Capacitor mfd had to use to high

[BNME8EZ]

I am probably all wet here and really don't know enough about it to speculate but I will anyway.

I see the action of the meter in response to the charge/discharge of the capacitor as being like a bell curve. All caps will follow that same curve just being a fraction or multiple of the smaller or larger ones. If that is the case then the time at each part of the curve would remain constant even though each part of the curve would be different. So as a result even though the time is different the average stays the same I guess a good analogy of what I am saying is a car on a race track. The amount of time on each curve and straight will be different the average time per lap would be very much the same. Then it is just a matter of adding a multiplier for scale.

[szw21]

It's probably possible to reduce those formulas into a simpler form good enough for this kind of measurement. So it is possible the Fluke formula is a general version of the cap charging formulae with approximations.

The issue with measuring caps using rise time is - not sure how accurate it is going to be and it will take a long time to measure higher value caps such as 5000 or 10,000uF. In general the application for the meter could be anything outside of measuring start and run caps.

These days I believe most meters use measurement ICs produced by various semiconductor manufacturers in the meter design. The product i worked on I mentioned above will be integrated into a custom IC and the end users of the product will not know anything about how the measurement is working under the hood.

Another architecture I've seen is a Wheatstone bridge type arrangement to measure capacitance but again I've not had first hand knowledge of designing or using something using this principle so I cannot speak to the accuracy of any meter using this as the basis for measurement.

[R600a]

It's probably possible to reduce those formulas into a simpler form good enough for this kind of measurement. So it is possible the Fluke formula is a general version of the cap charging formulae with approximations.

The issue with measuring caps using rise time is - not sure how accurate it is going to be and it will take a long time to measure higher value caps such as 5000 or 10,000uF. In general the application for the meter could be anything outside of measuring start and run caps.

These days I believe most meters use measurement ICs produced by various semiconductor manufacturers in the meter design. The product i worked on I mentioned above will be integrated into a custom IC and the end users of the product will not know anything about how the measurement is working under the hood.

Another architecture I've seen is a Wheatstone bridge type arrangement to measure capacitance but again I've not had first hand knowledge of designing or using something using this principle so I cannot speak to the accuracy of any meter using this as the basis for measurement.

I have noticed that fluke meters are lightning fast on capacitance but my Fieldpiece takes its time and thinks for a while before giving a final reading.
Any idea what the difference is?
I have noticed that my Fieldpiece always agrees with the factory test numbers on those few capacitors that have the actual as tested capacitance marked on them so I trust it.
Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

[szw21]

I have noticed that fluke meters are lightning fast on capacitance but my Fieldpiece takes its time and thinks for a while before giving a final reading.
Any idea what the difference is?
I have noticed that my Fieldpiece always agrees with the factory test numbers on those few capacitors that have the actual as tested capacitance marked on them so I trust it.
Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

Without knowing what methods they use it would be speculating - but if they are both using the frequency excitation method the longer you wait the more clock cycles you can collect and will give you a more accurate frequency measurement so the FP may be collecting more cycles and hence the delay. But of course they could be using entirely different methods.

A schematic of the meters and the associated parts will give some clues but I don't think they give schematics out. If you ever have the meters open to the circuit board level you can see what ICs you can spot and then Google the part numbers.

[BBeerme]

Doesn't look like you got an answer for this one. Parallel capacitors are easy, you just add the capacitance of the two capacitors. So 10 mfd and a 15 mfd in parallel would be a total of 25 mfd.

Here's an interesting relationship:
The formulas for calculating series and parallel capacitors is the opposite of resistors.

Simple example:
A) Parallel capacitors you add the two, and the result will be higher than the highest of the two. Series capacitors will be lower than the lowest one.

B) Parallel resistors will be lower than the lowest of the two. Series resistors will be higher of the two.

Thanks for the unequal capacitors wired in series calculator.

Do you have one for unequal capacitors wired in parallel ?

PHM
-------

[bdserv]

I am probably all wet here and really don't know enough about it to speculate but I will anyway.

I see the action of the meter in response to the charge/discharge of the capacitor as being like a bell curve. All caps will follow that same curve just being a fraction or multiple of the smaller or larger ones. If that is the case then the time at each part of the curve would remain constant even though each part of the curve would be different. So as a result even though the time is different the average stays the same I guess a good analogy of what I am saying is a car on a race track. The amount of time on each curve and straight will be different the average time per lap would be very much the same. Then it is just a matter of adding a multiplier for scale.

If you charge a capacitor through a resistor with a constant voltage source it will reach 63% of the applied voltage in one time constant which is equal to R x C. So, if you know R, V, and the time it took to get to 63% of the applied voltage then you know C.

Alternatively, C = (i * t)/V so if you charge the capacitor using a constant current source then there is no charge curve - the rise in voltage is linear because the current is constant. So, you can calculate C if you know the current level, time, and voltage rise. Some meters do use this method.

[rundawg]

Do you have one for unequal capacitors wired in parallel ?

Check your email. I sent you a pdf on this.

I don't post technical info in this OPEN forum.

[bdserv]

One more method which we used to use to measure very low inductance but could also be used to measure capacitance is a "pinger".

If you connect the unknown capacitor in parallel with a known inductor and then energize the pair with a single narrow rectangular voltage pulse they will resonate together and produce something similar to the waveform below. All you need to do is measure the base width of one cycle in that waveform and you can calculate the capacitance.

