Shopping for LCR meter. What is the difference between 100 KHz and 1 KHz test ?

[Wartex]

I'm thinking of buying a $460 LCR meter from Tonghui.

http://cgi.ebay.ca/Portable-Handheld-LCR-Meter-100-100KHz-Mini-USB-TH2822C-/330553226395?pt=BI_Electrical_Equipment_Tools&hash=item4cf681f49b

Total noob here, what is the difference between 100 KHz and 10 KHz test mode? I understand capacitors and inductors will behave differently under different frequencies? Dave mentions that "100 KHz" is the industry standard yet there are tons of LCR that go into megahertz range. Also, the BK and Agilent he showed are 10 KHz max models. So, what's the difference and why should I care?

[nukie]

Different Henry/Farad values at different frequency. So in order to match your operating frequency of the circuit you are working with.

[vk6zgo]

Different Henry/Farad values at different frequency. So in order to match your operating frequency of the circuit you are working with.

Close ,but no cigar!

The lower frequency test is less likely to show up any stray reactances which may modify the effective value of the

component.

Also,if the thing tests for"Q",a lower frequency may give misleading results  for low values of L & C.

100KHz is a compromise,like many things in Electronics (& life),higher frequencies would be more appropriate if you

are doing a lot of RF work,but are more expensive to provide in inexpensive instruments.

VK6ZGO
 

[Psi]

i believe that's what he said.

[vk6zgo]

i believe that's what he said.

Not quite,it might be what he meant, but it read as if different instruments were necessary if the final product is to be used in different frequency ranges.
Although the lower frequency unit may  have some problems testing components for HF use,the HF unit will be perfectly satisfactory
 testing components for LF use.
My point was,that 100kHz is a compromise,& in the real world,we very seldom have the ideal instrument available.

VK6ZGO

[saturation]

In June 2011, Agilent will release newer LCR meters, I would wait to see these as they are offering basic accuracy of 0.2%, which you don't normally find in $290+ LCR meter, which is what Agilent announced it will list price for.  For almost $500, I would trust Malaysian Agilent builds living up to their spec sheet, than the Chinese model.

Here is some information on the new products:


    Enhanced LCR meters with 2x improved accuracy (0.2%), wider test range with up to 100kHz selectable test frequency
    Enable detailed component analysis: D, Q, ?, Z, ESR, DCR
    2x longer battery life (16 hours)
    Automatic component identification and measurements with Ai
    Connectivity to PC and data logging function

    Quick, basic LCR measurements start from $290

 

U1731C
   

$ 290
   

20000 Count Dual Display 120HZ/1KHZ Handheld LCR Meter with alligator clips, 9V alkaline battery, Certificate of Calibration, Quick Start Guide

U1732C
   

$ 350
   

20000 Count Dual Display 100HZ/120HZ/1KHZ/10KHZ Handheld LCR Meter with alligator clips, 9V alkaline battery, Certificate of Calibration, Quick Start Guide

U1733C
   

$ 400
   

20000 Count Dual Display 100HZ/120HZ/1KHZ/10KHZ/100KHZ Handheld LCR Meter with alligator clips, 9V alkaline battery, Certificate of Calibration, Quick Start Guide


On LCR meters:

I doubt any handheld LCR meter is going to give you a real world LCR reading much farther than the basic cheapo meter, consider that you can find an LCR meter for $30,000, what can a $500 or $160 meter do?  What's the difference?   At least one, the test frequencies of expensive LCR meters are swept continuously from Hz to GHz in top models, and inbetween for lesser priced ones, cheaper ones sample single frequencies among many.

The better LCR meters do not specify accuracy easily, because it can vary for each frequency.  The better ones give you a bode plot that represents the DUTs frequency response curve for impedance.  From there the meter calculates DF, Q, ESR etc.,

Characteristics can change a bit if the DUT model used assumes series or parallel parasitic LCR, associated with the DUT.  So, an LCR meter must be able to switch configuration between the 2 models, and calculate the changes.

