Reference for LCR or ESR meters

[The Electrician]

A continuing problem for builders and buyers of LCR and ESR meters is how to check the performance and accuracy of the meter.

Precision capacitors are expensive.  An alternative is to have an ordinary capacitor that has been measured with a precision instrument.

I want to help.  I will measure some capacitors with a Wayne-Kerr 6440B and mail them to forum members.  The 6440B:

http://www.waynekerrtest.com/global/html/products/LCR/6430%206440.htm

is specified to have a basic accuracy of .02%, but that's marketing hype; it applies only to resistors.  For reactive components, the basic accuracy is .05% in the instrument's sweet spot.

I have a bunch of polystyrene caps that are old enough to have become reasonably stable.  Their temperature coefficient is good enough that a user at normal room temperature can probably count on .2% accuracy.  These would not be good ESR references because they're rather small--.01uF, but they would be excellent capacitance standards.  I can also supply some polycarbonates which also are reasonably stable, and larger at 2.2uF.

It might be useful to include an electrolytic or two with known ESR at room temp, although they wouldn't be very stable.

If anyone is interested, PM me.  I will bear the cost of mailing, so no one need send money.

I'm also interested to hear what thoughts people have about this.  I could also include some small inductors, but they would have ferromagnetic cores, and a .05% measurement of a not so stable component won't yield a .2% reference component.

[dannyf]

Having calibrated parts is helpful but not absolutely necessary to determine if a meter is accurate (sufficiently).

Let's say that you have a resistor of known value, and a capacitor of unknown value.

Put the capacitor on the meter and it should give a reading of C + ESR; Now, put the resistor in serial with the capacitor. You should have a reading of C + (ESR + R). Otherwise, the meter is faulty.

If you don't have a known resistor, pick a wire, and mark the middle point on the wire. Measure the resistance from one end point to the next end point, you should have a reading of R; measure the resistance from one end point to the middle, you should have a resistance of R/2.

Alternatively, you can combine the two techniques to see if a meter can measure small resistance on a capacitor.

[mrkev]

  Great idea

[SeanB]

Good idea, I like that as a reference component for checking, I generally use known values to check, often used components that have aged well. Resistors and good quality film capacitors generally age well and stay accurate if not run hot or at high voltage relative to max ratings.

[DL8RI]

Hi,

i have an ESI 2110 Video Bridge with basically the same specs.
Maybe it's interesting for you to cross-check.

And we could look at stability after shipping and so on...

[plesa]

I can made the measurement on my Agilent E4980, preffered only for EU countries.

[Co6aka]

But what about stability; temperature, humidity, mechanical stress, gravity, cosmic rays, alignment of the planets, government intervention, etc?

[robrenz]

Great idea

example here

[The Electrician]

I can made the measurement on my Agilent E4980, preffered only for EU countries.

I think it would be great if somebody like you could provide for the EU.  I can mail to the EU and it's not too expensive for just a few capacitors, but even better if mailing from there.

I ran some tests on the polystyrene caps I have.  I measured at room temperature (15C) and then put the cap in the refrigerator (0C) for a couple of hours.  The capacitance changed by .1% for a 15C decrease in temperature.  I don't think it would be unreasonable to expect one of these capacitors to be within .2% of the value I measure for a normal range of room temperatures.  Humidity won't be a problem with polystyrene and as far as mechanical shock goes, well, be gentle.

We're not trying for metrology lab precision here, after all!  Anybody who wants better can get some of these:

http://www.ebay.com/sch/i.html?_odkw=standard+capacitor&_osacat=92074&_from=R40&_trksid=p2045573.m570.l1313.TR0.TRC0.Xstandard+capacitor+genrad&_nkw=standard+capacitor+genrad&_sacat=92074

and pay for them!

[The Electrician]

I'll be sending out the first batch next week.  I've gotten 2 requests so far.

Something that would help me would be this--in your PM request, please type your mailing info, name and address, in exactly the form it should be on the package.  That way I can just print it, cut out the text and tape it on the package.

   Thanks

[BravoV]

I'm also interested to hear what thoughts people have about this.  I could also include some small inductors, but they would have ferromagnetic cores, and a .05% measurement of a not so stable component won't yield a .2% reference component.

