Author Topic: do i really need an expensive LCR meter??  (Read 2311 times)

0 Members and 1 Guest are viewing this topic.

Offline e0ne199

  • Regular Contributor
  • *
  • Posts: 81
  • Country: id
do i really need an expensive LCR meter??
« on: September 27, 2020, 01:51:56 am »
hello everyone... i am sorry i am asking you all about this trivial (and probably stupid to some of you) question, because i am still very newbie about this.. forgive me if i am wrong  ;D
i have already had de-5000 that i purchased last year, but i have a question... does LCR meter like de-5000 cover every application involving capacitor and inductor?? if not, what kind of application is an expensive LCR meter suitable for?
any answer is appreciated  :popcorn:
 

Online blueskull

  • Supporter
  • ****
  • Posts: 14374
  • Country: cn
  • BA7LKP
Re: do i really need an expensive LCR meter??
« Reply #1 on: September 27, 2020, 02:00:46 am »
Depending on how extreme are you going with.

If you are after characterizing LCR at a bias current and/or voltage, then you need to build and calibrate-out a bias injection jig.

If you want it to be built for you, you need a much more expensive unit, like E4980A with some options -- and it only injects voltage, not current. To inject current, you need to invest more on third party jigs.

If you are after dynamic LCR at a certain excitation waveform, you need a (potentially high power) resonant tester, an those things are basically only built to orders. Expect high $4-digit to low $5-digit.

Also if you need to measure extremely low inductance or capacitance, you need a VNA, not an LCR meter, and some RF knowledge like impedance matching and transmission line & test jig calibration.

For me, the majority of the time I measure inductance is with a VNA since most of the time I work in the sub-nH to 10nH range, but that's a very niche field and most people would never need to do that.

For common everyday use, I have an LCR tweezers.
 

Offline coromonadalix

  • Super Contributor
  • ***
  • Posts: 2963
  • Country: ca
Re: do i really need an expensive LCR meter??
« Reply #2 on: September 27, 2020, 02:01:35 am »
normally a good lcr can or could do  serial or parallel testing,  but it will depend of the circuit in witch you will probe

up to 100khz frequency test is okay for  many scenarios

of course the use of kelvin probe help a lot

annnd he list goes on and on  like blueskull wrote
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 1384
  • Country: us
Re: do i really need an expensive LCR meter??
« Reply #3 on: September 27, 2020, 02:15:19 am »
The original post made me chuckle.  I have lots of LCR meters and which I use depends on the application.  Measuring a capacitor at 1000 Hz will give different results from measuring it at 100 MHz.  Accuracy is sometimes more important.  Leakage too, and ESR and ESL and self resonant frequency.  And that's just for capacitors.

Inductors are much more tricky.  Resistors not so much but still there are plenty of considerations that need attention.

Of course bias is another issue that I won't deal with here.

The short answer is that you need a measuring device that corresponds to the application.
 

Offline coromonadalix

  • Super Contributor
  • ***
  • Posts: 2963
  • Country: ca
Re: do i really need an expensive LCR meter??
« Reply #4 on: September 27, 2020, 02:43:11 am »
+1  :-+
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 2033
  • Country: us
  • Retired, now restoring antique test equipment
Re: do i really need an expensive LCR meter??
« Reply #5 on: September 27, 2020, 02:47:28 am »
The most important point is that raised by Bob91343.  The DE-5000 is good in the audio range, but for RF applications there are other instruments that work well above 100 kHz.
 

