Maker Pro
Maker Pro

LC meter accuracy

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
It's my opinion that any cap, inductance, or R instruments intended to read very low values cannot have test leads and or alligator clips. They introduce far to much error when reading low values. Low values require direct and solid connections to absolutely stationary test points. The fact that some meters allow for zeroing doesn't really negate this issue.

Chris
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
It's my opinion that any cap, inductance, or R instruments intended to read very low values cannot have test leads and or alligator clips.

A very good point.

Mine has short leads (5cm) with alligator clips. This effectively limits the readings at the lower end of the capacitance range.

This is the device I have (however mine now sits inside a box :))

Note the minimum capacitance reading vs. precision. I guess this is to maintain your confidence by hiding the very low readings when the device is zeroed but disconnected.
 
Last edited:

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Wow Steve, I'm swallowing hard accepting their stated inductance range! Can it really measure down to .001uH??? Even with short leads as shown I'm very sceptical

Chris

Edit: Yeah, I see what you're saying now. They're also claiming .01pF!!!
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
I've never tried the nH readings, the lowest I've gone is a couple of uH.

The resolution is 0.001 uH or pF, and it does tend to wander around and I don't especially trust it.

However when testing small capacitors (I think 2p2 was the smallest I tried), the results tend to be highly repeatable and within an expected range. Keeping your hands away is a good idea at the low end :D

I poke the zero button with a stick.

If you're interested I can run some tests and give you some slightly more real figures.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Sure I'm interested. It isn't every day that I see specs that resolve down to .01pF and .001uH.

Chris
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
Heh. Precision vs accuracy. It might only be accurate to +/- 1 pF.

But yeah, I'll try to run some tests later today.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
Here's some results (I was given other duties :))

Firstly, the meter doesn't read anything under 0.1pF. I think that's a way of hiding the small readings when nothing is connected.

I find that mine reads 0.15pF or thereabouts even when zeroed. If I lift it above the bench, it reads 0.00 (therefore something less than 0.10).

Moving my hands cose to the probes changes the readings by up to 2 or 3 pF. Changing the position of the short test cables can change the reading by up to maybe 0.5pF.

The zeroing function clearly DOES NOT zero out the lead capacitance!!! If you zero it with a small capacitor connected, the meter reads pretty much the same as if you zero the meter then connect the cap.

The capacitors I used for the first test were five 0805 1.8pF caps connected in series using short lengths of wire wrap wire.

Here are some readings of these capacitors. There were 3 readings of each of the 5 caps.

Code:
                   2.17         2.43         2.09
                   2.36         2.31         2.06
                   2.36         2.25         2.03
                   2.34         2.12         1.89
                   2.29         2.16         1.92
The meter clearly reads high. The last colum contains tests were care was taken to keep the test leads separated as much as possible.

The capacitors were connected in series. Reading from end to end, the capacitors measured 0.28pF and 0.32pF (Thae calculated vaue would be about 0.36pF). These measurements required the leads to be much further apart than usual. If I ben the string of caps to allow the test leads to be as cose together as when I was reading a single cap, the value rose to 0.48pF.

During all of these tests, the value in the last digit varied by +/-1 or 2 digits (more if the leads could move)

I then grabbed a strip of 1nF 440VAC capacitors. these were also measured 3 times (results in pF

Code:
                   933         934         935
                   901         894         897
                   930         925         928
                   966         958         954
                   938         938         935
The results show it reading a ittle low, but interestingly the readings were rising. If I shorted out the cap, the value returned to a lower value (all of these were shortly after the shorted cap was connected).

Connecting a singe 0.1uF 275V X2 cap resulted in a reading of 100.2 or 100.3nF.

Lest you think I have very accurate capacitors, some 47nF caps read as 42.31nF and 42.48nF. The vaues vary by up to +/-5 in the last digit. My "good" meter reads these as 43.6nF and 43.9nF.

A dedicated capacitance meter reads 44.3nF and 44.7nF, 1.7pF, and 99.3n for the various capacitors.

