Question
SY, i'd like to ask fews newbie's questions :
1. Do I really two power supply? Can We just buil one power supply ? Does that will have any effect?
2. If we have to use 2 power supply, the what is the transformer's capacity for the driver? i have one small transformer (270VAC/20 or 35mA without CT). can I use that?
3. Could you explain me why you use regulated B+ for screen grid while you use simple CLC supply for Driver? Because i usually build amps with regulated at driver (even I use DC for heaters) and use standard CLC for Output?
Thanks
Ferry
SY, i'd like to ask fews newbie's questions :
1. Do I really two power supply? Can We just buil one power supply ? Does that will have any effect?
2. If we have to use 2 power supply, the what is the transformer's capacity for the driver? i have one small transformer (270VAC/20 or 35mA without CT). can I use that?
3. Could you explain me why you use regulated B+ for screen grid while you use simple CLC supply for Driver? Because i usually build amps with regulated at driver (even I use DC for heaters) and use standard CLC for Output?
Thanks
Ferry
Ok, sy!
Can I use the AN TRANS-200 (8k 15w)?
How do I adapt the circuit to this?
And not least what spec's do I need for the mains transformer?
If I go for this design, it will be my first tube amplifier from scratch!
I need a design that will give me some good sound in my rf-3's to inspire to continue diy'ing.
Can I use the AN TRANS-200 (8k 15w)?
How do I adapt the circuit to this?
And not least what spec's do I need for the mains transformer?
If I go for this design, it will be my first tube amplifier from scratch!
I need a design that will give me some good sound in my rf-3's to inspire to continue diy'ing.
Arnoldc's idea is a good and simple one. Even with a conventional chassis, it could have feet say 1 cm tall and the LED's mounted on the bottom in a simple grid on a panel. Then the chassis would "hover" on a pool of red light, like the automobile guys do to their cars. The "look of power" and maybe not so bright as to ruin the pleasure of looking at the tubes/valves.
and maybe not so bright as to ruin the pleasure of looking at the tubes/valves.
SY,
Do you have any more information about this amp?
What I would like to know is Zo, distortion pattern and more technically things than just the pattern of the red light district LESs.
I can see that you have put in some large grid stoppers that will help in case of blocking and that the forward roll off frequency is quite low (indicates bad OPTs?).
Any comments?
Do you have any more information about this amp?
What I would like to know is Zo, distortion pattern and more technically things than just the pattern of the red light district LESs.
I can see that you have put in some large grid stoppers that will help in case of blocking and that the forward roll off frequency is quite low (indicates bad OPTs?).
Any comments?
Re: Question
Using a separate supply for the driver lets you run a higher voltage there- good for linearity and stability. And the second supply only neds to deliver 10mA or so, so it can be pretty inexpensive. The transformer you have should work perfectly.
With pentode output stages, the screen supply is MUCH more critical than the plate supply. So regulation is called for. As a benefit, it gives a very convenient way to adjust idle current.
With the driver supply separated and the driver running class A, regulation doesn't bring much to the table.
Mas Penk said:
Using a separate supply for the driver lets you run a higher voltage there- good for linearity and stability. And the second supply only neds to deliver 10mA or so, so it can be pretty inexpensive. The transformer you have should work perfectly.
With pentode output stages, the screen supply is MUCH more critical than the plate supply. So regulation is called for. As a benefit, it gives a very convenient way to adjust idle current.
With the driver supply separated and the driver running class A, regulation doesn't bring much to the table.
EsE, I don't know that transformer. Is it made for push-pull? If so, it should work.
Jan, there will be much more info in the actual article. Bug Bas about publication date. Regarding measurements, they're in the article, too, but rest assured that I do NOT believe in designing without careful measurement.
BTW, good pickup on the seemingly odd value for grid stoppers!
Jan, there will be much more info in the actual article. Bug Bas about publication date.
Regarding measurements, they're in the article, too, but rest assured that I do NOT believe in designing without careful measurement. BTW, good pickup on the seemingly odd value for grid stoppers!
