find out value of pF capacitors

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
hi there

I'm looking for a way to calculate value of capacitors of pF(pico Farad) range. the thing is when I use a RLC meter, looks like the device is limited to high ranges of pF to nF and so I can not measure value of for example 1pF...
I used this technique
Circuit.JPG
using a microcontroller, I generate 8MHz pulse and pass it through first: a Lowpass filter and then: a peak detector
C1 is the value I'm trying to measure and it is around 1pF to 20pF(I do not knoe the right value, using this circuit I'm trying to measure it).
I've wired the circuit on a breadboard
then I use the equation Vo=Vs(1-exp(-t/tau))
Vs=3.3(STM32F103)
t=(here)62.5*10^-9
tau=R1*C1 and R1=4100ohm

I use a multimeter at the end of circuit where you can see "equivalent voltage"(I use a multimeter because it seems that it has no effect on my circuit for changing values due to very large input resistance ect.)

there are some issues:
1-it seems like the breadboard itself has some pF capacitor adding to my measurements(even for example if I just insert a piece of wire in a hole of breadboard)
2-it seems like the frequency also effects the value of output voltage and so the calculated value of C1.

can anyone help me with suggestions?
 

BobTPH

Joined Jun 5, 2013
5,227
You answered your own question right in your post. The problem with measuring such small capacitors is that the test leads themselves will have comparable capacitance. A breadboard will have capacitance that swamps a 1pF capacitor.

Bob
 

DickCappels

Joined Aug 21, 2008
8,812
To add to @BobTPH 's diagnosis, you can still get a pretty good reading by laying the test leads out and not quite touching the capacitor you intend to measure. Note the capacitance and then make the connection and note the change in capacitance. You might have to repeat this many times and then find the average change in capacitance to reduce "random" errors.

Remember to keep your hands away from the test leads and the component under test.
 

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
You answered your own question right in your post. The problem with measuring such small capacitors is that the test leads themselves will have comparable capacitance. A breadboard will have capacitance that swamps a 1pF capacitor.

Bob
I know, but how can I figure out the value anyway?
 

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
To add to @BobTPH 's diagnosis, you can still get a pretty good reading by laying the test leads out and not quite touching the capacitor you intend to measure. Note the capacitance and then make the connection and note the change in capacitance. You might have to repeat this many times and then find the average change in capacitance to reduce "random" errors.

Remember to keep your hands away from the test leads and the component under test.
sorry, but I didn't get what you meant by "test leads"? the capacitor is directly wired using no extra wires, jumpers, ect.
 

SamR

Joined Mar 19, 2019
4,292
Anything under 5-10pF is a crapshoot even doing as Dick says and using the relative button to zero the meter. That usually gets me within 1-2pF. But then, just why do you really need less than 5pF accuracy? How accurate is your meters specification? Did you measure your device temperature and compensate? Did you let your meter "warm up" to a steady state? So many variables...
 
Last edited:

DickCappels

Joined Aug 21, 2008
8,812
sorry, but I didn't get what you meant by "test leads"? the capacitor is directly wired using no extra wires, jumpers, ect.
Many LCR meters have wire leads with which to connect the component under test. My old B&K meter has both test leads and "slots" into which the component leads can be poked directly. I bought it when surface mount circuits were still rare and exotic.

It might help to cut the capacitor leads to the length they will be while in use. Of course this only applies to leaded capacitors.
 

SamR

Joined Mar 19, 2019
4,292
Just for grins and giggles I grabbed a no-name chinesium 5pF ceramic capacitor from stock. Gave all my meters a 30-minute warm-up and zeroed them. Move my hands away and noted the zero shift. Then connected the nominal 5pF cap and measured it. Here were the results...
IMG_1004[1].JPG
Throwing out the Siglent reading (2nF range @1-120%) the best I can say is it is ~6pF.

Edit: Yes, the 2 LCR meters have insertion slots instead of leads, much better! Although I do zero the bench meters with the leads attached and then note the drift when my hands are removed from the area of the meter and leads.
 
