65V and 215V sound about right.
I disagree with that. The OLG without the global feedback is whatever the gain is when the loop is boken. The fact that the unbypassed resistor is providing additional current feedback is irrelevant. Otherwise we'd have to neglect the inherent anode feedback too, and any other inherent feedback in the ciruit, and start from first principles assuming the triode operates like a pentode!SY said:
)
Alfetta, no, a close friend is an expert designer/artist type and she did the work. I'll pass on the kind compliments!
Merlin, here's the way I think about it: with the loop closed, the voltage applied to the cathode is the sum of the divided-down output voltage and the current feedback. But the current feedback is reduced by the voltage feedback factor, since the input tube will now swing a proportionately smaller signal current.
In the case of the Bevois, there's about 10V of swing at the plate for full power out, which corresponds to a signal current of 10/R, where R is the plate load resistor. I don't have "Valve Amplifiers" to hand at the moment, but if memory serves, R was something like 33k. So the signal current is about 0.03mA. I can't recall the cathode resistor value, but it was something like 680R, giving a voltage swing of about 20mV. Now, the divider from the output places something like 1-2V at that point. So the effect of the current feedback with the loop closed is fairly negligible, not because of shunting, but because it's swamped by the applied voltage from the voltage feedback divider.
cgrums, I think this is a fine place for that discussion.
Merlin, here's the way I think about it: with the loop closed, the voltage applied to the cathode is the sum of the divided-down output voltage and the current feedback. But the current feedback is reduced by the voltage feedback factor, since the input tube will now swing a proportionately smaller signal current.
In the case of the Bevois, there's about 10V of swing at the plate for full power out, which corresponds to a signal current of 10/R, where R is the plate load resistor. I don't have "Valve Amplifiers" to hand at the moment, but if memory serves, R was something like 33k. So the signal current is about 0.03mA. I can't recall the cathode resistor value, but it was something like 680R, giving a voltage swing of about 20mV. Now, the divider from the output places something like 1-2V at that point. So the effect of the current feedback with the loop closed is fairly negligible, not because of shunting, but because it's swamped by the applied voltage from the voltage feedback divider.
cgrums, I think this is a fine place for that discussion.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.