My feeling, is you don't specify parameters enough, then the application really doesn't require anything but the cheapest LCR meter you can find to give you basic accuracy.  Any added calculation made by mid-priced LCR meters is not necessarily more accurate since the measured value on the meter may change with the actual frequency of your application, that is even 100kHz is as much just a sample and not  "real world" if your LCR applications are in the MHz+ range.

[tinhead]

Tonghui is producing LCR meter you can't really pay (up to 9000 USD). You can actually trust in their accuracy,
these THD2828C are really good value for money (A and B series are out of production already, so don't get fooled
and don't buy them).

Agilent is for sure better choice, but only if you need things like additional warranty or calibration support - this is
actually not given for "low cost" Tonghui models (only available for these expensive models).

[EEVblog]

Agilent are doing their best to get me one of these for review before the release date.
But basically, I believe it's just the B model with improved performance.
I will have both, so can do a comparison.

Dave.

[saturation]

Probably a simple way to see what these lower cost LCR meters do compared to the $30,000 ones is that the method used to measure LCR can provide less stable readings by their instruments electronics when measuring certain values of L or C at a specific frequency. 

B&K manuals are clearly written on theory of operation, but in a nutshell for large capacitance use low frequency such as 100Hz, and smaller, [ nF pF etc.,] use high frequencies, such a 1000Hz.  You can use either frequency to measure, just choose whichever is stabler, the accuracy is mostly the same for either frequency.  Because it can change a tiny amount depending on the frequency used, they label the whole meter as 'basic' accuracy, a kind of 'summary' of the total accuracy of the device.

So think of the various frequencies as a 'range' that best shows the LCR readout, like manual range on a  DMM.

So, in toto, the higher the frequency the more likely you'll be able to read small value L C better.

This is not necessarily the case for all designs, the best way to find out is get the spec sheet and see the limits of measurable L C for that meter.  Many can read much larger or smaller L C with just 100 and 1000 Hz alone, or thereabouts.

Handheld provide a more 'ideal' value of LCR.  It will presume the reactance is stable throughout its applied frequency range as defined by, say for capacitance, by its standard formula and so impedance is purely reactive:



However, in read world use, L C R are loaded with parasitics and its actual reactance will change at a unknown frequency, and now ESR and other non-ideal issues appear, so better LCR meters can test a range of frequencies rather than just a sample of them, the best ones sweep the frequency range, hopefully the upper range is the range of your application.

These issues occur mostly in the RF range and up.

So in summary, paying more for 100kHz frequency LCR meter is not better unless that meter allows you read smaller value LC than a meter limited to lower frequencies, be it 10kHz or 1 kHz.  It won't be more accurate, for example look at the accuracy of the Agilent meter offering despite the price range, it just has a potential of reading smaller ideal LCR values.

A word on what 'accurate' means.  The simplest way to compare meters is the capacitance % reading referenced at 1kHz, and can be NIST traceable. 

http://www.cenam.mx/memsimp06/Trabajos%20Aceptados%20para%20CD/Octubre%2026/Bloque%20D/D3-ELECTRICA%20IV/D3-1.pdf

With a reference capacitor at NIST, current studies show that various frequencies alter the measured capacitance value by less than 1 ppm.

http://rsi.aip.org/resource/1/rsinak/v74/i9/p4212_s1?isAuthorized=no


See also:

https://www.eevblog.com/forum/index.php?topic=3625.msg48674#msg48674

[Mechatrommer]

simply checkout the caps datasheet. the caps value itself will change with certain parameter. temperature, voltage, frequency etc. so thats bogus on the component level alone. discussing the measurement accuracy? duh! thats a headache!

[saturation]

You could, but only if the item is guaranteed to specification, if the spec sheet says typical, it can be off.  The actual device you use is often different due to subtle manufacturing  issues; its only a guide so you need to test each item individually if your design requires tight specifications.

simply checkout the caps datasheet. the caps value itself will change with certain parameter. temperature, voltage, frequency etc. so thats bogus on the component level alone. discussing the measurement accuracy? duh! thats a headache!

[Ronnie]

I would trust Malaysian Agilent builds living up to their spec sheet, than the Chinese model.

I agree with Saturation. Agilent products manufactured in Malaysia are high quality and user's manuals are nicely written.