electrical guru ?? testing capacitors

[glennster]

having problems with one of the shop mig welders, snap on 110v. intermittent heat issues when welding. tested the diodes according to the manual and they check out. thinking maybe a capacitor problem . it has 4 12000 uf 50v capacitors in it. heres the label off the cps

nippon
chemi-con

36de 95 deg c
12000 uf 50v
0304l 24
positive


any ideas how to test these?? i have a dvom meter.
thanks glenn

 

[John T]

IT takes a REAL capacitor tester to do it properly which will indicate its capacitance and leakage etc. However, a capacitor must read a "basically" open circuit to straight DC once its charged and NOT be permanent short circuit to DC or else its baddddddddd. A capacitor will pass AC remember. Soooooo a DC VOM can to some extent tell you if its badddddddddd but not necessarily how good it is or its capacitance value.

OPEN circuit (To straight DC) once charged, is good..........
NOT a short circuit to DC, thats BAD..........

With my old analog Simpson VOM I used to run what I call a kickback test as follows:

IM NOT SAYING THIS WILL WORK ALL THE SAME FOR A DVOM, ONLY THAT IT WORKED FINE FOR MY OLD ANALOG SIMPSON ON SMALL CAPACITORS. I bet a motor shop can test them for you

1) Put it on R x 1 scale and insure it eventually reads an open circuit (if a short its baddddddd)

2) Discharge it and put meter on R x 10,000 scale then apply the meter and the needle should first initially swing over to the short side as its charging but then kick back to the open circuit side as it charges up.

THIS IS NOT A PERFECT TEST (before anyone has a calf lol) but it can give you an idea AND IT CAN SHOW IF ITS BADDDDDDDDD like a dead short or to some extent its open and doesnt accept charging current. This test ALWAYS WORKED FOR ME. Others may have different tests or methods absent a true capacitor tester and they will hopefully add to my findings and experience.

Dont forget to discharge them before and after these tests.

Hope this helps

John T

 

[glennster]

john, how do you charge the capacitors?

 

[Bob M]

A quick test for open/shorted capacitor:

Disconnect one lead on each cap to isolate it. Set your DVM to resistance (ohms) x 10,000 or autorange resistance.

Now for each cap short the cap terminals with a screwdriver or jumper wire to remove any residual charge. Then put one DVM probe on each terminal and watch the display.

A good cap will show near 0 ohms the instant the probes make contact. The reading will then immediately climb gradually and smoothly into the megohm range over the manner of a few seconds. (If you wish to repeat the test, be sure to short the cap terminals first to discharge it).

A bad cap will either show constant 0 ohms (internal short) or infinite resistance (open).

Note! All this test looks for is shorts/opens. It does NOT measure whether the cap's labeled capacity is still there!

---

Incidentally this test works with an analog VOM too – use the R x 10k range and observe meter needle motion.

 

[CH(upstate,NY)]

Current from the meter. Remember, the meter puts out a set voltage depending on ohm range and is measuring current flow.

 

[John T]

CH has ya covered, the VOM will only charge them up to maybe 1.5 or 9 volts depending on its battery. If in a operating circuit they charge up to whatever voltage is applied. Its NOT charging them up you really need to be worried with, its just first discharging them then seeing how well they accept a charge of current and that they arent shorted youre interested in.

John T

 

[Anonymous-0]

I "measured" a motor start capacitor once with my Simpson 160 (miniature 260) within 20% of its nameplate value.

I did like you said, shorted and watched the ohmmeter reading rise. In the analog ohmmeter the mid scale (inline with half scale on the voltage scale) ohmmeter reading is the series resistance of the ohmmeter. Likely 10 ohms on Rx1, 100,000 ohms on the R x 10K. Then after shorting the capacitor I timed the meter movement and when stopped timing when the meter passed 63% of full scale on the voltage scale. So I checked one time constant of the Resistance of the meter times the capacitance. T = RxC. T / R = C. So on that test if I was on the R X 10,000 with R = 100,000 ohms and T was four seconds, C would be 0.000040 F or 40 microfarads.

Now on the welder capacitor, a good capacitor ought to take 12 seconds to 63% voltage with a 1000 ohm charging resistor like the meter on R x 100. Electrolytic capacitors have a wide tolerance, far wider than my inaccuracy counting seconds, "thousand and 1"

Gerald J.

 

[gene bender]

Why wont the famous Snap=on waranty take care of the problem?

 

[glennster]

gene, its an older snap on muscle mig. its experiencing a power roll off while welding. the manual has some test proceedures in it, and i went thru them, the diodes checked out fine, so i figured it could be the caps. but its not that either. looks like i'll have to send this one in to the shop for repairs.

 

[glennster]

here is a link to the welder, its on the spendy side at 3g. 100% duty cycle, and does spot, stitch and continious weld
spark here

 

[MarkB_MI]

Capacitors almost always fail catastrophically. In the case of big electrolytic capacitors like in your welder, they generally 'splode and you can tell that they're bad by looking at them. I doubt that your heat problem is caused by a bad capacitor. They only smooth out the voltage to reduce the 120 Hz ripple you get with full-wave rectified AC. You could probably completely disconnect them and not affect the operation of the welder. The value of a capacitor will change over time, but it's just not that big of a deal for a welder.

There should be a thermal overload switch somewhere in the unit, it might be acting up.

If you can monitor the voltage at different points in the welder while you have a helper weld, you should be able to see where the voltage is dropping off. Remember that you have AC coming into the diode bridge and DC coming out.

 

[buickanddeere]

Lacking a proper cap tester.
An analog ohm meter works as good as anything for a go/no go test. Then megger for high voltage testing.

 

[Anonymous-0]

Hi Glenn - I fix x-ray equipment and all of these capacitor tests have technical merit. I saw in one of your replies that you have determined that the caps are O.K. If you are getting an intermittent arc (snap) and it is an older piece of equipment look carefully at the internal wiring harness, especially if the are any high current connections on a molex type plug somewhere. Try to include the wiring harness in the cap test (point A to point B, etc) to see if you have an intermittent open somewhere. We run into this type of failure from time to time, especially on older equipment.

 

[Wilson]

john really been worried that you might develop a hernia fron lugging that heavy banjo Hey buddy how's the sunny south?????????

 

[bc]

As Mark said, I'd look for a bad thermistor or dirt and dust on it keeping it from being cooled. Designed to open up at a certain temperature and break the circuit. It will carry dc voltage till it opens. Jumper across it and see if welder keeps going. Spray it with some component freeze spray, aka propane to see if that closes it back up. Component freeze spray was the secret for spraying around printed circuits boards looking a component that was open. Freezing the component made it shrink and make contact inside until it warmed up again.

If nothing else take the machine outside in the cold or get some dry plastic baggies full of ice or snow and lay on the components.

 

[gene bender]

Those things are fine but when thr problems arise then the money starts. I decided to go with MILLER as the brand x mite be cheap but who is going to fix it. I know some of the big factorys have their own repair service as they just have too many to be sending out for service also some welding services do a good job in repairing them.

 

[Anonymous-0]

Capacitors increase the strike voltage and you will notice if they are disconnected.Big capacitors can lose capacity over time.DC welding arc is much smoother than ac.Testing caps with a 1.5 volt battery is foolish.Capacitors should be tested at their working voltage for leakage.