LCR Meter Accuracy

[y2khris]

Hi guys, I have been looking at LCR meters for a while now and it is one piece of equipment that I am having a hard time pulling the trigger on.  I have seen the DER EE DE-5000, the various hand held Agilent, and the full size Agilent (most likely HP as I have a somewhat limited budget) and I was wondering what is the best way to go in terms of accuracy, flexibility, data logging and longevity.  I have been using a function generator and oscilloscope to measure ESR and have made a calculator tool that let's me put in a few of the values and I get the ESR of the capacitor.

I figure the full size ones are more accurate because they offer a wider frequency range to test with but I am unsure of what is actually needed and what is the best bang for the buck (I know it's probably the Der EE, but I would rather not).

I have been looking at the following models:
HP 4262A (the 4261A only goes up to 1KHz)
Agilent U1731C (Seems kind of pricey for a handheld tool that only goes up to 1KHz)
Agilent U1733C (Goes up to 100KHz, but costs around $100 more than the U1731C)
HP 4274A (5 1/2  digits, versus the ones above that are 3.5, the 4275A goes up to 10MHz but only goes down to 10KHz)
HP 4263B (Accurate, nice display, modern but it costs an arm and a leg and I would have to save up for a while to get one)

I know somebody out there knows enough about this stuff to help me out. 

[nctnico]

I used to own a HP4274A but I sold it again because for it's size it couldn't do much and for measuring low resistance (milli-Ohms) it was useless. For now I'm using this one:
http://www.ebay.com/itm/Auto-LCR-Digital-Electric-Bridge-Resistance-Capacitance-Inductance-ESR-Meter-/171199475262

but it doesn't have data logging or remote control so even though it is pretty accurate it's use is very limited.

One that interests me very much is this one because it has a reasonable frequency span, can apply DC voltages/currents and it costs around $1200 including shipping:
http://www.ebay.com/itm/Hi-accuracy-0-05-LCR-Meter-50Hz-200KHz-14K-Frequency-Points-TFT-LCD-USB-RS232-/272298505240
It is from a brand called ZXP (http://www.zxptest.com/eng/ ) and with some digging the English manuals can be found for similar models. There are tons of similar LCR meters on Ebay from Applent and Tonghui; I get the feeling a lot gets copied.

[TurboTom]

At that price tag I wouldn't hesitate to get the Applent: http://www.ebay.com/itm/Applent-AT826-USB-Handheld-LCR-Meter-100kHz-LCR-Meter-Portable-LCR-Meter-New-/261672893309?hash=item3ceceb7b7d:g:dkgAAOxyaTxTU58C

I've got one of these (paid a little more than this) and didn't regret the purchase. Compact, reasonable frequency range and pretty accurate.

Cheers,
Thomas

[TheSteve]

You could contact this guy and see if he has anymore U1733C's - they are a nice product.
https://www.eevblog.com/forum/buysellwanted/fs-agilent-branded-u1733c-lcr-meters/

[Deckert]

Hi,

The problem (for me) is that an LCR meter is not something I use that often - I use my multimeters far more. But when you do need an LCR meter, boy, you really do need a decent one.

I made a video a while back, detailing my experience with the ISO-Tech LCR-1701 LCR Meter. I'm not advocating that you get it, especially not if you can source the DE-5000, but it uses the same chipset (from Cyrustek) as the DER EE DE-5000 meter, so you'll get an idea of what it's capable of. The DE-5000 is just not available in my country.

https://www.eevblog.com/forum/testgear/iso-tech-lcr-1701-teardown-and-review/

Most importantly, I've realised that I need a handheld device, not one tied to a bench. YMMV due to different requirements though. As a hobbyist, fixing the odd switch-mode supply and testing reclaimed caps, it's rather useful from time to time.

If you're mainly interested in measuring ESR, then have a look at the Peak Atlas ESR70. Steve Gardner over on SDGEE did a really great review of it (look for video SDGEE #010 on YT).

--deckert

[Dwaine]

I used to own a HP4274A but I sold it again because for it's size it couldn't do much and for measuring low resistance (milli-Ohms) it was useless. For now I'm using this one:
http://www.ebay.com/itm/Auto-LCR-Digital-Electric-Bridge-Resistance-Capacitance-Inductance-ESR-Meter-/171199475262
but it doesn't have data logging or remote control so even though it is pretty accurate it's use is very limited.