Noob question, about tempco stability, does an air core inductor better than the one with ferromagnetic core ?

If it does, I'm thinking of using like one of these below that are easy to find, and use a hammer to whack out the ferrite part since they're brittle and easy to break, and will leave out only the bobbin with the winding but air core one.

[The Electrician]

I would think it would be more stable without the ferrite, but the inductance will be much lower.  There are magnetic materials that have been formulated to have a low tempco of permeability. 

[cellularmitosis]

Electrician, this is a great idea.  Have you thought about setting up a small store on tindie.com and offering a small assortment of calibrated capacitors as a product?  Might be the most convenient way to handle the ordering process, and also give you some more publicity!

[The Electrician]

I'm not interested in taking it that far.  I just want help those people on this forum who need a reference for checking their meters.

But, if somebody else wants to turn it into a small business, more power to them. 

[SeanB]

I think Frankie is starting that, or at least he is supplying check components for sale. Just ask him Iloveelectronics on the forum.

[plesa]

I can made the measurement on my Agilent E4980, preffered only for EU countries.

I think it would be great if somebody like you could provide for the EU.  I can mail to the EU and it's not too expensive for just a few capacitors, but even better if mailing from there.

I ran some tests on the polystyrene caps I have.  I measured at room temperature (15C) and then put the cap in the refrigerator (0C) for a couple of hours.  The capacitance changed by .1% for a 15C decrease in temperature.  I don't think it would be unreasonable to expect one of these capacitors to be within .2% of the value I measure for a normal range of room temperatures.  Humidity won't be a problem with polystyrene and as far as mechanical shock goes, well, be gentle.

We're not trying for metrology lab precision here, after all!  Anybody who wants better can get some of these:

http://www.ebay.com/sch/i.html?_odkw=standard+capacitor&_osacat=92074&_from=R40&_trksid=p2045573.m570.l1313.TR0.TRC0.Xstandard+capacitor+genrad&_nkw=standard+capacitor+genrad&_sacat=92074

and pay for them!

Send me PM if someone is interested. Also what about finding some standartd components kit e.g from Farnell and measure them? It will make it more consistent.

[Rufus]

It is worth remembering that most 'real' LCR meters measure everything ratiometrically against precision resistance standard(s).

There isn't anything to calibrate so there is no calibration to 'go off'.

Subtle measurement system faults which don't result in no or completely garbage measurements tend to just introduce noise.

[The Electrician]

It is worth remembering that most 'real' LCR meters measure everything ratiometrically against precision resistance standard(s).

There isn't anything to calibrate so there is no calibration to 'go off'.

Subtle measurement system faults which don't result in no or completely garbage measurements tend to just introduce noise.

Those with access to "real" LCR meters don't need the simple references I'm offering; I'm aiming at beginner/hobbyist folks. 

Now that the holiday season is over, I'm making the measurements needed.  I'll be mailing out the caps in the next few days.  I will probably include some inductors as well.

I'm going to include a polystyrene cap and a polycarbonate cap; I've measured the temperature coefficient of these two caps.  I'll also include an OSCON electrolytic.  The OSCON has much better low temperature performance than a regular aluminum electrolytic, so the variation in capacitance for a +- 15 degree C variation in measurement temperature is less than 1%.  However, I can't say how the OSCON stability with aging is going to be; that would be an experiment I could perform.

All the measurements have been made at an ambient temperature of about 15 degrees C.  If the user makes no compensation for his different ambient, an accuracy of about .2% (for the film caps) can be expected for an ambient up to perhaps 30 degrees.

[Rufus]

It is worth remembering that most 'real' LCR meters measure everything ratiometrically against precision resistance standard(s).

Those with access to "real" LCR meters don't need the simple references I'm offering; I'm aiming at beginner/hobbyist folks. 

By 'real' I meant not multimeters that measure capacitors by timing charging ramps or the variety of crude capacitor ESR meters. The cheap Mastech etc hand held LCR meters are 'real'.

[The Electrician]

It is worth remembering that most 'real' LCR meters measure everything ratiometrically against precision resistance standard(s).