Offline e0ne199

  • Regular Contributor
  • *
  • Posts: 81
  • Country: id
Re: do i really need an expensive LCR meter??
« Reply #6 on: September 27, 2020, 10:59:35 am »
haha thank you so much for your explanation here  ;D
at least i get some enlightment about the importance of expensive LCR meter 😀
i mostly work on digital electronics, and perhaps analog on basic level... again thank you so much for your explanation here  ;D
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 2033
  • Country: us
  • Retired, now restoring antique test equipment
Re: do i really need an expensive LCR meter??
« Reply #7 on: September 27, 2020, 02:00:59 pm »
Things that go wrong with "digital" circuits are often "analog" in nature.  For example, if you need to measure the self-resonant frequency of a bypass capacitor, a high-frequency unit with very careful fixturing is required.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 12299
  • Country: us
  • DavidH
Re: do i really need an expensive LCR meter??
« Reply #8 on: September 28, 2020, 04:03:20 pm »
i have already had de-5000 that i purchased last year, but i have a question... does LCR meter like de-5000 cover every application involving capacitor and inductor?? if not, what kind of application is an expensive LCR meter suitable for?

Network analyzers give much more detailed information by essentially testing at every frequency.  I have had decoupling capacitors which tested good at low frequencies but were in fact bad.

LCR meters do not measure leakage or dielectric absorption in capacitors, or saturation or hysteresis in inductors, which are important in some applications.  Blueskull mentioned voltage coefficient of capacitance.

 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 1384
  • Country: us
Re: do i really need an expensive LCR meter??
« Reply #9 on: September 28, 2020, 05:18:27 pm »
This brings me to one of my favorite topics, the nanoVNA. It measures capacitance and inductance and resistance and much more over an amazing frequency range.  It showed me how ineffectual an electrolytic capacitor is at high frequency.  It showed me the effect of lead length on components of all sorts.

I have learned more from this little cheap device than from nearly any other test equipment I have used.
 
The following users thanked this post: cdev, 4cx10000, Jacon, Trader, rernexy

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 12986
  • Country: gb
    • Having fun doing more, with less
Re: do i really need an expensive LCR meter??
« Reply #10 on: September 28, 2020, 05:26:08 pm »
haha thank you so much for your explanation here  ;D
at least i get some enlightment about the importance of expensive LCR meter 😀
i mostly work on digital electronics, and perhaps analog on basic level... again thank you so much for your explanation here  ;D

You are actually working on analogue electronics. Exception: if you are working on photon-counting or femtoamp circuits :)

Whatever equipment you use, you have to understand how the measurement results do and do not relate to your circuit.

So, first understand and define what is important in your circuit, and second find a way of measuring that.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1682
  • Country: de
  • Testfield Technician
Re: do i really need an expensive LCR meter??
« Reply #11 on: September 28, 2020, 08:16:28 pm »
Hi,
Quote
Inductors are much more tricky.

For inductance measure we use three methods, depending on the DUT.

Generally we use a GenRad LCR/Keysight LC1733C, for the most situations.(The DE-5000 matches their results very close/identical, so I´m happy to have one for private use).
When the airgap of a coil must be adjusted, we use a parallel resonant circuit with known capacitor.
For powercoils we use a selfmade device which can deliver pulses up to 200A, calculating the inductance by an µC.( so was the intention, but most we´re using a scope parallel, because of the ringing)



Online wizard69

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: us
Re: do i really need an expensive LCR meter??
« Reply #12 on: September 29, 2020, 06:55:05 pm »
Bob above probably gave the best answer.    I look it this way, everybody should have at least one good hand portable multimeter but that doesn't mean you will not want to have a good bench meter at some point in time.  Even on the bench having more than one bench meter may make sense as they are not all the same and sometime you just need to monitor more than one node.   

So a LCR meter like the DE-5000 is a good start.   At some point in your knowledge expansion and challenges you will realize that you need something different and even have a good idea what the features will be on such a device.

Think of it this way; if you buy a 100 MHz scope and at a latter date pick up a wiz bang 1Ghz scope, that 1GHz scope doesn't make the old scope useless.   On the contrary you might reach for that 100MHz scope more often as a general purpose unit.   It really depends upon your needs which by the way can change over time or for some stay the same.   If you are focused on audio equipment for example you might setup a repair / build bench and never have a need for new equipment for years.

hello everyone... i am sorry i am asking you all about this trivial (and probably stupid to some of you) question, because i am still very newbie about this.. forgive me if i am wrong  ;D
i have already had de-5000 that i purchased last year, but i have a question... does LCR meter like de-5000 cover every application involving capacitor and inductor?? if not, what kind of application is an expensive LCR meter suitable for?
any answer is appreciated  :popcorn:
 

Offline 13hm13

  • Regular Contributor
  • *
  • Posts: 200
  • Country: de
Re: do i really need an expensive LCR meter??
« Reply #13 on: December 17, 2020, 05:05:55 pm »
Not sure what the OP means by "expensive".
My meter for anything LCR is one of several $20 "M Tester" types that are found in several threads here on EEVBlog.