One that interests me very much is this one because it has a reasonable frequency span, can apply DC voltages/currents and it costs around $1200 including shipping:
http://www.ebay.com/itm/Hi-accuracy-0-05-LCR-Meter-50Hz-200KHz-14K-Frequency-Points-TFT-LCD-USB-RS232-/272298505240
It is from a brand called ZXP (http://www.zxptest.com/eng/ ) and with some digging the English manuals can be found for similar models. There are tons of similar LCR meters on Ebay from Applent and Tonghui; I get the feeling a lot gets copied.

How are you liking the XJW01 LCR meter?  I'm thinking about buying the XJW01 LCR meter, but it's a toss up between it and the Applent AT2817A or AT817D.

[y2khris]

I used to own a HP4274A but I sold it again because for it's size it couldn't do much and for measuring low resistance (milli-Ohms) it was useless. For now I'm using this one:

So, regarding the HP4274A, not taking the low resistance measurements into account, you got rid of it because it doesn't have many capabilities that new LCR meters have?

One that interests me very much is this one because it has a reasonable frequency span, can apply DC voltages/currents and it costs around $1200 including shipping:
http://www.ebay.com/itm/Hi-accuracy-0-05-LCR-Meter-50Hz-200KHz-14K-Frequency-Points-TFT-LCD-USB-RS232-/272298505240
It is from a brand called ZXP (http://www.zxptest.com/eng/ ) and with some digging the English manuals can be found for similar models. There are tons of similar LCR meters on Ebay from Applent and Tonghui; I get the feeling a lot gets copied.

Seeing as how they are similar in price, what do you think of the capabilities of the HP 4263B versus the ZXP?

[y2khris]

At that price tag I wouldn't hesitate to get the Applent: http://www.ebay.com/itm/Applent-AT826-USB-Handheld-LCR-Meter-100kHz-LCR-Meter-Portable-LCR-Meter-New-/261672893309?hash=item3ceceb7b7d:g:dkgAAOxyaTxTU58C

I've got one of these (paid a little more than this) and didn't regret the purchase. Compact, reasonable frequency range and pretty accurate.

Cheers,
Thomas

The screen looks really nice on the Applent; how does it compare to the EE 5000?  Is it more accurate than the aforementioned HP LCR meters, maybe with the exception of something expensive like the 4263A?

[nctnico]

I used to own a HP4274A but I sold it again because for it's size it couldn't do much and for measuring low resistance (milli-Ohms) it was useless. For now I'm using this one:

So, regarding the HP4274A, not taking the low resistance measurements into account, you got rid of it because it doesn't have many capabilities that new LCR meters have?

It is nice but the number of test frequencies is very limited.

One that interests me very much is this one because it has a reasonable frequency span, can apply DC voltages/currents and it costs around $1200 including shipping:
http://www.ebay.com/itm/Hi-accuracy-0-05-LCR-Meter-50Hz-200KHz-14K-Frequency-Points-TFT-LCD-USB-RS232-/272298505240
It is from a brand called ZXP (http://www.zxptest.com/eng/ ) and with some digging the English manuals can be found for similar models. There are tons of similar LCR meters on Ebay from Applent and Tonghui; I get the feeling a lot gets copied.

Seeing as how they are similar in price, what do you think of the capabilities of the HP 4263B versus the ZXP?

According to the specs (14000 different test frequencies) the ZXP would allow to make a fine grained plot of Q versus frequency of a component where the 4263B can only do a few standard frequencies. In my case the 100kHz of the 4263B would not be high enough. The 4263B is also missing DC bias abilities which are handy to check the voltage dependant behaviour of high capacity MLCC capacitors.

[pascal_sweden]

I have been using a function generator and oscilloscope to measure ESR and have made a calculator tool that let's me put in a few of the values and I get the ESR of the capacitor.

Can you elaborate more on the manual method using a signal generator and an oscilloscope?

How come that there is no software tool for Rigol or Siglent, that uses LXI to automatically control the LXI-enabled signal generator and the LXI-enabled oscilloscope in order to measure the ESR value and the Q factor?