Those with access to "real" LCR meters don't need the simple references I'm offering; I'm aiming at beginner/hobbyist folks. 

By 'real' I meant not multimeters that measure capacitors by timing charging ramps or the variety of crude capacitor ESR meters. The cheap Mastech etc hand held LCR meters are 'real'.

Thanks for the clarification; I thought you meant high end professional meters.  But those "real" meters, costing maybe a few hundred bucks, will have "precision" reference resistances that may not be as good as the ones found in the multi-tens-of-thousand dollar Agilent and other high end instruments.  I have no idea how good they may be; I've seen no comprehensive test data.  And there may be other error effects.  I think it's good to have around a known cap standard to verify that the phase sensitive detector is still working.  I haven't noticed a tear-down where the reference was pointed out, and its apparent quality was wondered about. 

Your point is well taken, though.  With that style of instrument one can do a fairly good check on calibration by just measuring a very high precision resistor.  I used to keep a .01% 1k ohm resistor around to check my early Digibridge.  The manual mentioned that there was no calibration adjustment.  Calibration consisted of checking accuracy; if it was off, the reference resistor was replaced.

Does anybody know if the owners manual for any of the new LCR meters based of the Cyrustek chip suggest a reasonable cal check would be to simply measure a high precision fixed resistor?

I'm going to do a long post on the deficiencies of meters that use a shortcut method that may fail under certain circumstances.  I'm hoping some of the people I'm sending these caps to will report their measurement experiences.  Then we may see that those meters don't always work so well.

[bingo600]

We're not trying for metrology lab precision here, after all!  Anybody who wants better can get some of these:

http://www.ebay.com/sch/i.html?_odkw=standard+capacitor&_osacat=92074&_from=R40&_trksid=p2045573.m570.l1313.TR0.TRC0.Xstandard+capacitor+genrad&_nkw=standard+capacitor+genrad&_sacat=92074

and pay for them!

Thank you for the nice offer(s)

I might send some "Red Wima" Caps to Sweden 

I was lucky to grab a HP 4440B decade cap , for 67€/88$

http://www.ebay.com/itm/Aglient-Hewlett-Packard-HP-4440B-Precision-Decade-Capacitor-40pF-1-2-F-0-25-/310825818187

Was told that it prob. was more precise than my Mastech 5308.

I goofed on the pics , as i forgot to calibrate the meter when using the Kelvin probe.

The result is better after calibration.

/Bingo

[BravoV]

I might send some "Red Wima" Caps to Sweden 

What is so special about this particular type of cap ? 

Compared to silver mica type ?

[bingo600]

I might send some "Red Wima" Caps to Sweden 

What is so special about this particular type of cap ? 

Compared to silver mica type ?

1: It's a very well behaved cap (for the price)
2: I have a lot of them 

/Bingo

[Jay_Diddy_B]

Hi group,

You can search for General Radio 1419 decade capacitor. The 1419K is mica and is a 0.5% decade box. The 1419A and B are polystyrene and are 1% The 1419-M are 1.5% on the higher decades and 1% on the lower decades.

The General Radio 1412BC is also 0.5% but these are more expensive.

Probably more expensive than the LCR meters 

Regards,

Jay_Diddy_B

[The Electrician]

I mailed the capacitors today.

I wrote the cap values directly on the caps themselves, where possible, so they don't get lost.

There is a list of 4 values; they are measured at 4 different frequencies, 100Hz, 1kHz, 10kHz and 100kHz, in that order.

The silvery, cylindrical capacitor is a polystyrene, the red rectangular cap is a polycarbonate, the one electrolytic is an OSCON cap.

There's a small inductor, about 27 mH.  This component varies quite a bit with temperature.  If I just hold it with my fingers, I can see the value drift as it warms up.

All these components were measured at a nominal 15 degrees C.  The polystyrene and polycarbonate caps have measured temperature coefficients of about -.008%/degree C, so you can compensate for your own ambient temperature.  I don't think it will be necessary.  Those caps have such low tempco, the error just from different room temps should typically be less than .2%.

The measured tempco of the OSCON electrolytic is about .3%/degree C.