[BNME8EZ]

This reminds me: has anyone ever tried to connect a scope to a A/C to check for issues? I ask because my son hooks up scopes to cars all the time to find bad valves, plugs injectors, sensors, etc. I asked him about it once and he said that they have known good to use as a comparison, and examples of known problems, then over time you learn what different waveforms mean.

Just seems to me that we often hook up pressure transducers with the electronic gauges so maybe a scope would show similar in our trade like bad valves, winding going out, etc.

[szw21]

This reminds me: has anyone ever tried to connect a scope to a A/C to check for issues? I ask because my son hooks up scopes to cars all the time to find bad valves, plugs injectors, sensors, etc. I asked him about it once and he said that they have known good to use as a comparison, and examples of known problems, then over time you learn what different waveforms mean.

Just seems to me that we often hook up pressure transducers with the electronic gauges so maybe a scope would show similar in our trade like bad valves, winding going out, etc.

Yes in some situations it would be a useful tool. For example troubleshooting inverter driven condensers and ECMs - the scope can allow you to look at the inverter drive waveforms to see what could be going on. Also it could be helpful to look at the signals from communicating thermostats and even view mains waveforms to look for any distortions.

Cannot see how it would help troubleshoot valve issues since valves are mechanical. The scope will help visualize electrical signals. Not much value for dc unless you are looking for noise, spikes or other issues riding on the dc.

Regardless it would be a useful tool in some situations. I think you can pickup a handheld scope for $100-$200. The signals in HVAC, automotive etc. are pretty slow you don't need a fancy scope. The ones that cost a lot are for measuring high frequencies - the ones that are designed for use in the GHz range are pricey.

[R600a]

Yes in some situations it would be a useful tool. For example troubleshooting inverter driven condensers and ECMs - the scope can allow you to look at the inverter drive waveforms to see what could be going on. Also it could be helpful to look at the signals from communicating thermostats and even view mains waveforms to look for any distortions.

Cannot see how it would help troubleshoot valve issues since valves are mechanical. The scope will help visualize electrical signals. Not much value for dc unless you are looking for noise, spikes or other issues riding on the dc.

Regardless it would be a useful tool in some situations. I think you can pickup a handheld scope for $100-$200. The signals in HVAC, automotive etc. are pretty slow you don't need a fancy scope. The ones that cost a lot are for measuring high frequencies - the ones that are designed for use in the GHz range are pricey.

That brought back a memory from when I first got my amateur radio license. A local amateur who owned a repeater had a scope for FM signals and helped me adjust my modulation remotely by watching my signal coming in to his repeater with that scope and telling me what pot to adjust which way to get my audio up to the standard I think it was 5 MHz of modulation.
He said that scope or I think I had some fancier name than scope but I'm going to call it a scope for now was $20,000 when he bought it years ago when he was still working.

Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

[szw21]

That brought back a memory from when I first got my amateur radio license. A local amateur who owned a repeater had a scope for FM signals and helped me adjust my modulation remotely by watching my signal coming in to his repeater with that scope and telling me what pot to adjust which way to get my audio up to the standard I think it was 5 MHz of modulation.
He said that scope or I think I had some fancier name than scope but I'm going to call it a scope for now was $20,000 when he bought it years ago when he was still working.

Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

I did not realize you had a radio license. Not sure what he was using but there are more specialized instruments for specific applications like analyzing radio waveforms, looking at the modulation index etc. But yes some of this stuff is expensive. Some stuff I've worked with for designing satellite modems are in the $75K range for price.

[R600a]

I did not realize you had a radio license. Not sure what he was using but there are more specialized instruments for specific applications like analyzing radio waveforms, looking at the modulation index etc. But yes some of this stuff is expensive. Some stuff I've worked with for designing satellite modems are in the $75K range for price.

I've been thinking about it since I posted that and I think he called it a spectral analyzer. It's been a few years though so I may be remembering wrong.

Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

[szw21]

I've been thinking about it since I posted that and I think he called it a spectral analyzer. It's been a few years though so I may be remembering wrong.

Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

Yes it certainly could have been. It shows the signals in the frequency domain whereas a scope shows the signals in the time domain. The old spectrum analyzers were analog and worked by mixing the rf signals with a local swept oscillator. These days spectrum analyzers are digital and directly perform Fast Fourier Transforms on the signals and display the results.

[szw21]

This thread made me think about trying out a cheap portable small scope since some have asked about scope before. So I tried out a $170 one from Amazon.

Here are a few captures.

Reference 1KHz square wave used to calibrate the probes. There is also a FFT display for spectral analysis. Cannot see any means to measure frequency components - the spectrum looks correct sine a squarewave has odd harmonics. Don't see this is as being very useful other than to get a visual indication of the spectral distribution of the signal

[szw21]

120V supply.

[R600a]

That blue line the zero cross line?

Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

[szw21]

Taken at the condenser. Upper trace - start winding - lower trace - run winding.

[R600a]

Taken at the condenser. Upper trace - start winding - lower trace - run winding.

That would be really interesting to see a side by side comparison with several different capacitors.

Sent from the Okie state usin Tapatalk
"Is this before or after you fired the parts cannon at it?" - senior tech

[szw21]

24VAC at the contactor