Last edited:

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
Anything under 5-10pF is a crapshoot even doing as Dick says and using the relative button to zero the meter. That usually gets me within 1-2pF. But then, just why do you really need less than 5pF accuracy? How accurate is your meters specification? Did you measure your device temperature and compensate? Did you let your meter "warm up" to a steady state? So many variables...
You're right, I should take temparature into count, simply I do not know the initial value of capacitor to use the equation
C2=(C1+kt)
Yeah, there exists soooo many variables. This means that I should precise my conditions more strictly. Can you give me some advice?
 

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
Many LCR meters have wire leads with which to connect the component under test. My old B&K meter has both test leads and "slots" into which the component leads can be poked directly. I bought it when surface mount circuits were still rare and exotic.

It might help to cut the capacitor leads to the length they will be while in use. Of course this only applies to leaded capacitors.
Your,re right and it sounds good to use the slots.
My test condition is that finally I'll make capasitor using PCB tracks l so at the end I wont have any access to metes etc. And I should measure it using a microcontroller. Any suggestions?
 

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
Just for grins and giggles I grabbed a no-name chinesium 5pF ceramic capacitor from stock. Gave all my meters a 30-minute warm-up and zeroed them. Move my hands away and noted the zero shift. Then connected the nominal 5pF cap and measured it. Here were the results...
View attachment 270076
Throwing out the Siglent reading (2nF range @1-120%) the best I can say is it is ~6pF.

Edit: Yes, the 2 LCR meters have insertion slots instead of leads, much better! Although I do zero the bench meters with the leads attached and then note the drift when my hands are removed from the area of the meter and leads.
Wow, You helped alot... I might do tge same
Thanks for spending time to help me out
 

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
I think he was simply trying to get someone to validate his procedure. As usual I got off on a tangent...
Yes I was really doubful as when I was changing the conditions the result were changing tooo, and it was becoming frustrating...
But your suggestions really opened my eyes. Thanks

PS.
Dear Sam, I'm she actually I know by default people mistake me with "he" and my incomplete profile might also influence
But anyways thanks aloooot
 

SamR

Joined Mar 19, 2019
4,292
Note: None of my meters, except the Extech, have any frequency adjustment. The default on the Extech (on turn-on) is 1kHz. It can do 100Hz, 120Hz, 1kHz, 10kHz, and 100kHz. To get to 8MHz I am afraid is going to take some serious money for a research-grade instrument.
 

DickCappels

Joined Aug 21, 2008
8,812
I still don't know what kind of accuracy you are shooting for, but I will go ahead and comment that if you can get your hands on a precision capacitor of about the same capacitance you expect, that should help you verify the calibration of your capacitance measurement method. If you are going to be using 10,000 pf or less, you can probably find some good precision NPO or COG capacitors with which to calibrate your instrument.

One thing that bothers me about your calculation is that I did not see the duty cycle of the pulse figured into it. That's important, you know ;-) While at it, the forward voltage drop in the diode will probably be very significant as may power supply variation, adding to the importance of calibration to a reference capacitor.

Edit:
I had to frequency compensate the feedback on a VHF closed loop amplifier and knew it was going to come out to about 0.5 pf but I did not want the signal to be over damped, so I had the PC designer lay out a "gimmick capacitor" -two traces side-by-side a little longer (higher capacitance) than I thought I would need. The placement of the capacitance with respect to the rest of the circuit was also very critical. When the first proto run of PCBs came in I carefully trimmed the lengths of the two parallel "plates" with an X-Acto knife until I had the waveform I wanted and informed the PCB designer of the correct length. Subsequent production runs gave entirely satisfactory results.
 
Last edited:

Thread Starter

AlaSH72

Joined Apr 3, 2022
30
I still don't know what kind of accuracy you are shooting for, but I will go ahead and comment that if you can get your hands on a precision capacitor of about the same capacitance you expect, that should help you verify the calibration of your capacitance measurement method. If you are going to be using 10,000 pf or less, you can probably find some good precision NPO or COG capacitors with which to calibrate your instrument.

One thing that bothers me about your calculation is that I did not see the duty cycle of the pulse figured into it. That's important, you know ;-) While at it, the forward voltage drop in the diode will probably be very significant as may power supply variation, adding to the importance of calibration to a reference capacitor.
you're right. the duty cycle is 50%. so you mean my calculations should go further like this:

Vc=Vs(1-exp(-t/tau))
Vo=(Vc*0.5)-0.5 (first 0.5 is for 50% duty cycle and the second is forward voltage drop of diode)
 
Top