Making a capacitor discharge tool

[redwire]

I'm making a new capacitor discharge tool. Asking for suggestions for sizing the discharge resistor.
Mr. Carlson's Lab High Voltage Capacitor Discharge Tool 1 is using 4 of 1kΩ 50W aluminum (mount to heatsink) resistors in series for 4kΩ 200W, so it's big, which is fine.
His Patreon but I don't have a paid subscription for the project details but it looks straightforward. The plastic box with aluminum plate can't get rid of any amount of that heat though.
Wirewound resistors can be overloaded by a factor of 5 for 5sec. and a factor of 10 for brief times. I've never tried say 20x.

I did some math and simulations:
-Microwave oven say 2uF cap at 2kV is a huge spike, 4kΩ resistor peak power 1,000W done after 40msec though.
-But then SS audio amps 20,000uF cap at 85V is slowww even with 1kΩ resistor is almost 2 minutes I'll fall asleep.
-SMPS say 680uF cap at 340V, 500Ω resistor peak power 231W done after 2 sec.

So I am thinking of using a switch to select all in series or all in parallel for the say 3-4 power resistors. That would allow to choose say 4kΩ or 1kΩ.
Or use lower resistance like 2kΩ for the HV caps but that is 2kW peak 1A and 10x overload for a 200W bank, for a microwave oven cap or RF transmitter. If that doesn't cause drama.

Just looking at ideas for now.

[eccerr0r]

Re: Making a capacitor discharge tool

Looks like a lot of people are coming to the right conclusion: one size does not fit all.

Technically speaking the wattage of the resistor can be a lot lower than the calculated wattage which is only valid *at the beginning* of the discharge. The voltage will drop and if you compute the RC time constant to be fairly short (say milliseconds), resistors can eat small spikes, even like 4-10x without difficulty - as long as it's very short in duration. Millisecond durations are even shorter...

A lot of the time I just use incandescent light bulbs for discharging capacitors with the added bonus of the light bulb going dark when it's finally discharged...

[sam_sam_sam]

Re: Making a capacitor discharge tool

A lot of the time I just use incandescent light bulbs for discharging capacitors with the added bonus of the light bulb going dark when it's finally discharged...

What type of incandescent light bulb do you use for discharging capacitors

[CapLeaker]

Re: Making a capacitor discharge tool

Well… how often are you actually discharging very high voltage caps? The REIL’s I wanted to fix, just had a small discharge resistor across the 3000V HV cap. That’s it.

[eccerr0r]

Re: Making a capacitor discharge tool

What type of incandescent light bulb do you use for discharging capacitors

Again one size does not fit all. Completely depends on the capacitor and charge that needs to be dissipated.

[redwire]

Re: Making a capacitor discharge tool

I'm tired of the hassle of looking for something to do it, a resistor, clips etc. and it usually goes unsafe in a rush. Mostly it's tube amps at 550VDC or SMPS at 340VDC, and the odd oven, although I had a 0.02uF at 20kV and it killed a multimeter. It's when you run into a charged cap unexpectedly that things go bad.

To discharge a microwave oven cap, I wear safety glasses and gloves and use a screwdriver if I can get at the terminals (they're usually recessed and with insulated crimps) then I'll use a jumper wire with pokers. Realistically, microwave oven caps, start/run caps have a 10MEGΩ internal discharge resistor so waiting a minute can also work- assuming that the resistor has not failed. It's very high voltage to say 3kV/1.3uF but low energy around 6 Joules.

I think I need three resistance ranges:
2-4kΩ for microwaves or HV gear around 2-3kV.
For tube amps, SMPS around 500VDC, around 500Ω-1kΩ.
For solid-state amps around 30-85VDC, around 100Ω.
In the end i.e. 3kΩ 150W, 1kΩ 150W and maybe 100Ω 50W.

I looked at a dozen different 25W and 50W wirewound resistors.
For aluminum-housed, first limiting factor is the dielectric insulation from resistor to housing. You don't want an arc from leads to housing. Mr. Carlson insists on a plastic box for that reason.

Dielectric rating (element to housing) is around 3,000VAC/DC for 50W and most 25W types with resin filling.

Max. element voltage around 1,250V so 3 of 50W in series gives 3,750V max. but 25W types are only good for 550V max.

Still number crunching, but I'm going to use switches for lower value resistors for lower voltages.

[redwire]

Re: Making a capacitor discharge tool

I'm settling on these resistor values and a DPDT center-off switch.
Three of 2kΩ 50W and one of 100Ω 50W.
Gives 6kΩ 150W or 667Ω 150W or 100Ω 50W impulse power and discharge currents are under an amp.
Next is the LED or buzzer circuit indicator.

[petehall347]

Re: Making a capacitor discharge tool

dont forget capacitors can have a habit of charging themselves up after being discharged

[lotas]

Re: Making a capacitor discharge tool

Here are the finished devices
https://aliexpress.com/item/1005003120177533.html
https://aliexpress.com/item/4001191972846.html

[redwire]

Re: Making a capacitor discharge tool

I did see those but rated 1,000 chinese volts and do they make anything strong?
The first one is 10k resistor unknown wattage and a review says it caught on fire at 700V.
The second one uses IGBT 1200V 15A but where is the heatsink for it?
I guess these look not so great and expensive for what you get. I can get 50W resistors for $1 and not worry about a fire or blown IGBT.
Powering the voltmeter off a wall wart is not a good idea with high voltage coming in, it would be best off a battery.

[CapLeaker]

Re: Making a capacitor discharge tool

… chinese volts… LMAO!

[redwire]

Re: Making a capacitor discharge tool

Just multiply by 1/2 to get the real volts, amps, watts

so I got more of my circuit done, will post it tomorrow.

[stj]

Re: Making a capacitor discharge tool

so with resistors, the voltage is usually dictated by the distance between the end caps because of the risk of arcing.
high voltage resistors are longer.
we used to get round this by using several resistors in series, shoved into a plastic or glass tube and filled with epoxy or RTV to keep moisture out.

[redwire]

Re: Making a capacitor discharge tool

I'm working backwards from power resistor specs. Under 1A peak discharge current I thought is reasonable. 2,500V for HV worst case.

Aluminum-housed 50W in the most auspicious RX24 series made by villagers in Elbonia are looking good for price- until I found the pic of the sand and airgap lol. Know of any better quality Ali vendors? Their price isn't much lower than NA disti's.
RX24 50W some have a dielectric voltage rating of 2,000VRMS for 1 minute, this is with either lead to case.
Arcol Ohmite HS50 is 2,500VRMS 1 minute 3,500VACpk, element 1,250V max.
Stackpole KAL50, TT WH50 are rated 3,000V.

Mr. Carlson was concerned about an arc from a resistor to the housing/metal box, so he mounted them to a metal plate and then on standoffs in a plastic box. I'm just using a die-cast box touching nothing lol.