Could be that your power supply is on the edge of oscillation. Please describe how much capacitance, the current draw, number of stages, what rectifier etc.
If you are using a centre tapped transformer, you can just use the relay to close the centre tap to ground, this works quite well. If you using a diode bridge, you should switch the circuit before the rectifier, on one (or both) of the power transformer leads.
If you are using a centre tapped transformer, you can just use the relay to close the centre tap to ground, this works quite well. If you using a diode bridge, you should switch the circuit before the rectifier, on one (or both) of the power transformer leads.
Hi there......We need MORE circuit details......e.g are you using NFB from o/p tranny sec to input tube ? ......it may not be phase shift stable with large voltage v.s gain conditions.
P-P circuits are more susp in this....Are you delaying B+ to o/p tube but not to input tube ?
Is circuit a home-made botch up ?
rich
P-P circuits are more susp in this....Are you delaying B+ to o/p tube but not to input tube ?
Is circuit a home-made botch up ?
rich
Abruptly switching DC into inductors...
Well of course you get a nasty thump when switch on. You are attempting to suddenly magnetise the core of your output transformer, and Lenz's law objects. Even worse, by switching off suddenly, you turn your output transformer into an ignition coil, creating an even worse thump.
Move the switch so that it switches the AC to the HT rectifier. That way, the HT will come up rather slower, and certainly decay slower. If I have interpreted your post correctly, this should substantially reduce the thumps.
Well of course you get a nasty thump when switch on. You are attempting to suddenly magnetise the core of your output transformer, and Lenz's law objects. Even worse, by switching off suddenly, you turn your output transformer into an ignition coil, creating an even worse thump.
Move the switch so that it switches the AC to the HT rectifier. That way, the HT will come up rather slower, and certainly decay slower. If I have interpreted your post correctly, this should substantially reduce the thumps.
Thanks for the replies
The power supply is like.
Solid State bridge, 4400mF, choke, 2200mF, 50mF MKP, relay, OPT. There is a 10k resistor in parallel of the relay to charge the OPT.
@SY: Yes, it’s fully charged.
@richwalters: Yes, the relay only switches on the HV of the output tube. The previous 2 stages get HV with no delay switch.
@EC8010: I think your explanation is what happens. But if I do move it on the AC point charging the caps would not give me a worst one?
The delay is there because I have some info that 6C33C-B has some kind of problem getting HV before heating up.
If you have some more info on the reliability of the tube that can take out the relay I would be more than happy. They cost just 10$.
The power supply is like.
Solid State bridge, 4400mF, choke, 2200mF, 50mF MKP, relay, OPT. There is a 10k resistor in parallel of the relay to charge the OPT.
@SY: Yes, it’s fully charged.
@richwalters: Yes, the relay only switches on the HV of the output tube. The previous 2 stages get HV with no delay switch.
@EC8010: I think your explanation is what happens. But if I do move it on the AC point charging the caps would not give me a worst one?
The delay is there because I have some info that 6C33C-B has some kind of problem getting HV before heating up.
If you have some more info on the reliability of the tube that can take out the relay I would be more than happy. They cost just 10$.
OK, that's a little clearer. You'd do best to ramp up the voltages a bit more slowly. Maybe a lot more slowly.
One neat way to warm up without severe thumps is to put the timing relay on the primary side of the B+ supply; the contacts are put across a 100 ohm power resistor (10 watts or so) in series with the primary. After 10-15 seconds of filament warmup (that's all you need to prevent cathode stripping, not a full warmup), the relay closes and shorts the resistor. (this all assumes that the filament transformer is separate from the B+ transformer)
One neat way to warm up without severe thumps is to put the timing relay on the primary side of the B+ supply; the contacts are put across a 100 ohm power resistor (10 watts or so) in series with the primary. After 10-15 seconds of filament warmup (that's all you need to prevent cathode stripping, not a full warmup), the relay closes and shorts the resistor. (this all assumes that the filament transformer is separate from the B+ transformer)
Hi EC8010, Sy;
I like the ideas I've seen so far for cutting back on the turnon thump. Here are a couple of other suggestions for helping Morfeas improve his amp (not just turnon thump); what does everyone think of them?
1. Put the smaller cap first in that CLC bridge - maybe even add a smaller first leg to his CLC filter just to take it easy on his PS trafo. Charging those enormous HV caps has got to be stressing his trafo!
2. Put an NTC (Negative Temperature Coefficient) thermistor inline with the B+ (in front of the caps) to help limit the current flow.
3. Drop the size of the filter caps and rely more on inductance to provide filtering.
4. Use a hybrid or indirectly heated valve rectifier instead of ss - obviously this won't work with the enormous current draw of those caps at startup. A 5AR4 has a textbook max lead cap size of 40uF IIRC, for example - I'd keep it under 30uF personally.
Just ideas. Good luck Morfeas!
All the best,
Morse
I like the ideas I've seen so far for cutting back on the turnon thump. Here are a couple of other suggestions for helping Morfeas improve his amp (not just turnon thump); what does everyone think of them?
1. Put the smaller cap first in that CLC bridge - maybe even add a smaller first leg to his CLC filter just to take it easy on his PS trafo. Charging those enormous HV caps has got to be stressing his trafo!
2. Put an NTC (Negative Temperature Coefficient) thermistor inline with the B+ (in front of the caps) to help limit the current flow.
3. Drop the size of the filter caps and rely more on inductance to provide filtering.
4. Use a hybrid or indirectly heated valve rectifier instead of ss - obviously this won't work with the enormous current draw of those caps at startup. A 5AR4 has a textbook max lead cap size of 40uF IIRC, for example - I'd keep it under 30uF personally.
Just ideas. Good luck Morfeas!
All the best,
Morse
@EC8010: hope so. I didn’t get that big bang at the lab with 89db speaker. But my main speakers are 100db.
@Sch3mat1c: ìF (microfarad)
@SY: unfortunately the PT is one. So I’m going to try it on the secondary AC HV line.
@Morse: there is a thermistor on the primary of the PT. PT is 630VA. No space available on the chassis for an extra tube.
Thanks to all for the replies.
@Sch3mat1c: ìF (microfarad)
@SY: unfortunately the PT is one. So I’m going to try it on the secondary AC HV line.
@Morse: there is a thermistor on the primary of the PT. PT is 630VA. No space available on the chassis for an extra tube.
Thanks to all for the replies.
It's best to use the current limiter on the primary side. They don't tend to like high voltages. If it causes the filaments to warm a bit more slowly, that's probably OK (that's the way Eico did it in some of their classic tube amps with SS rectification). The main thing is to not do the sort of hot switching you've got set up now.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.