Low Cost Multimeters vs Voltage References

I have three meters I use a lot: UNI-T UT61E and UT136B and the Extech EX330. I measure them against the AD584 voltage reference, and also against the TL431 chip.

Datasheets:

Both of these chips are readily available from ebay sellers.

[UpcycleElectronics]

Check out the LM4040 voltage reference. It's available in several different voltages and is very cheap. The SMD version can be found on AliEx 10 for around $2. It's spec'd at +/-0.1% instead of the common TL431A @+/-1%
The best TL431 is rare but the "B" version of the IC is +/-0.5%.
The Analog Devices chip is very old and I think it is accurate to 0.6% (IICR).
The LM4040 is commonly used on the AVR Transistor Tester OSHW project for those that are familiar with the original project from the German forums.
-Jake

[ronniepirtlejr]

Did you know Unity-T UT 61E can be trimmed to your specific voltage reference by taking the back off?
Now this all depends on how much trust you have in your reference voltage source!
You can find info about how it's done on YouTube. That is (If you're not happy with it.)
The trimmer pot is in the top left side of the board. Extremely touchy!
The little Uni-T UT136B I'm not sure about?
Neither am I sure about your extech330 having a trim pot?

[Aga-Bubamara]

you are aware that 2.5, 5.5, 7.5, 10v are not exact.. as this particular chip is not trimmed and it is a lowest grade of this chip, , it can be anything between 2.4925 -2.550v and same ratio all along with other voltages... only when there is a calibration pot on the board, only then you can trust it's been calibrateted or trimmed to produce 2.5 v exact and so on... I made a few circuits myself using ad584b and ad586 which is 2ppp-5pp BR version of the chip.. plus that one is a J or H version which means 15ppm-20ppm .. what that means .. tha means that voltage will change for 20 parts per million for every degree celsius from some strating point... so you can not check you meters assuming it's 2.5 exact... as it's not... I have a 6.5 didgit bench multimeter and none of those chips produce a exact voltage unless trimmed and you stick with the same temperature at the time of calibration.. especially chines chips are away beyoned the specs.. chines chips are sold on ebay as those that are dicarded of the QC so PPM is away higer then it should be so the temperature coefficient too.. temperature can affect this chip for 5 degrees 100ppm

[yambo]

In addition to what Aga B says below, youre forced to deal with variables such as the calibration of the reference device used to calibrate your meters at the factory as well as how accurate that device was along with temperature and other variables. Same with your meters, how well have they maintained their accuracy etc etc. - but your exercise still shows youre getting more than enough accuracy for most average uses a hobbyist or DIYer would need. These little reference boards are really just ballpark devices but in using them ive found they are good for checking a new meters basic function and the more meters you check the more you get to know the boards characteristics and just how accurate it may be, Bottom line they are well worth having for the average DIYers hobby bench, if you need more keep looking.

[MartinPHE]

Don't forget your meters error% + digits, read your meter manual. I assume you warmed up the references to stability as well before testing and took into account the temperature/voltages specified by datasheets.

[pcs]

Why do people give thumbs down without saying why? Grr
Nice test Martyn
I had a VC97, and found it fine but swapped it for an Aneng AN8002 and find it better, it`s much smaller.
Dave Jones (EEVblog) tried his best to knock a similar Aneng device but had to admit for a hobbyist meter it was exceptional..
I suppose expensive professional meters will protect you much more but for low voltage/ low current applications it`s fine.

[pcfred]

You should check the batteries in the UT61E. It being your oldest meter they may be going a bit flat anymore? Low batteries can throw readings off sometimes too.

[mikeoliver]

you should look into the Aneng 8002. It is a nice 6000 count budget meter with great features.

[wegi]

2.500V 2.507V OK ? What with a voltage drop on the probes ? Doesn't exist ?

[DeeegerD]

Your voltage standard should have come with the measured voltages. 2.5/5/7.5/10 is not the actual voltage. (example: 2.5 might be 2.498 or something similar).