[Deckert]

Hi,

Manual measurement using a scope and a sig-gen:


And measuring ESR:


Why use very expensive tools when you can use something that's a lot simpler, quicker more accurate and a lot more portable?
 

--deckert

[Vtile]

There is also application note for this age old Gen&scope method from IIRC tektronixs. Atleast for hobbyist use it might be suitable, I have seen a few methods of this. The one that finds 1/2 division frequency were simpler to use with low end gear .. IIRC.

This is basic voltage divider and calculation against known resistor.

http://www.tek.com/document/application-note/capacitance-and-inductance-measurements-using-oscilloscope-and-function-ge

[shteii01]

Hi,

Manual measurement using a scope and a sig-gen:


--deckert

Cool stuff.  Thank you for posting it.

But.  That dude got Tek 2467!  That was top of the line 'scope.  I would not want to do all that stuff on purely analog scope.

[drkirkby]

Hi guys, I have been looking at LCR meters for a while now and it is one piece of equipment that I am having a hard time pulling the trigger on.  I have seen the DER EE DE-5000, the various hand held Agilent, and the full size Agilent (most likely HP as I have a somewhat limited budget) and I was wondering what is the best way to go in terms of accuracy, flexibility, data logging and longevity.  I have been using a function generator and oscilloscope to measure ESR and have made a calculator tool that let's me put in a few of the values and I get the ESR of the capacitor.

I figure the full size ones are more accurate because they offer a wider frequency range to test with but I am unsure of what is actually needed and what is the best bang for the buck (I know it's probably the Der EE, but I would rather not).

I have been looking at the following models:
HP 4262A (the 4261A only goes up to 1KHz)
Agilent U1731C (Seems kind of pricey for a handheld tool that only goes up to 1KHz)
Agilent U1733C (Goes up to 100KHz, but costs around $100 more than the U1731C)
HP 4274A (5 1/2  digits, versus the ones above that are 3.5, the 4275A goes up to 10MHz but only goes down to 10KHz)
HP 4263B (Accurate, nice display, modern but it costs an arm and a leg and I would have to save up for a while to get one)

I know somebody out there knows enough about this stuff to help me out. 

Okay, I know this topic is a bit old, but the HP 4263A is a cheaper than the 4263B you found too expensive. The differences between the A and B models are given on the Keysight website and seem pretty minor, to be honest.

I don't know if this link will work - it loads very slowly for me:

https://www.keysight.com/main/editorial.jspx?cc=US&lc=eng&ckey=1498485&nid=-32776.536879695&id=1498485

If not, search for the text "What are primary differences between the 4263A and 4263B?" (blame Keysight for missing out the word "the" before "primary" - not me)

but the shows that:

1) The 4263B can select from 3, 4 or 5 digits, but the 4263A can only display 5 digits. That is a pretty insignificant feature! I'm sure we can round that if we want.
2) The 4263B has current and voltage monitoring, but the 4263A does not.
3) The 4263B has a "show settings" feature, to display all the settings, whereas the 4263A does not.
4) The 4263B can display the open/short/load compensation values, but the 4263A can not. But the 4263A still does the compensation.

None of those differences would put me off the 4263A. I would certainly choose it over the Chinese meters people are talking about. Note the Chinese meters are not supplied with coaxial cables, but the 4TP connection on the HP meters should use coax, and the coax has to be a specific length (1 m).

I picked up a used HP 4263A in August 2018, that had Agilent calibration stickers on the screws, but no cal sticker on the front. I rather suspect that means it had been back for calibration, but not in the last 5 years. I only did a quick test, but a 100 ohm 0.005% resistor read 100.00 ohms, with a few nH of inductance!

The annoying thing about the 4263A and 4263B is their limited choice of test frequencies. (100, 120, 1000, 10000, and 100,000 Hz), with 20 kHz as an option.

I have a 4284A (20 Hz to 1 MHz), which sell for a lot of money. That has over 8000 test frequencies, but they are very poorly chosen. There is 20 Hz, and 20.01 Hz, but nothing between 750 kHz and 1 MHz. So it has some frequencies so close together to be useless, and others so far apart.

Dave.

[David Hess]

It is difficult for me to recommend anything except the DER EE DE-5000 for casual use because of its low cost.  It covers the basics including measurements at 100kHz which are commonly specified.