I'm actually quite happy that all the meters do a pretty similar job
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Steve, thanks for taking the time to run those test. Believe it or not I still don't own a digital C or L meter. I think I once purchased a small kit intended to be used with a DMM or AMM but I was sidetracked by something (it's always something) so it's buried somewhere in my storage bay.

I still have my Heathkit LCR bridge (gas eye) that I built when my hair was black. It wasn't and still isn't useful for measuring small values of L though.

I'm passing this along for all the young that take our current technology for granted...

When I was doing mostly RF work my Lx meter was a sweep gen that swept an LC tank built into a small die cast aluminum box with BNC in out connectors. A detector was also built into the box. C was known so it was just a matter of calculating Lx using the resonant freq formula. The sweep gen has a marker gen input so I adjusted the marker gen freq for a beat at resonance peak on my scope. I then plugged the freq into my HP25 where I stored the short program to find Lx.

Chris
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
I still have my Heathkit LCR bridge (gas eye) that I built when my hair was black. It wasn't and still isn't useful for measuring small values of L though.

I'm pretty sure this meter will have the same sort of trouble.

I rarely have needed to measure really low inductances, but a few tries using loosely coiled pieces of wire didn't read on the meter.

I may have more time today (when I am not cooking for some guests) to try out something a little more structured.
 

The Electrician

Jul 6, 2012
117
Joined
Jul 6, 2012
Messages
117
It's my opinion that any cap, inductance, or R instruments intended to read very low values cannot have test leads and or alligator clips. They introduce far to much error when reading low values. Low values require direct and solid connections to absolutely stationary test points. The fact that some meters allow for zeroing doesn't really negate this issue.
Chris

You're quite right. This works:

attachment.php
 

Attachments

  • Fixture.JPG
    Fixture.JPG
    28.1 KB · Views: 228

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
Here's an inductance measurement.

I have 4 turns of wire wrap wire with a diameter of about 4mm. The coil is about 6mm long. Oh, the wire is about 0.25mm in diameter.

This page calclates the inductance as 0.0324uH. My meter reads 0.065uH

Another coil which is 16 turns of the same wire with diameter 9mm and length 43mm.

My meter reads 0.659uH, the calculated value is 0.446

After resetting, the readings are 0.065uH and 0.646uH.

A 2.2uH inductor I have measures 2.010 uH
A 59uH inductor reads 46.75uH
A 100uH inductor reads 95.46 uH
a 1mH inductor reads 1.008mH

I think it's pretty clear that the precision exceeds the accuracy!

Interestingly, the meter foesn't read anything for the lower value inductances unless the inductr hangs out in free space. If it is resting on my bench, it reads 0.000
 

techiesteve

Jul 27, 2013
17
Joined
Jul 27, 2013
Messages
17
Steve, your test results make interesting reading, as I have the same LCR meter. At least it gives more accurate readings than my 15 year + old iso-tech LCR meter, that's more than useless at the low end, but then I bought it for measuring mH's.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
You're quite right. This works:

attachment.php

Yeah, like that! By the way, that was dirty pool! I spent over 20 min searching it and the instument that uses it, Then I spent another 20 or so min reading through this .pdf!

http://www.google.com/url?sa=t&rct=...5YCQDg&usg=AFQjCNFtAmOcw3KY5hLigwCdRQU2G5P89w

We need a "Grrr!" emoticon! :p

Here's an inductance measurement.

I have 4 turns of wire wrap wire with a diameter of about 4mm. The coil is about 6mm long. Oh, the wire is about 0.25mm in diameter.

This page calclates the inductance as 0.0324uH. My meter reads 0.065uH

Another coil which is 16 turns of the same wire with diameter 9mm and length 43mm.

My meter reads 0.659uH, the calculated value is 0.446

After resetting, the readings are 0.065uH and 0.646uH.

After resetting what?

Chris
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
So, I take it, the tests surpass your expectations.

Chris
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
I'm happy with its performance. I didn't buy it as a precision instrument and I'm not surprised that it gets a bit weird at very low values of inductance and capacitance.

As expected its stability gets better with increasing values, but I probably wouldn't rely on anything past the first 2 digits unless I was just trying to match parts.
 
Top