Using Energy Rated Resistors for High Voltage Capacitor Discharging - A Discharge Probe Project

When my fascination for electronics began (I think that was in the crib), I played with a lot of tube equipment. My first job was working in a television repair shop at the age of 14. A television picture tube can hold a charge for a long time. The standard anode voltage was about 25,000 volts. What business does a 14 year old kid have messing around with that kind of hazard? Needless to say, I've been drawing sparks with a screwdriver for a long time.

General

Guitarists love tube amplifiers, and so do I. The capacitors in tube amplifiers can hold a dangerous charge. Some drain their charge fairly quick, and some take longer. Before you work on one, you really need to get those capacitors discharged. There are a few ways to accomplish this.

1. Dead Short Across the Capacitor - Hold a screwdriver against the chassis to ground the screwdriver (or attach a clip lead from the chassis to the screwdriver shaft) and touch the tip of the screwdriver to the non-grounded lead of the capacitor. POW! The problem with this approach is, the instantaneous current in the capacitor when shorted is very high. Tests and studies have been conducted by capacitor manufacturers that show very high current transients can damage the capacitor. The oxide layer, which is the dielectric of an electrolytic capacitor, is degraded each time this happens. There is a also a high current when the amp is powered up, but the transformer resistance, transformer inductance, power filter choke inductance, and power line resistance does a great job limiting the inrush current to a moderate level. Some modern amps use series thermistors to relieve the inrush current.
2. Working the Amp Down - If the standby switch is on, play a guitar like a rock & roll star until there is no longer any sound coming out of the speaker. The theory is, the signal going through the amp will drain the energy stored in the capacitors. The problem is, when tube filaments lose power as soon as the power switch is turned off. The emission of a tube dies very quickly thereafter. If the tubes don't conduct due to a cooled filament, the amp doesn't draw much current and will not sufficiently discharge the capacitors.
3. Patience - The capacitors in most amps with drain to a fairly low voltage if you just give it some time. But, they will still have some voltage across then for a while. If the voltage is low, let's say below 50V, discharging it with method one is a little less harmful. It's still not a nice thing to do to a capacitor
4. Discharge Through a Resistor - A resistor placed across the capacitor can discharge the capacitor safely. I heard of folks shorting the plate terminal of a tube, which uses the plate resistor as a built-in discharge resistor. The problem is, the plate resistor for most stages is about 100k ohms. That could take as long as 25 seconds to discharge a 100µF capacitor from 500V to 50V. A smaller resistance would work faster, but the trouble is determining the correct wattage to select.

There is no doubt a discharge resistor is the right approach to limit the instantaneous current in the capacitor. The challenge is what resistance and wattage to choose. The resistance is easy. Here is the equation for the voltage across a capacitor when a discharge resistor is placed on it.

Where:
E = the initial voltage on the capacitor
e = the base of natural logarithms 2.71828
t = time is seconds
= tau or R x C
Vc = The voltage on the capacitor after t seconds

Of course, this equation is theoretical based on an ideal capacitor. In reality, a capacitor has leakage and will discharge faster on its own leakage current to various extents.

Being conservative let's determine a worst case discharge capacitance and voltage. A very standard example is the (2) 70µF capacitors in series in a Fender Twin Reverb. This is 35µF total. Other capacitors are also tied to this node through an inductor and various resistors. Let's say the largest amount of capacitance that would practically need to be discharged is 100µF, and the largest voltage is 500V. Let's calculate the voltage in the capacitor will be when discharged through a 200 ohm resistor after 1/10 second.

That's really nice. It would be practically zero after 1 second.

Now let's determine the power rating. Let's see what the instantaneous power is in a 200 ohm resistor.

Of course the current power will rapidly go to zero, but realistically there is not a good way to determine the wattage since we don't know how fast the inside of the resistor can heat up. One common discharge resistor I have seen is 150 ohm 10 watt. But, how do I know I'm not damaging the resistor every time I discharge it?

Energy Rated Resistor

There are companies who make resistors that have an energy rating. It is easy to calculate the energy in a capacitor. We can forget about power.

Where:
Wc = Energy in Joules
C = Capacitance
E = Voltage

The energy of 100µF capacitor at 500V is,

The Ohmite Corporation has a nice selection of energy rated resistors. The 30 series are leaded wirewound components are axial lead parts.

Let's consider how to package the discharge probe. Some discharge resistor applications simply connect a resistor with wires on both ends, and use heat shrink tubing to insulate the resistor in the middle. The issue with that is, the high voltage is on one wire all the way down to the resistor. If the insulation becomes defective, it could become unsafe. The ultimate would be a probe tip in which a resistor can fit inside. Most voltmeter probes are too small for a resistor like we are considering. Some folks have used a ball point pen body for such an application. The largest diameter ballpoint pen I could easily find was the Wide-Body Bic offering.

I performed a HIPOT test of the pen body at 1,500 volts and there was no measureable current. Use this pen or any other at your own risk. I can't guarantee that next months lot of Bic pens will not have some impurities or chemical changes that will degrade its dielectric strength.

The 10 watt series 30 offering is too large in diameter at 0.406". However, the 5 watt series fits nicely. The largest resistance value on the 5 watt series is 75 ohms. At 75 ohms, the initial current is 6.66A. Since the fusing current is 5.55A, I'm not comfortable with this choice. Putting (2) 75 ohm resistors in series will fix that problem.

Resistance: 150 ohm (Ohmite 35J75R)
Initial Current: 3.33A (fusing current of resistor is 5.55A)
Energy: 12.5 joules (resistors are 18.77J x 2 = 37.54J)
Discharge Voltage: 0.636V in 0.1 seconds

Construction

Disassemble the pen from the clip side only as shown. Discard the encircled parts.

Click on picture for a higher resolution image

I used a 6D finishing nail for the probe tip. It was a perfect diameter for the hole in the pen. I was quite surprised I could solder to a nail. I cut the head off and chucked it up in a drill press. I used a file while the bit was rotating to round the tip off and remove the sharp end.

Click on picture for a higher resolution image

The distance from the resistor body to the probe end of the pen is ~1.7". I made the distance from the end of the probe to the resistor body ~2". The resistors and wire must be connected very close to fit the length of the pen body. I used a Pomona 565-AL-B-24-0 clip lead with one clip removed. Don't forget to thread the wire through the clip part of the pen before soldering the wire to the resistor. I also put a generous amount of clear silicone sealant (RTV) around the nail probe where the solder joint is located just prior to insertion. This should help to hold the probe in place against any axial force exerted by the probe tip.

Here's the finished probe.

Click on picture for a higher resolution image

If you use P-Touch software for a Brother label printer, I linked the label here.

These parts are available at Mouser.

Ohmite 35J75R Wirewound Resistor - Mouser Web Page Link
Pomona 565-AL-B-24-0 Clip Lead Mouser Web Page Link

On a final note, it should be noted that capacitors will have some energy in them after they have been discharged. Many capacitors have a "memory". If they were previously charged, they like to remain charged. This is called dielectric absorption. It's a good idea to connect a dead short clip lead across the capacitor after discharging them with a resistive probe.

©2010 Terry Downs Music™