If I wanted something better but at a significantly higher cost, I would look for a low frequency network analyser instead.

[drkirkby]

It is difficult for me to recommend anything except the DER EE DE-5000 for casual use because of its low cost.  It covers the basics including measurements at 100kHz which are commonly specified.

If I wanted something better but at a significantly higher cost, I would look for a low frequency network analyser instead.

The DE-5000 was a product made by the reputable company IET labs. I don't have a direct link, but here

https://www.ietlabs.com/de6000-lcr-meter.html

it mentions the DE-6000 is an improvement on the DE-5000.

Now there are other DE-5000's around, that will no doubt claim the same specifications as the IET Labs instruments, but I suspect are probably inferior.

A network analyzer and an LCR meter are very different sort of instruments. VNAs are not designed to work at very low or very high impedances - only impedances close to that of their bridge, which is normally 50 ohms.

I have here
* HP 8720D VNA (50 MHz to 20 GHz)
* HP 8753 VNA (300 kHz to 3 MHz)
* HP 4291B impedance/'material analyzer (1-1800 MHz)
* HP 4284A precision LCR meter (20 Hz to 1 MHz)
* HP 4263A LCR meter (100 Hz to 100 kHz). This is surplus to my requirements, as I only bought as it was sold with the Kelvin test leads.

For very low or very high impedance values, the LCR meters are best. For more modest impedance values, instruments based on the RF-IV technique, like the 4291B are more suitable and have a wider frequency range. For the widest frequency range, VNAs are best, but they are not suitable for low or high impedances.

Yes, there are tricks one can do with a VNA to handle higher or lower impedances, by not using a direct connection. But they will not approach the accuracy of an LCR meter.

The instrument I would like to get my hands on is an HP 4285A LCR meter (75 kHz to 30 MHz), but unfortunately, they are damn expensive.

Dave

[David Hess]

That is why I said "low frequency network analyser" instead of vector network analyser.  They make special ones for testing components.  Analog Devices even sells DDS ICs specifically intended for them.

It is one of those things which I have been tempted to design for myself and maybe make available.

[drkirkby]

That is why I said "low frequency network analyser" instead of vector network analyser.  They make special ones for testing components.  Analog Devices even sells DDS ICs specifically intended for them.

It is one of those things which I have been tempted to design for myself and maybe make available.

Can you give me an example of such an instrument bf one of the manufacturers of high-end test equipment, such as Anritsu, Rohde &Schwarz, IET labs, Keysight, Tektronix, Keithley etc? What impedance range do they cover, and what sort of frequency range?

Any such instrument must be able to measure the vector of the impedance.

I've personally never seen anything commercially available that can cover such a wide frequency range as the LCR meters, but it is not something I have looked for to be honest.

I just performed a quick literature review and did not manage to find much, but I'm sure there must be some "review papers" on the subject of impedance measurements.


Dave

[nctnico]

VNAs are not designed to work at very low or very high impedances - only impedances close to that of their bridge, which is normally 50 ohms.

This statement isn't quite true. There are several ways a VNA can be build and the ones with a lower frequency range (tens of Hertz or lower) have several methods of measuring which allows to measure very low and very high impedances.

[ogden]

Can you give me an example of such an instrument bf one of the manufacturers of high-end test equipment, such as Anritsu, Rohde &Schwarz, IET labs, Keysight, Tektronix, Keithley etc? What impedance range do they cover, and what sort of frequency range?

Here you go:

https://www.omicron-lab.com/products/vector-network-analysis/bode-100/

[David Hess]

Can you give me an example of such an instrument bf one of the manufacturers of high-end test equipment, such as Anritsu, Rohde &Schwarz, IET labs, Keysight, Tektronix, Keithley etc? What impedance range do they cover, and what sort of frequency range?

Any such instrument must be able to measure the vector of the impedance.

I do not know about the others but HP/Agilent/Keysight has had them for a long time.  Today they are calling them "impedance analyzers".  Internally they are essentially a low frequency one port vector network analyser optimized for precision instead of frequency range.

https://www.keysight.com/en/pc-1000000382%3Aepsg%3Apgr/impedance-analyzers?nid=-33831.0&cc=US&lc=eng
https://www.keysight.com/en/pc-2461685/impedance-analyzers?nid=-32775.0.00&cc=CZ&lc=eng

Hioki has them also but I do not know anything about the company.

