Capacitor Leakage Tester

Intro: Capacitor Leakage Tester

This tester can be used to check smaller value capacitors to see if they have leakage at their rated voltages. It can also be used to test insulation resistance in wires or to test a diode's reverse breakdown characteristics. The analog meter on the front of the device gives an indication of the current going through the device under test DUT and the multimeter gives the voltage across the DUT.

NOTE OF CAUTION: THIS DEVICE DEVELOPS VOLTAGES UP TO 1000 VOLTS WHICH CAN BE LETHAL IF THIS DEVICE IS MISUSED. ONLY BUILD THIS DEVICE IF YOU UNDERSTAND THE SAFETY PRECAUTIONS FOR WORKING WITH HIGH VOLTAGES.

STEP 1: How It Works

I have capacitor testers but not a leakage tester which actually measures the current going through a capacitor at its rated voltage. As capacitors age, they start becoming leaky and this tester will demonstrate if they are exhibiting this characteristic. Unfortunately, this tester will not deliver enough current at high voltage to test capacitors of about 1 mfd and above so it isn't very useful for testing electrolytics but excellent for anything below this in value. The best way to test electrolytics is by measuring it's ESR (Equivalent Series Resistance) but that's for another Instructable.

This circuit uses an Astable Multivibrator using (2) 2N3904 transistors running at about 10 kHz. This frequency was picked because the 10-1 ratio miniature transformer worked most efficiently at this frequency. The signal is coupled from the second transistor via a 15 nF capacitor to the gate of a IRF630 MOSFET which is biased at 4.5V between the two 1 megohm resistors. One of the resistors is a variable resistor and it varies the size of the signal getting into the gate and therefore varying the voltage on the output. The drain of the IRF630 is connected to the primary of a 1-10 ratio step up transformer where it is stepped up from approximately 25 volts peak to to around 225 volts peak. This voltage is then applied to a Cockroft-Walton voltage multiplier. The end product is around 1000 volts DC which is applied to two outside terminals with the positive side going through a 0-400 microamp meter movement to the positive terminal. The outside terminals are banana terminals so they fit most standard size meter probes.The 9 volt battery current is supplied through a momentary push button switch when a test is to be made.

STEP 2: Starting Construction

I first took the box and drilled the necessary holes for the potentiometer, push button switch, meter and the two holes for the banana plugs. The box had top and bottom halves so I put all the holes into the flat part of the top side except the banana plug jacks which were drilled into the lower half.

STEP 3: Install Components on Top and Bottom Halves of Box.

Using the correct sized drill bits, drill holes for the potentiometer, push button and switch in the top half of the box and in the lower half, for the two banana plug sockets. The meter opening will need to be drilled, reamed and filed to get it to the right size. Do not install meter at this time as the meter plastic cover needs to be taken off and a new scale needs to be made.

STEP 4: Making the Cockroft-Walton Voltage Multiplier.

I made the voltage multiplier on a piece of vectorboard that was 3 inches by 1 1/2 inches which allowed the components to fit neatly with lots of room. The 13 capacitors and 13 diodes were connected with their own wires together and soldered in place. The AC input goes in one end between two terminals and the positive 1000 volts output is taken from the last capacitor and the right hand terminal of the AC input. This board is transformer isolated from the other board.

STEP 5: Making the Multivibrator Board.

The multivibrator was made on a 3 by 1 3/4 inch piece of vectorboard with the components connected together by their own wires and pieces of bared copper wire. The voltage control potentiometer was connected to the multivibrator board and also the push button switch. The output of the transformer was connected via short leads to the voltage multiplier board. Once the multivibrator board was completed, it was confirmed that it operated at 10 kHz by looking at it through an oscilloscope. The MOSFET was mounted without a heat sink and the whole assembly with the miniature transformer mounted with lots of room to spare.

STEP 6: Making a New Meter Scale.

Take off the plastic cover that's covering the meter. It's secured with tape. Cut a piece of white bond paper to size and shape and very carefully make a scale with 4 equal divisions and mark the beginning as 0 and the end as 400. The divisions should read 0, 100, 200, 300, 400 and write microamps on the bottom. Secure the new scale with paper glue and put the meter cover back. The meter can now be installed on the top cover with hot melt glue.

STEP 7: Wiring Everything Together.

Wire everything together as seen in the schematic and the above photos. The high voltage wiring should either be done with regular hookup wire with a sleeve of heat shrink tubing slipped over the wire. I used old high voltage wire salvaged from an old television.

