|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
24.169.151.61
In Reply to: shared versus separate filament suppies.... posted by Vinnie on January 20, 2007 at 06:05:22:
With directly heated tubes, the heater circuit is also the cathode. With adjustable grid bias, you can share the heater winding, but don't do it otherwise.
Follow Ups:
But why? I would like to know what problems are created by not having the separate supply.
thanks
If your tubes are matched exactly, no problem. But they could drift apart, one could need replacement, who knows. If they're cathode biased, they won't draw the same current, and the hot one may "run away" - goodbye!. You need to separate heaters (separate cathode resistors) or have adjustable bias for each.
I have separate catode resistors, but a common supply for the 300B filaments (per my first post). The other thing I did was use twin hum pots with the single supply. Does fit your description of what works?
The cathode resistors are in parallel, since the cathodes are in parallel on the same heater winding - they are not separate. If one tube opens up, the other will self-bias to near twice normal current - not good. It's OK as long as they are matched...
Roger on that, I think I get the idea now. I will install another filament tranny and report back on the results.
Now in a push pull Class A amplifier with a filament transformer and a hum pot for each output tube, assuming the tubes are matched in terms of gain and idle current for a given bias, the wipers can be tie to a common cathode resistor. This has the advantage of eliminating the need for a cathode resistor bypass cap while still being able to null out the hum. One less cap in the signal path is always a good thing.With single ended output, the cathode resistor bypass cap is unavoidable and 100% in the signal path.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Ok then, I will fix it right. Hope it helps with the residual hum too. Another question you made me think of is how does one trace out the "signal path"?
Thanks for the help.
Some talk of "primary" and "secondary" (order of importance) audio path.
I won't argue with them but any part where the voltage or current is being modulated by the audio is in the signal path.Example;
"Grounded cathode gain stage"
Parts list;
1. Grid resistor
2. tube
3. cathode resistor
4. cathode resistor bypass cap
5. plate resistor
6. decoupling capacitor (last cap in power supply just before the plate resistor)
7. coupling cap (the cap the feeds the output signal to the next stage)
8. next stage grid resistor.1. This one's obvious. This is as far as the original signal gets. All other signal is just a response to this signal.
2. It's the changes within the tube, caused by the changing grid bias (input signal) that creates the output signal.
3. The current for the tube flows through the cathode resistor.
4. With the cathode resistor bypassed, only the DC current flows through the cathode resistor, the signal (AC component) of the tube's current flow flows through this cap.
5. All the current (AC and DC) flows through this resistor.
6. This cap is the "path" for the AC component to get back to ground without having to go all the way back through the power supply.
7. All of the usable output signal to the next stage flows through this cap.
8. Back to the grid resistor where it all starts for the next stage.
So all of these parts are "in the signal path".
With DHT's the filament and the cathode are one in the same.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Thanks Tre', this will be very useful. Since your last post I installed a second filament transformer for the 300b's. It may be safer now, but the hum is much worse. I am guessing it's a grounding or layout issue, so that is what I will start on next. Know of any good web sites on how to do star grounding layouts? Also, do all the filament leads to the tubes from the filament trannys have to be separated? As it is now I have bundled them from the PSU chassis to the front end chassis and then spread them out. Wondering if that could be a source of hum?
Thanks!
Vince
I would keep the filament wires separated and each pair should be twisted. Keep them as far away from other parts as you can, close to the chassis. Did you adjust the hum pots for minimum hum?The main thing about grounding is you only want one spot where the chassis is grounded. Run wires to everything that needs to be grounded from one chassis point. This includes the power supply.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
One last question..... how do you treat the star grounding for dual chassis? Would I have a star or bus bar on the front end (grounded at one spot on the front end chassis)with one wire that goes to the star on the PSU chassis and then take one wire to the mains ground?
Thanks
Vince
Each chassis has it's own star ground. The only ground between chassis is the IC (or in the case of separate power supply chassis, there is a ground wire that feeds to the star ground in the receiving chassis). The only chassis that is earth grounded in my system is the preamp.BTW, I keep the filament (even when they are DC) in a separate wire from the B+ when running between PS chassis and signal chassis. I like to shield the B+ line. Filament supplies can be very noisy and you don't want that getting into your B+.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Doesn't the psu chassis have to be earth grounded for safety sake? By the way I worked the psu layout over for several hours last night per your suggestions and the hum is way down. Might try shielding the B+ and see if I can get rid of a bit more. Thanks for all the help!
Vince
Safety? What's that? :-)I look at it this way, the supply is fused. If something goes wrong the fuse will blow. I've been at this for 30+ years and I'm not dead yet!!!
In the studio we use ground lifts all the time to get things as quiet as possible, haven't killed a client yet!
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
If the power amp is earth grounded but the preamp is not earth grounded, could this cause a ground loop or other type of ground problem?
In my system I can only ground one thing or I get ground loop hum. I don't get a ground loop hum with the power amp grounded and the preamp ungrounded but the system is quieter with the preamp grounded and the power amp ungrounded.Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
(nt)
Another way around that is to use an isolation transformer on the preamp. That's what I have done and it works well.
Vince
No, you don't have separate cathode resistors. With only one filament winding for both 300b's the two cathode resistors are in parallel one way or another.If you have a hum pot for each tube, with the wiper of each hum pot going to it's own cathode resistor, those hum pots are in parallel and those cathode resistors are in parallel because it's all tied together at the filament transformer winding.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Well that cuts it then. Better throw in another filament tranny and see what happens.
This post is made possible by the generous support of people like you and our sponsors: