Hi tubelab,
My only comment was that the failure mode is normally flashing over between spirals if your applied voltage does not create excessive dissipation right away. I wouldn't be surprised if a normal overheat also causes flash over as the element goes open. As you've discovered, lot's of fun!
In short, in most sane applications I find that metal oxide resistors are more than satisfactory. They are not as quiet as metal film types, but they are much less noisy than carbon composition types are. Those ceramic compositions look interesting too.
-Chris
Well, yes. That too. However, I'll bet that the dielectric strength of the coating to external objects is a fair bit higher than the rating. Add to that, it's common practice to do as you have, mount those resistors off the PCB. Mounting them over holes is even better for dissipation.
My only comment was that the failure mode is normally flashing over between spirals if your applied voltage does not create excessive dissipation right away. I wouldn't be surprised if a normal overheat also causes flash over as the element goes open. As you've discovered, lot's of fun!
In short, in most sane applications I find that metal oxide resistors are more than satisfactory. They are not as quiet as metal film types, but they are much less noisy than carbon composition types are. Those ceramic compositions look interesting too.
-Chris
tubelab.com said:
From my short experiments, beam tubes appear to draw about one third the screen current of an otherwise equivalent pentode (comparable plate voltage and total idle current). This is to be expected - beam tubes are supposed to have the screen grid aligned in the shadow of the control grid, which substantially reduces the number of electrons that make it to the screen. I understand there may be some pentodes out there that do have aligned screens, but I haven't seen any yet myself.
Johan Potgieter said:
When NFB is applied ig2 rises up sharply on clipping, it is a nice feature: I use it to control anti-clipping optical compressor.
According to my musical tastes slight compression is the lesser evil than clipping, even though tube clipping sounds less harsh than SS clipping.
Screen stopper
I don't know whether I'm qualified to have an opinion on this, but would it be reasonable to use zener diodes instead of screen stopper resistors? The typical applications in some tube manuals specify a specific screen voltage, and it seems to be that that could be conveniently obtained by using zener diodes of appropriate rating. That should also work well with an output transformer having screen taps.
I don't know whether I'm qualified to have an opinion on this, but would it be reasonable to use zener diodes instead of screen stopper resistors? The typical applications in some tube manuals specify a specific screen voltage, and it seems to be that that could be conveniently obtained by using zener diodes of appropriate rating. That should also work well with an output transformer having screen taps.
Re: Screen stopper
"Stopper" is a jargon term that means stopping of parasitic oscillations by adding of energy losses to an oscillating system, such as a resistor in series with a capacitor or in parallel with an inductor. Zeners are devices that have such combination of properties like relatively high voltage drop, but relatively low dynamic resistance. Also, they have a capacitance that is much higher than typical resistors have.
Zeners that regulate voltage in respect to cathode of the pentode may be used to decrease distortions.
Zener from B+ source to screen grid reduces constant part of B+ voltage on a screen grid, while almost all variations of B+ voltage will be applied to it. Is it what do you need, or not?
FRE said:
"Stopper" is a jargon term that means stopping of parasitic oscillations by adding of energy losses to an oscillating system, such as a resistor in series with a capacitor or in parallel with an inductor. Zeners are devices that have such combination of properties like relatively high voltage drop, but relatively low dynamic resistance. Also, they have a capacitance that is much higher than typical resistors have.
Zeners that regulate voltage in respect to cathode of the pentode may be used to decrease distortions.
Zener from B+ source to screen grid reduces constant part of B+ voltage on a screen grid, while almost all variations of B+ voltage will be applied to it. Is it what do you need, or not?
The last tube amplifier I had was a Dynaco stereo 70 which I assembled in about 1961. The schematic shows the EL34 screens directly connected to the screen taps on the output transformers, but I've seen some amplifier schematics with resistors connecting the screens to the output transformer taps. I had assumed that the resistors were intended to drop the voltage and was unaware that they were to damp parasitic oscillations. I don't know exactly when zeners became available, but probably it wasn't 'till after about 1960.
Of course zeners would not damp oscillations, so I can see that if oscillations are a problem, resistors should be used. I guess that that shows that making changes to circuits could have unexpected results.
After I move into my new house, I'll try to find a schematic for an amplifier using 6V6s in push-pull mode.
Of course zeners would not damp oscillations, so I can see that if oscillations are a problem, resistors should be used. I guess that that shows that making changes to circuits could have unexpected results.
After I move into my new house, I'll try to find a schematic for an amplifier using 6V6s in push-pull mode.
The usual series resistors of 1K in EL34 screens are advised by Mullard. In graphs with and without those resistors, distortion is lower with the resistors, especially at higer outputs. I have taken this as a peculiarity of EL34s (manufacturers are not often at fault on their own products!).
I have never used zeners as screen droppers, but can see no reason why they would not work, especially in UL where one is limited to the same h.t. while it is sometimes advantageous to have lower Vg2. I would bypass them with say 470nF, and for the EL34 one can have the 1K extra. (Zeners are shorts to signals.) one must just remember that zeners have a temperature co-efficient; at 100V it can be quite 10V, which can have an effect on maximum power output. (Amplified zener packages can be built - but that is another story.)
I have never used zeners as screen droppers, but can see no reason why they would not work, especially in UL where one is limited to the same h.t. while it is sometimes advantageous to have lower Vg2. I would bypass them with say 470nF, and for the EL34 one can have the 1K extra. (Zeners are shorts to signals.) one must just remember that zeners have a temperature co-efficient; at 100V it can be quite 10V, which can have an effect on maximum power output. (Amplified zener packages can be built - but that is another story.)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.