But I want something more reliable (= repeatable), ergonomic and accurate.


===========================
I was thinking something like this $100 USD Japanese model...
DE-5000 Handheld LCR Meter


Product description
1. 100 / 120 / 1k / 10k / 100k Hz test frequency
2. Dual 19999/1999 display
3. Backlight
4. Ls / Lp / Cs / Cp / Rs / Rp / dissipation factor (D) / quality factor (Q) /phase angle/ ESR LCR auto selection
5. Auto Ranging mode
6. Auto Serial / Parallel mode
7. Components sorting function
*Selectable tolerance ±0.25%, ±0.5%, ±1%, ±2%, ±5%, ±10%, ±20%, -20% +80%
8. Low battery indication
====================

Or splurge a bit ($366.00 USD) on...


Schweinhund---du bist bored genug to look at mein signature --- Dummkopf!!!!!!!!!!
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 1384
  • Country: us
Re: do i really need an expensive LCR meter??
« Reply #14 on: December 17, 2020, 07:51:21 pm »
The trouble with the 5000 and so many others is that we are trying to measure something that isn't a constant.  Every component has parasitic components that confuse and ruin otherwise simple measurements.

Every device has its limitations.  I think that most of the time we just want to see if a part is good, and most gear determines that well enough.  But if you care to take it a step farther, you find yourself mired in the quicksand of physical principles.  Stray parameters are one thing.  But nonlinear components are another (all parts are nonlinear).

If you are into the intricacies of these parts, you won't find that too many instruments will tell you what you want to know.  Self resonance, dielectric absorption, magnetic saturation, distributed capacitance, reflected energy, and many more parameters will confuse the best of instruments and those using them.

The worst thing a device manufacturer can do is to keep silent about the method his unit uses to make measurements.  If they don't tell you, don't buy it.  Otherwise you will have no clue what to think when a measurement doesn't agree with what you think it should be, or what some other instrument indicates.  The most important information is the frequency at which the measurement is taken.  Or if it isn't at any particular frequency, such as charge rate, oscillator pulling, etc.
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 2033
  • Country: us
  • Retired, now restoring antique test equipment
Re: do i really need an expensive LCR meter??
« Reply #15 on: December 17, 2020, 09:38:14 pm »
The DE 5000 makes no misleading claims about its measurements.  It measures the complex impedance at each of a discrete set of frequencies, and the frequency is displayed.  Normally, this reads out as a capacitance (or inductance) and a series or parallel resistance, or with a D or Q value.  True, it doesn’t directly measure the self-resonant frequency, nor the dielectric absorption (I’m not aware of any instruments that directly measure the latter).
My only problem with the measurement is that if you measure an inductor at a selected frequency above self resonance, it does not indicate that the reactance has changed from inductive (positive) to capacitive (negative) and reports an incorrect inductance.  I wish it had a mode to read out the reactance (with sign) and resistance as separate parameters.
« Last Edit: December 17, 2020, 10:04:33 pm by TimFox »
 
The following users thanked this post: cdev

Offline Electro Fan

  • Super Contributor
  • ***
  • Posts: 2657
Re: do i really need an expensive LCR meter??
« Reply #16 on: December 17, 2020, 10:08:13 pm »
Different horses for different courses so you might eventually need more than one horse.

A DE-5000 would be a reasonable first horse.  For in circuit ESR testing a Peak Atlas ESR70 PLUS might be a good second horse.  And by all accounts a nanoVNA (some version, they are continuously on the move) might also be a great addition to the stable.

