I plan on using a couple of .33 200VDC Vitamin Q coupling caps in the design below instead of the .47 caps shown. My question is this. Are the 200VDC Vitamin Q caps rated high enough voltage wise for this position? AC wise they will be fine. I'm thinking that since there really won't be a DC voltage drop across them (since they are connected to the grids and not to ground). Am I looking at this correctly? Thanks.
G
G
Attachments
Use a 400V cap there. The grid is at ground potential. And the cap will see the full B+ at turn-on, not to mention what happens with large signal swings. You might have a 200V cap survive because of the current limiting inherent in the grid resistor, but I wouldn't want to chance it, since the downside will involve smoke, flame, and melting glass.
Don't know what you've selected for C2 (10uF/400Vdc) but the Vishay 735P series available through Mouser have extremely low ESL and ESR - using them for your B+ bypassing ought to help preserve maximum bandwidth (heck, I'd use on for the bulk filter capacitor and a smaller 1uF/0.1uF combo near the plate resistor for bypassing.
For coupling, I'd use one 1uF/400VDC 735P series cap, unless you absolutely must have paper dielectric, in which case "line to ground" safety capacitors might be worth looking into (they invariably use paper dielectric because of its unparalleled self-healing properties).
For coupling, I'd use one 1uF/400VDC 735P series cap, unless you absolutely must have paper dielectric, in which case "line to ground" safety capacitors might be worth looking into (they invariably use paper dielectric because of its unparalleled self-healing properties).
> isnt the 15k resistor dropping 150V at 10mA? Therefore, only 150V is seen at the coupling cap.
At turn-on, the B+ rises quickly to 300V or more. The tube sits there stone-cold, unconductive, for 5 to 10 seconds. C3 sees 300+V in series with 15K+330K= 345KΩ.
Assume the unloaded power supply is 300V. (I suspect it will be over 400V, which makes things worse.)
Because of the resistance, the cap charges to 300V "slowly". But not as slow as a tube. In 0.114 seconds it will charge to 63%, which is 189 volts. By 5 seconds when the tube is starting to conduct, the cap will be around 295 volts.
Over-voltage damage to a cap is instant, unlike tubes where a slight overvoltage just shortens life.
Most caps have some margin on voltage rating, so the cap may not die the instant it hits 201 volts. However if it does die when you put more than rated volts on it, you have no good cause to complain.
The old "metalized" caps had non-uniform (rough paper) insulation. Some parts of a "200V" cap would break-down at 100V, other parts at 600V. The metal was VERY thin, so if you got a break-down in a high-current cuircuit, the small area of metal around the weak-spot would burn-away and become useless. "Self-Healing". At the factory, they would blast the caps with the rated voltage (blowing-out the weak areas) and then check for minimum capacitance. So if you put 250V across a 200V 0.1µFd cap, it would quickly become a 250V 0.09 or 0.08µFd cap. 400V surges could burn-away half the metal and leave you with 0.05µFd. In lowest-price radios, this was acceptable. In "good work", you don't want to trust any "self-healing" scheme.
In many circuits, the worst voltage surges are not in-operation but at turn-on. Sharpen your pencil and your mind and watch what happens on paper as the circuit comes to life.
At turn-on, the B+ rises quickly to 300V or more. The tube sits there stone-cold, unconductive, for 5 to 10 seconds. C3 sees 300+V in series with 15K+330K= 345KΩ.
Assume the unloaded power supply is 300V. (I suspect it will be over 400V, which makes things worse.)
Because of the resistance, the cap charges to 300V "slowly". But not as slow as a tube. In 0.114 seconds it will charge to 63%, which is 189 volts. By 5 seconds when the tube is starting to conduct, the cap will be around 295 volts.
Over-voltage damage to a cap is instant, unlike tubes where a slight overvoltage just shortens life.
Most caps have some margin on voltage rating, so the cap may not die the instant it hits 201 volts. However if it does die when you put more than rated volts on it, you have no good cause to complain.
The old "metalized" caps had non-uniform (rough paper) insulation. Some parts of a "200V" cap would break-down at 100V, other parts at 600V. The metal was VERY thin, so if you got a break-down in a high-current cuircuit, the small area of metal around the weak-spot would burn-away and become useless. "Self-Healing". At the factory, they would blast the caps with the rated voltage (blowing-out the weak areas) and then check for minimum capacitance. So if you put 250V across a 200V 0.1µFd cap, it would quickly become a 250V 0.09 or 0.08µFd cap. 400V surges could burn-away half the metal and leave you with 0.05µFd. In lowest-price radios, this was acceptable. In "good work", you don't want to trust any "self-healing" scheme.