Another one I remember running across.

Analog Devices makes a dedicated DDS IC for this and even an evaluation kit.  I am a little surprise there is not a commercial produced based on the AD5933 but maybe I have just not found it yet.

I've personally never seen anything commercially available that can cover such a wide frequency range as the LCR meters, but it is not something I have looked for to be honest.

I just performed a quick literature review and did not manage to find much, but I'm sure there must be some "review papers" on the subject of impedance measurements.

I always have troubling finding them because nobody uses a common name to describe them.  Essentially they are the vector network analyser version of an LCR meter.

[ogden]

Internally they are essentially a low frequency one port vector network analyser

Incorrect. Fact that 1-port VNA and IV impedance analyzer both have 1 port, does not make them equal.

Further reading:

https://literature.cdn.keysight.com/litweb/pdf/5950-3000.pdf

Chapter "2.6 Difference between RF I-V and network analysis measurement methods"

[bugi]

Analog Devices makes a dedicated DDS IC for this and even an evaluation kit.  I am a little surprise there is not a commercial produced based on the AD5933 but maybe I have just not found it yet.

May have something to do with "Capable of measuring of 100 Ω to 1 kΩ with additional circuitry" (and only 1k+ as is) (and note, it is about impedance, not just resistance). It seems to get quite tricky in that 100 to 1k range, and I don't think there would be many customers for a general purpose measurement device that does not go below 100 ohm.
That chip seems to be aimed at special purpose devices, where its features and limitations have a better fit for the purpose, and where it is possible to tweak the circuitry to match the specific needs.

It is difficult for me to recommend anything except the DER EE DE-5000 for casual use because of its low cost.  It covers the basics including measurements at 100kHz which are commonly specified.

The DE-5000 was a product made by the reputable company IET labs. I don't have a direct link, but here

https://www.ietlabs.com/de6000-lcr-meter.html

it mentions the DE-6000 is an improvement on the DE-5000.

Now there are other DE-5000's around, that will no doubt claim the same specifications as the IET Labs instruments, but I suspect are probably inferior.

There are already some threads here having some talk on that DE-5000 (and 6000) stuff, and my poor memory has collected something along the lines that DER is now making the same thing, maybe it was the original manufacturer of it for IET or something, and is now selling them to sooo many different sellers. The devices seem to now come with varying changes in what components are left out and/or replaced for saving a cent or two. They seem to work ok (with some quirks) and more or less "to the specs" (at least once the user remembers to do the calibration process). Also, it has been said that the IET's DE-6000 "improvements" were quite insignificant, and definitely not worth the price difference to DER's DE-5000. Of course, now IET part of the story is just a historical note, since IET discontinued both models (and "replaced" 6000 with a huge bench box with price as "request a quote", which typically means too expensive to be shown publicly...).

[Wolfgang]

... there is a Keysight E5061B-3L5 with a range from 5Hz (!) to 3GHz.
This seems to be the holy grail of VNA impedance measurements, and there are a load of Keysight appnotes how to measure very high or very low impedances.
Problem: Severe wallet drain.

[David Hess]

Analog Devices makes a dedicated DDS IC for this and even an evaluation kit.  I am a little surprise there is not a commercial produced based on the AD5933 but maybe I have just not found it yet.

May have something to do with "Capable of measuring of 100 Ω to 1 kΩ with additional circuitry" (and only 1k+ as is) (and note, it is about impedance, not just resistance). It seems to get quite tricky in that 100 to 1k range, and I don't think there would be many customers for a general purpose measurement device that does not go below 100 ohm.

The application notes show that the measurement range can be extended even beyond that but even with those limitations, it would be useful where common LCR meters are not.  I suspect though that anybody who knew how to make best use of it does not need it and could just make their own design without it from scratch which is certainly what I would consider doing.

That chip seems to be aimed at special purpose devices, where its features and limitations have a better fit for the purpose, and where it is possible to tweak the circuitry to match the specific needs.

I suspect the only reason the part is economical to produce is that it has a wide applicability to specialized transducer measurements where it can operate over a limited range of frequency and impedance.