kodabmx,
I hope you have bleeder resistor(s) on your B+(s). If you rely on a neon to indicate the presence of B+, and the neon dies, you may be surprised! Most bleeders take at leas 15 seconds or more to discharge the B+ (it might be enough time for you to remove the bottom plate before getting your hands inside).
A 500V B+ with a 50k Ohm bleeder and filter caps totaling 200uF will have the following discharge:
1 RC time constant = 50k Ohms x 200uF = 10 seconds
4 RC time constants = 40 seconds
In 4 RC time constants, it will discharge to 16% of the original voltage . . . 16% x 500V = 80V.
Grab that, and it will smart!
5 time constants will discharge to 5% of the original voltage, 500V x 5% = 25V. OK
I hope you have bleeder resistor(s) on your B+(s). If you rely on a neon to indicate the presence of B+, and the neon dies, you may be surprised! Most bleeders take at leas 15 seconds or more to discharge the B+ (it might be enough time for you to remove the bottom plate before getting your hands inside).
A 500V B+ with a 50k Ohm bleeder and filter caps totaling 200uF will have the following discharge:
1 RC time constant = 50k Ohms x 200uF = 10 seconds
4 RC time constants = 40 seconds
In 4 RC time constants, it will discharge to 16% of the original voltage . . . 16% x 500V = 80V.
Grab that, and it will smart!
5 time constants will discharge to 5% of the original voltage, 500V x 5% = 25V. OK
Actually, only 1.8% of the initial voltage remains after 4 RC time constants. Starting at 500 V, only 9.1 volts will be left, which is quite safe to touch.
Actually, 5 time constants gets you to a bit less than 1% of the original voltage (0.67% to be more exact.) Starting with 500 volts, that would leave about 3.4 volts on the cap, again, quite safe to touch.
With the high-efficiency LED circuits I've been talking about, R is unlikely to be as low as 50k, and may be ten to twenty times larger. If you really are using a (rather unusually large) 200 uF filter cap, and a 1M LED resistor, then one RC time constant is 200 seconds - more than three minutes - and waiting for five time constants translates to at least a 15 minute wait.
Five time constants or not, please be safe - always check that B+ voltage with a DMM before you assume it's safe to touch!
-Gnobuddy
Gnobuddy,
You are absolutely right. I missed the voltage discharge after 5 RC time constants by almost 1 order of magnitude.
I often use two 25k Ohm 5 Watt bleeder resistors in series. For a 300V B+ supply, that is 1.8 Watts in 10 Watts of resistors, they do not get too hot, and should not fail. The remaining voltage is about 11V after 3 time constants (200uF x 50k Ohms x 3 = 30 seconds).
200uF total capacitance for a CLCRCRC filter (with a very small input cap) is not unusually high (or at least not when you have to get less than 100uV ripple from a single ended triode amp that does not have regulated B+.
You are absolutely right. I missed the voltage discharge after 5 RC time constants by almost 1 order of magnitude.
I often use two 25k Ohm 5 Watt bleeder resistors in series. For a 300V B+ supply, that is 1.8 Watts in 10 Watts of resistors, they do not get too hot, and should not fail. The remaining voltage is about 11V after 3 time constants (200uF x 50k Ohms x 3 = 30 seconds).
200uF total capacitance for a CLCRCRC filter (with a very small input cap) is not unusually high (or at least not when you have to get less than 100uV ripple from a single ended triode amp that does not have regulated B+.
Last edited:
Sorry, I forgot there are two alternate universes when it comes to valve circuits! 🙂
-Gnobuddy
- Status
- Not open for further replies.