STEP 8: Once the Unit Is Assembled Test With Scope

Looking at the signal taken at the gate of the MOSFET on the far left picture, we see a 9 volt positive going sawtooth waveform with an approximately 1 microsecond negative going spike caused by the input capacitance of the MOSFET. The second waveform shows the drain of the MOSFET where it connects to the transformer. The waveform is more rounded off until it hits a peak of 20 volts. Note the 25 volt spike at the beginning of the waveform as the primary of the transformer tries to resist the change in current passing through it. The third waveform is of the signal as it comes out of the transformer and is applied across the voltage multiplier input. Here it's approximately 225 volts peak or 159 volts RMS. This will be multiplied in the voltage multiplier to approximately 1000 volts DC.

STEP 9: Trying Out the Capacitor Leakage Tester.

In the first picture the meter is applying approximately 400 volts to a small modern capacitor rated at 400 volts and there is very little leakage, around 25 microamps. The second the same 400 volts is applied to an old fashioned paper capacitor also rated at 400 volts, it's very leaky, passing through 10 times the current. If this capacitor was in a circuit, I would replace it, the other one I wouldn't.

Comments

[dta1161]

Where in the world did you find the transformer?

[Franks Instructables]

If I recall it was one that I had bought years before at the Radio Shack. If you have any old ac adapters that you can sacrifice, it will have a 10-1 transformer in it and will work fine for this purpose as a step up transformer. Lots of scrap electronics have small transformers in them, just do a resistance measurement of the windings. You could also wind your own on a toroid. Good luck.

[dta1161]

Yeah, I really miss Radio shack. They had some things unavailable elsewhere.
Thanks for your reply

[rodger14803]

I have a silly question...I just don't see it: In the schematic, where are the terminals I connect the capacitor under test to?
I think this circuit is just what I am looking for.....

Thanks!
Rodger

[Franks Instructables]

No it isn't a silly question, I looked at the schematic again and realized that it could be confusing. The most positive point is the top right hand side of the voltage multiplier and the most negative point of the voltage multiplier is the lower side of the secondary of the transformer. The most positive point would be the positive terminal and the negative side of the meter would be connected to the most negative point of the voltage multiplier. The positive side of the meter could be the negative test terminal.

[rodger14803]

Thanks for the quick response! So just to be clear: One lead of the "capacitor under test" would attach to the anode of D15 and the other lead of "the capacitor under test ' to the junction of C16 and T1. Am I understanding correctly?

Thanks again!

[Franks Instructables]

Precisely! Good luck Rodger.
Frank

[kwric]

Hi Frank, wow this looks too good to be true, I’ve been looking for a leakage tester for my old valve radio projects. Mainly capacitors below 1uF as I always change the higher value electrolytics due to their age, usually more than 60 years old.
You warn about the dangers of 1000v, but isn’t it the current that is more dangerous, and this is very low current.
Thanks again...
Kevin

[Franks Instructables]

If you had heart problems or had a pacemaker or an implanted defibrillator, I don't think you'd want to get zapped with 1000 volts. Why take the chance? Also, while the voltage and current might not kill you in the classical sense, it could cause uncontrolled muscle movements that could cause a person to fall banging their head etc. It's best just to have a warning on there to just point out that electricity is dangerous, don't take any unnecessary chances.

[jjsaldar]

Good morning Mr. Frank
This interesting project is intended for which type of capacitors (mica, polyester, ceramic, paper, electrolytic?).
What is the range of voltages and capacitance I can measure for the capacitors I want to determine their leakage.

I appreciate your help and thank you in advance for your response.
Cordially,

John J. Saldarriaga

[Franks Instructables]

You can test all the capacitor types you mentioned at their rated voltage. I wouldn't test electrolytic capacitors above 2 uFd with this device, because the Equivalent Series Resistance tests are much better for electrolytics. This device will test devices up to 900 volts but set the voltage with a meter before you check the capacitor. You can also use this device as an insulation tester.

[rollow2]

Thanks for explaining that, did compare both mosfet's spec sheets but obviously my knowledge was limited. Will either find a 630 or go with 2 9v batteries, there's room in the box I created for it.

[rollow2]

Hi, I only get about 167V on output at the transformer, used a 640 because it seems a decent replacement. FYI, the transformer works as expected at 10KHZ. Using a sine wave generator at 2.72V gives approx. 25.7V on output.
What would I have to do to get to 225? Other then that, thanks a lot for your instruction. Very handy tool for old caps.

[Franks Instructables]

The IRF640 has an internal resistance of .18 ohms and the IRF630 has an internal resistance of 4 ohms. The lower resistance has changed the output voltage. Try increasing the supply voltage to 12 volts or even 24 volts, that will increase the level of your output signal. If you want to have the convenience of the small form factor of the 9 volt battery, there are small 12 volt batteries that you can buy that are designed for key fobs. They're called A23 batteries and you can buy them in a two pack. You could easily make a battery holder for one or two of these in series for more output voltage. The transistors should have no trouble handling the higher voltage. Just experiment with different supply voltages until you get the output voltage that you want. Good luck.