EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: jheatac on March 18, 2020, 08:37:22 pm
-
Hello,
I am looking at testing some 15,000uf 45v capacitors from a welder. This is way outside the capacities of my LCR and multimeters.
I was going to use a signal generator, resistor and oscilloscope.
Most of the videos and directions for doing this are dealing with small pf and nf capacitors. Is there anything I will need to do differently for the larger capacitor?
-
Build a curve tracer and observe the reactance of the Cap
You only need a dual channel oscilloscope providing it has XY mode
-
Wire the capacitor being tested in series with a resistor R that goes to a SPDT switch. The switch should connect either to a voltage source, probably 30 or 40 VDC in your case, or ground.
Use a scope to watch the voltage on the capacitor and then switch from ground to the DC supply. A DSO in single sweep mode is best. Measure the time t in seconds it takes to charge from 0 volts up to 63% of the DC supply voltage.
Then the capacitance is C = t / R
You can choose a value of R to get a reasonable time t for whatever capacitor you are testing. If you don't see an exponential curve rising from 0 volts, it isn't acting like a capacitor and is likely bad.
-
Build an oscillator from your capacitor with an NE555, then measure the oscillation frequency.
https://duckduckgo.com/?q=ne555+capacitance+meter
If you're worried about leakage and dielectric absorption in the capacitor, then charge it to some 30V for a few minutes, then disconnect it from the power source and measure the voltage every few minutes with a DMM, or add an opamp as buffer for really high input impedance.
-
@jheatac are you trying to measure the Q (quality factor) to see if they are good or actually measure their capacitance? If its just capacitance, you can use a current source to measure the capacitance with your scope. Check out my post on Stack Exchange to see how to do this, where I was trying to measure 2000F super capacitors.
https://electronics.stackexchange.com/questions/387155/how-to-measure-very-large-capacitance-e-g-super-ultra-capacitors (https://electronics.stackexchange.com/questions/387155/how-to-measure-very-large-capacitance-e-g-super-ultra-capacitors)
Pictures of my test setup at the time.
https://imgur.com/a/gSO81IM (https://imgur.com/a/gSO81IM)
Instead of a DMM you can use your oscilloscope to get delta t and delta V. Then the capacitance is simply i*(Δt/ΔV).
-
If you want a quick and dirty method, set the PSU to 40V and 15mA current limit and connect it to a dead cap with DMM showing cap voltage and maybe another DMM showing current into the cap so you can fine tune the current limit if needed.
Use a stopwatch to figure out how long it's taking for the voltage to rise by 1 volt.
For 15mA and a 15000uF cap it should be 1v per second
0.015 A / 0.015000 F = 1 volt per second
This method doesn't work so well for smaller caps because you need such a low current limit and it tends to drift.
But for 15000uf a 15mA limit should be doable on most psus.
Be careful with the charged cap you end up with at the end :) 15000uF at 45V will give a decent bang :-DD