In many circuits, the worst voltage surges are not in-operation but at turn-on. Sharpen your pencil and your mind and watch what happens on paper as the circuit comes to life.
Hi,
Absolutely...there's more to it though...I often find people calculating caps under static conditions forgetting what happens when an AC signal is superimposed...
The turn on surges can be lessened by using tube rectifiers and ...they're so much easier on components, aren't they?
So, now you know why grandma's radio lasted forever...no excuse why your tube amp shouldn't last forever...
And it can sound great all along...
Cheers,
Absolutely...there's more to it though...I often find people calculating caps under static conditions forgetting what happens when an AC signal is superimposed...
The turn on surges can be lessened by using tube rectifiers and ...they're so much easier on components, aren't they?
So, now you know why grandma's radio lasted forever...no excuse why your tube amp shouldn't last forever...
And it can sound great all along...
Cheers,
Positron,
I see your confusion: DC coupling=infinite value cap? Not quite.
DC coupling: A transient can cause grid current to flow. Once the transient is over, the circuit is almost immediately* back to normal.
AC coupling using oversize cap: A transient causes grid current to flow, causing the charge across the cap to rapidly change (due to the lower grid impedance under these conditions). Once the transient is over, the circuit takes a finite time for the cap to restore it's original charge.
The smaller the cap value, the faster the circuit normalizes.
This is one reason that transformer coupling is favoured by some of us over CR coupling.
Cheers,
Almost immediate* assumes a perfect power supply.
I see your confusion: DC coupling=infinite value cap? Not quite.
DC coupling: A transient can cause grid current to flow. Once the transient is over, the circuit is almost immediately* back to normal.
AC coupling using oversize cap: A transient causes grid current to flow, causing the charge across the cap to rapidly change (due to the lower grid impedance under these conditions). Once the transient is over, the circuit takes a finite time for the cap to restore it's original charge.
The smaller the cap value, the faster the circuit normalizes.
This is one reason that transformer coupling is favoured by some of us over CR coupling.
Cheers,
Almost immediate* assumes a perfect power supply.
No confusion
I see where you are coming from, no confusion there, but I was thinking of what the string was initially, mainly concerned about, bass response. With such a small time constant, are we going to get flat bass response vs DC coupling? Just a question.
I was aiming at another point too; in general how the size of the cap affects the sonic qualities before grid current flows. I could have been alot more clear. Sorry about that.
Steve
I see where you are coming from, no confusion there, but I was thinking of what the string was initially, mainly concerned about, bass response. With such a small time constant, are we going to get flat bass response vs DC coupling? Just a question.
I was aiming at another point too; in general how the size of the cap affects the sonic qualities before grid current flows. I could have been alot more clear. Sorry about that.
Steve
Hmm,
Looks like I was the one a little confused with the subject
So, in answer to your question, there is no difference in bass response between DC coupling and an infinitely large coupling cap.
As for cap sonic qualities, that's a contentious issue with many opinions. I just try to choose the best quality cap (of the right kind) that I can afford.
Am I closer now?
Cheers,
Looks like I was the one a little confused with the subject
So, in answer to your question, there is no difference in bass response between DC coupling and an infinitely large coupling cap.
As for cap sonic qualities, that's a contentious issue with many opinions. I just try to choose the best quality cap (of the right kind) that I can afford.
Am I closer now?
Cheers,
Understand
I understand DH. I could have been more clear. I was thinking of DC vs, say the .47uf cap.
As you mentioned, infinite value cap under perfect conditions should sound like DC coupled. But it seems even a 1uf cap starts to color the sound, regardless of brand, even if just a little.
Wonder also about the field each layer of "foil" pesents to other layers when charging and discharging, since current is flowing through the foil.
Any thoughts?
I understand DH. I could have been more clear. I was thinking of DC vs, say the .47uf cap.
As you mentioned, infinite value cap under perfect conditions should sound like DC coupled. But it seems even a 1uf cap starts to color the sound, regardless of brand, even if just a little.
Wonder also about the field each layer of "foil" pesents to other layers when charging and discharging, since current is flowing through the foil.
Any thoughts?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.