Several posts about LCR meters around, here are a couple:
https://www.eevblog.com/forum/testgear/what-lcr-meter-should-i-get/
https://www.eevblog.com/forum/testgear/der-ee-de-5000-vs-mesr-100-accuracy/25/
 
The following users thanked this post: Trader

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1682
  • Country: de
  • Testfield Technician
Re: do i really need an expensive LCR meter??
« Reply #17 on: December 17, 2020, 10:08:31 pm »
Quote
Different horses for different courses so you might eventually need more than one horse.

Exactly.

The DE-5000 could be the "main horse", as it got really nice performance at all and for the money let´s call it incredible.
At work, we got a GenRad Bench LCR and a Keysight handheld LCR, both yearly calibrated and the DE-5000 performs very well in comparsion to these ones(brought my DE-5000 to work and tested it).
So you can´t go wrong the DE-5000.
But we´re not only using these two ones, depending on the DUT we use a resonant circuitry with more testvoltage or our selfmade inductance-checker, which creates a pulse with high current ( up to 200A) and calculating the inductance according to the voltage-time-area.
So we use three possibilities to measure an inductive element.
The obstacle is, to know which method is the best suitable for the actual DUT.

Quote
I was thinking something like this $100 USD Japanese model...
DE-5000 Handheld LCR Meter

Bought it too from a japanese ebay-seller, in the same configuration.
At the end, I´ve paid appx 150€ including tax, never regret it so far.


Offline cdev

  • Super Contributor
  • ***
  • Posts: 6072
  • Country: 00
Re: do i really need an expensive LCR meter??
« Reply #18 on: December 17, 2020, 11:30:00 pm »
A VNA particularly the nanovna2 along with a test fixture(s) that allows you to measure series or shunt or shunt-thru impedances (can be very inexpensive, or home made) is a hell of a useful device for learning. Its not the same as an ESR meter, although it definitely could be used to measure ESR. (but not at anything near 60 Hz)

I have been working from this to understand this better.

https://www.eevblog.com/forum/rf-microwave/fun-with-crystal-filters/?action=dlattach;attach=996013

(its an article from MWRF entitled Accurate VNA Measurement mwrf_11053_21jrevise(1).pdf and it applies to all VNAs)

You need to make test jigs to do something like measure a component's complex impedance in something that approximates a real world setting. How you measure it depends on which kind of reactance is likely to dominate the situation.

The jig is more and more important to accuracy as you go up in frequency.

At HF it doesn't matter as much but at much higher frequencies it seems as if it would become very difficult to get accurate measurements without taking great pains to make sure the transmission line you're using was 50 ohms or whatever it is supposed to be. You could use some other dielectric material and copper tape, probably, also.

I'm sure there are any number of ways you could creatively likely make a jig that wouldn't require you to solder the component in place but its not as simple as one would expect not having some experience with a VNA. So the experience is absolutely invaluable for learning about RF.  On the other hand if you have no interest in radio stuff and dont plan to, or even if you do and are looking for a very useful test bench tool, get the ESR meter first. It would be a PITA using a VNA to try to measure ESR as a matter of routine. On the other hand, having a VNA is a lot of fun and the best thing for learning RF ever.
« Last Edit: December 17, 2020, 11:41:21 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Offline Electro Fan

  • Super Contributor
  • ***
  • Posts: 2657
Re: do i really need an expensive LCR meter??
« Reply #19 on: December 18, 2020, 03:59:34 am »
The DE 5000 makes no misleading claims about its measurements.  It measures the complex impedance at each of a discrete set of frequencies, and the frequency is displayed.  Normally, this reads out as a capacitance (or inductance) and a series or parallel resistance, or with a D or Q value.  True, it doesn’t directly measure the self-resonant frequency, nor the dielectric absorption (I’m not aware of any instruments that directly measure the latter).
My only problem with the measurement is that if you measure an inductor at a selected frequency above self resonance, it does not indicate that the reactance has changed from inductive (positive) to capacitive (negative) and reports an incorrect inductance.  I wish it had a mode to read out the reactance (with sign) and resistance as separate parameters.