SY, I have one question about the Maida regulator. I read (probably Morgan Jones) that the current through the zener diode should be at least 5mA for it to operate without producing to much noise. If my math from after 2 in the morning is right, in your design there is about 1,3mA going through the zener. Is that sufficient? Did you try higher currents? What where the results?
Many thanks
Erik
Many thanks
Erik
SY, Thank you for you explanation about power supply.
But my friend (who also has plan to build this amp too) doesn't have second power transformer for the driver and his plan is put 2 bridge diodes for generating power supply. One for B+1 and Screen Grid and the other for driver.
Is that okay or it will any detrimental effect on the amp?
rgds
Ferry
But my friend (who also has plan to build this amp too) doesn't have second power transformer for the driver and his plan is put 2 bridge diodes for generating power supply. One for B+1 and Screen Grid and the other for driver.
Is that okay or it will any detrimental effect on the amp?
rgds
Ferry
Attachments
Well, remember that you're going to want 350V or a bit more for the driver, but 320V or a bit less for the output stage. That's tough with one transformer. You also sacrifice (potentially) some stability unless you use another regulator for the driver. And at that point, it's cheaper and easier to bite the bullet and buy a small 270V/20mA (or more) transformer.
It is possible to use one transformer feeding two rectifiers. One rectifier operates into a capacitor input filter and gives 350 V. The other rectifier operates into a choke input filter and can be tweaked to give 320 volts.
A very similar scenario, with one transformer and two rectifiers again. the rectifier that feeds the power amp is a tube rectifier with 30 volts loss. The rectifier that feeds the driver stage is SS. The cap values can be tweaked to get the desired voltages.
A very similar scenario, with one transformer and two rectifiers again. the rectifier that feeds the power amp is a tube rectifier with 30 volts loss. The rectifier that feeds the driver stage is SS. The cap values can be tweaked to get the desired voltages.
You don't need to regulate B+2. Just the screens.
Tubelab, that is a good approach, though separating the supplies completely means that any sag in the output tubes' B+ under hard drive will not be reflected back to the driver. I think that helps tighten up the bass and definitely banishes any tendency toward motorboating.
Tubelab, that is a good approach, though separating the supplies completely means that any sag in the output tubes' B+ under hard drive will not be reflected back to the driver. I think that helps tighten up the bass and definitely banishes any tendency toward motorboating.
Mas Penk said:
That rc network is a compensation to maintain overall stability when global feedback is applied.
If you decide to change the OPT you HAVE TO modify that network, maybe you need also to bypass the Rfdb with another rc network.
You must do that because with high mu and high rp pentode you can generate instability using global feedback.
The way to set the optimum values is to fit variable resistors and capacitors in place, apply 10kHz square wave to the input, monitor the output across a dummy load and adjust the compensation until you see a clean square wave. This method is suggested by Morgan Jones in building valve amplifier.
Mark
MJ's method is a good one, quite classic. I prefer to not use the cap in the feedback network (read Crowhurst to see why not), but rather, put a dominant pole in the open loop response sufficiently low that the feedback cap is not necessary. If the open loop gain isn't high enough, this might cause a little bit of rolloff at the top of the passband, but that's something I can live with.
In this case, using old Dynaco transformers, the amp has no problem hitting 30kHz at better than -0.1dB. Newer and better transformers could only help.
In this case, using old Dynaco transformers, the amp has no problem hitting 30kHz at better than -0.1dB. Newer and better transformers could only help.
poobah said:A.K.A. "feedforward compensation"... cool.
Yeah, and the forward roll off frequency often tells something about the quality of the OPTs. Leakage inductance can be a pain in some cases...
The problem with forward roll off is distortion at high frequency, there will not be so much to feedback (compensate) when putting in a low pass filter. If fc is low and the global feedback is moderate (~20dB), this indicates not much (or anything at all) for compensation at high frequencies.
