Choke bleeder resistor


2017-11-25 11:50 pm
Is there a "rule of thumb" for bleeder resistor calculation for let us say a LC PSU for a power amp where a rail fuse is implemented?
The bleeder resistor purpose is not for discharge of cap but to "eat" the energy of the choke if fuse blows to protect the caps from high voltage.
In my case I have a 11mH choke with approx. 1.5 A bias (24VDC) and I am thinking of a resistor at about 150 ohm.
No one seems to take this issue serious (for cap discharge only the resistor is usually much larger)?
The resistor was meant to be part of the load so placed same place you would place a cap bleeder but before the rail fuse so there always is a load the choke can deliver its energy to. I was not thinking of "short" the choke with the resistor. So it was just if there was an easy way to calculate such a resistor if fuse blows that the "spark" from the choke does not destroy the caps.
Ah, I see what you mean.
The choke acts as a resonator or reactor when current is drawn, when no current is drawn there is no reactance. I think you mean keeping a permanent load ... but how will you reduce the resistive load as the required load is increased I wonder.

I see no advantage in loading a choke in the way you describe, as all the energy is wasted in heat and there is no advantage to the power supply, if anything when you do draw current from the load, with your resistive load active, the choke will not work correctly.


Paid Member
2007-09-15 8:14 am
Why not add a highish wattage zener and lowish value series resistor in series with the zener across the rail. That would dump any excess voltage/energy.

Can't think :) what polarity is this unwanted pulse of energy going to be?
This was answer to Jon Shell......Mooly came just in front :)
Yes, but if you suddenly break the DC current in a choke and the choke has no way to get rid of the stored energy it will create high voltages spikes (like an ignition choke in a car) so it was just have a load if rail fuse burns…...a load just a fraction of the normal load when amp is on but a load that reduces the spikes to a level where the filter cap is not damaged. Maybe it is not a real problem and therefor I have not been able to find any information :)
I have another choke loaded PSU where I just by "guessing" have a "bleeder". One time the fuse burned and the flash light from fuse was quite big.....and fuse was all black afterwards so something goes on...…
Also when the fuse breaks in the primary of a large toroid transformer it can be quite "wild".
To Mooly:
This could be an ok solution and there also exist those "discs" you sometimes have at the 240 VAC side to take high voltage spikes at the primary (they have same form factor as a NTC disc). I can't remember the name on those…..
A zener that works as zener in both directions?
Yes, I can search the web for such a low voltage type.
I found the attached in my "back log" but those are for line level voltage.


  • DSC_4062_00001.jpg
    583.9 KB · Views: 79


2012-12-19 7:24 pm
If the fuse blows and the choke continues the current, the capacitor between rectifier and choke get discharged.
When the voltage across the capacitor reach zero and there is still current left it gets charged in the opposite direction, not good.
Remedy is simple, put a diode over the capacitor.
Now I understand what you want to do.
Sounds to me like a problem that isn't there unless the power supply is liable to be unregulated, in which case not a good design.
Use a shunt regulator; zener diode and a transistor. When the voltage gets too high, the zener turns the transistor on and shunts the excess voltage.
A varistor will act as a short circuit and bear in mind the power or Joule rating is for a few micro seconds only otherwise it will go bang.
I remember Guido from Tentlabs told me to add a big 5 watt zenerdiode across the capacitor feeding his shunt regulator. Because i used a rather big henry choke with high dcr as an input choke i was afraid that if no load present for whatever reason the voltage might rise well above the maximum rating of the cap. So mounting a 22 volt zener across the 25 volt cap would make a perfect protection.. an easy way to avoid using a 40 volt cap where the actual voltage would be around 18 volt and keeping the input voltage for the shunt always well below the maximum input voltage. Greetings, Eduard
Now I have attached an quick schematic to show what my idea is. I forgot the rail fuse which is after the cap. I will have two of those pr. mono block to make +-24V for both.

So first a SMPS 24VDC and then a small common mode choke to take the worst switching noise. Then a serial DC choke 11mH (5A rated) for additional filtering. My idea with DC choke is also that during power up it should limit the charge up of cap so SMPS does not go into protection. It will probably be some "try and error" as I don't know how intelligent the "error recovery" is made in this type of SMPS. The DC current will be about 1.5A during operation.

The main idea was to try a SMPS based PSU and make it "super duper" for audio. The 4 pole Jensen Capacitor caps are quite expensive so will not blow them up.

Do you think a diode reversed is enough to protect the caps?
I have ordered the varistors so can do both. So bleeder resistor like 1k + varistor + reverse diode?


  • DSC_4063_00001.jpg
    109.1 KB · Views: 62


Paid Member
2007-09-15 8:14 am
You have to to member that SMPS doesn't always like (and need) lots of capacitance.

If you are experimenting then you must rely on measurements to determine what works best... and it could be that small values such as 100uF or lower work best.

Lots of capacitance also impairs the regulation of many SMPS. They don't need it.