What are the chances the positive/negative signage could be added to the DE5000 with a user installed firmware upgrade?  I’m guessing slim but if it was possible that would make the product even more popular.  Alternatively it would probably make for a popular second generation product. 

Seems like if someone slammed together a $100 DE5000, a $50-$100 nanoVNA, plus the in circuit ESR capabilities of the Peak Atlas that would add up to a very popular ~$299 product, maybe in a couple versions - one handheld and one a bench unit maybe with a bigger/better display.
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 2033
  • Country: us
  • Retired, now restoring antique test equipment
Re: do i really need an expensive LCR meter??
« Reply #20 on: December 18, 2020, 02:13:42 pm »
Since the LCR hardware essentially derives the real and imaginary components of the complex impedance, I assume that a firmware change could follow my suggestion.
I have an old Wayne-Kerr admittance bridge that is calibrated in capacitance and parallel conductance (at a single frequency) that can show positive and negative values of both:  negative capacitance results from measuring an inductance, and the frequency (1592 Hz = 10,000 / 2 pi) was chosen to use a table of reciprocals (in the manual) for ease of calculation.
LCR meters and network analyzers are quite different, so it probably is not a good idea to put them in the same box, but for high frequency measurements (up to several GHz), the network analyzer with appropriate test fixtures is the way to go.
 

Online Bud

  • Super Contributor
  • ***
  • Posts: 4589
  • Country: ca
Re: do i really need an expensive LCR meter??
« Reply #21 on: December 18, 2020, 03:59:50 pm »
While VNAs provide good accurate measurements they are not convenient for casual measurements of LCR because they require  user calibration. It takes several operations and will become tiring quickly. For LCR you'd want something capable of instant On.
Facebook-free life and Rigol-free shack.
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 1384
  • Country: us
Re: do i really need an expensive LCR meter??
« Reply #22 on: December 18, 2020, 07:35:28 pm »
Not so, Bud.  I have my nano set up so that all I do is switch it on and I can read component value.  If I want to concentrate on a particular frequency I can select that.  If I want more precise values I can calibrate but I have found that my calibration holds for a particular setup and I don't have to do it each time.
 

Offline cdev

  • Super Contributor
  • ***
  • Posts: 6072
  • Country: 00
Re: do i really need an expensive LCR meter??
« Reply #23 on: December 18, 2020, 08:47:29 pm »
It wouldn't be hard or expansive to build a nanoVNA2 into your bench with its screen and have it permanently connected to your computer and fixture and calibrated to come up with that as the default calibration plane and load, open short and through.

If you usually work say with smd components somebody made a test jig which they showed off in a thread that used pogo pins and maintained a 50 ohm impedance. Probably worked fine up to maybe 500 MHz.

I really wish that some manufacturer made something like veroboard but with the vias and breakable and a 50 ohm trace built in, cheap with the right price, that would make making fixtures a piece of cake.

It would basically be like little RF crackers you could break apart and make into things. 
« Last Edit: December 18, 2020, 08:50:42 pm by cdev »
"What the large print giveth, the small print taketh away."
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 2033
  • Country: us
  • Retired, now restoring antique test equipment
Re: do i really need an expensive LCR meter??
« Reply #24 on: December 18, 2020, 09:11:09 pm »
At my former employer, we had two -hp- units to measure impedance at high frequencies:  an -hp- 4191A impedance analyzer, and an 8753A network analyzer with reflection bridge.  Both used an APC-7 hermaphroditic connector at the test port, and needed to be calibrated with short, open, and 50 ohm load.
For surface-mount components (e.g. 1206 and 0805), at -hp-'s recommendation, we purchased a jig from CoilCraft that mated with the APC-7.
Their current version (for APC-3.5 connectors, the metrology grade of SMA) can be found at  https://www.coilcraft.com/en-us/other/smd-d/
It works for 0603 through 1812 devices.
You might want to emulate this design, since the jig is probably expensive.
 
 
The following users thanked this post: cdev


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf