Hi I've had a few people ask if I can do a video on how I maintain the calibration of my multimeters I've got here in the Eev blog lab now. uh I've mentioned before that I don't actually maintain these meters like I don't actually pay to get them formally calibrated or calibrated in quote marks with the calibration certificate and or adjustment or whatever because it's it's just not warranted uh for the uses I do. But I've got a couple of uh, little transfer standards that I can use to verify and maintain uh, the calibration of, you know, basic volts and ohms and amps of my multimeters. This won't be an in-depth tutorial as well I Just wanted to show off my little transfer standard here and my HP 3478a now uh I'll link in the video to this um EDC Electonic Development uh Corporation MV 106 uh DC voltage standard that I've got I I bought this for like 20 bucks on eBay It was an absolute bargain and I took it to the local Uh Calibration laboratory Trio Smartcow they're just uh down the road here and I'll link in the video to that and you'll see that this was actually absolutely spot on.

It didn't need any adjustment whatsoever and it's a six decade device so I can adjust it Uh well. anywhere from Uh 10.0000 molts up to Uh 99.99999 volts. So it's got three ranges there: 10 MTS 100, MTS, and 10 volts. And this allows me to check the Uh DC calibration of all the meters I've got.

Now the best multimeter I've got here in the EU V blog Lab is a a classic HP 3478a and these are an old but a really great bench meter you can pick up pretty cheaply on anywhere on eBay for anywhere from 100 to 200 bucks including calibration certificate uh sometimes and they're really quite a stable meter. and I highly recommend one. Highly recommend them if you can get one at a reasonable price. Once again I bought this on eBay not sure how much I paid for it I think it was under a hundred bucks or something like that and I have not adjusted this one either.

It is absolutely spoton. Now what I've done here is I've just let both of these uh warm up because the 3478a here uses a Uh heated Vol a controlled temperature voltage reference in it. so it does take some time to warm up and also a voltage standard like the Uh MV 106 Here you have to. you know, in theory you should probably leave these sort of things on all the time if uh, you know if you take your calibration and your voltage standards seriously.

but you know I don't do that in the lab here. It's just for curiosity sake and mucking around, but we'll find that this is more than good enough for measuring for checking the Uh calibration of the multimeters I've got here. So uh, I'm using this Mv16 as a trans what's called a transfer standard and that means it's I've had it tested in a proper, certified, traceably calibrated calibration laboratory and it was absolutely spoton now. Temperature coefficient: I Won't go into this.

This is not a a calibration tutorial by any means, but temperature is going to play a big role in these. That's why I let these stabilize I've had them on for an hour or so. The current temperature in here in the lab is about 24.5 degrees. Uh Celsius here.

but I found the temperature in the lab when it varies from, you know, 21 degrees with the air con on up to 26 if I Come in at the first time a day and it's summertime and it's a bit warm here in the lab. It really doesn't matter these things Don't Drift at all to the resolution that I've got now I Just wanted to show off how spot-on these two are. Remember I got these both on eBay we're looking at you know, like 150 bucks total worth of instrumentation is probably more like at 200 by the time I included. uh, postage here, but they're absolutely spot on.

So what I've got is on the 10vt Range one point and that's that's what that light there for decimal point 000000 volts and so that's a 5, 1/2 digit range and the 3478a, 5 and 1/2 digits as well. There it is. Look, it's absolutely bang on I have not needed to touch this since I Bought bought it on eBay and watch this I can just dial in the one there and it goes up two three. It's absolutely spoton.

Look at that. Could you ask for anything better than that? Yeah, really? In, you know, to get better than that, you got to have to have you know, a certified uh, you know, a proper certified cowb. This is absolutely brilliant. So these two together are more than capable of of uh, checking the calibration of my fluke 875 and my Agilant U1272a and my Gossen meters and stuff like that.

So more than good enough because they're an order of magnitude better than the other multimeters I've got. But by uh, having these voltage standard with the 5 and 1 half digit multimeter I can check I mean if this is still spot on in a month's time, I know that it's going to be pretty rare that both of them have absolutely drifted absolute up or absolute down. The odds of that happening, you know, are pretty low. So I can I have a good level of confidence that as long as these two continue to match, then they haven't really drifted anywhere at all.

In absolute terms, M Uh, let's go down to 10 molt. So I'm in the 100 molt range and I'm generating uh, 10.4 M volts and ah, once again, you know it's jumping around. There's a bit of noise there I might need to let it settle some more or something like that, but it's pretty darn close to spot on. Even when we're measuring 10 molts isn't that scary? So I can go down even further and we're talking 1 molt there.

but you know I mean we're We're really down into the noise there. Who knows what this is picking up I don't know. it's a low impedance source so it should be okay. But really, that is quite remarkable I can just dial in whatever I want.

let's go up arrange to. Well there we go. we've gone up to 10. so we put uh, the maximum 10 volts and it's spot on.

I Love it. And at this level here, when I've jumped up a range 10.11 1111 I've still got five uh, digits past the decimal place there, but this is only displaying four. so I can tweak this one. and yeah, it's going to go up by that one.

so this one has an increased digit resolution over that. but I can set that to 2222. Not a problem at all. A Love it! What an eBay score! And as for a resistance standard here, I've made up this little box here using two.

Um, I think they cost me like $20 per resistor or something you can get them from Digi key There's the vishe part number for this: 10K Ohm resistor at nominal value at 25 C Is there rated to plus - 0.05% or 50? uh PPM Accuracy plus minus or absolute? Uh plus -.2 PPM per degree C Tempco and over the load life of it, it's only going to vary by another 05% or 505050. PPM And I've got a 1K resistor as well. that one's not 005, it's 0.01 So these are pretty good. and there you go.

if I whack that one on there, it's basically least significant digit smack on 1K So once again, with these uh two I can see if they've drifted or not. So I use these as my uh, absolute reference here in the lab and once again, it's you know it's uh, if they both. if it varies, um, over time, then I know it's uh, drifted. either the resistor itself or the meter drifted then I can do something about it.

But if it basically if I want to calibrate my meters I just uh, leave these running for a while spot. Check them and Bingo! I've got myself my fairly confident absolute voltage and resistance standard. And of course, if you got voltage and resistance, you can actually measure current as well. Not a problem.

So I've got the capability to at least measure up to Uh 10 volts. uh DC ranges I don't have a voltage standard that generate higher than that? I've got voltage power supplies that can go up to 1,000 volts, but they're not an absolute, you know, very low Tempco uh voltage standard because it's all about that temperature coefficient, the stability. People think it's about the absolute, accuracy. it's not.

You can have a 5% absolute accurate uh reference standard and it can be worth you know, a million bucks because it's got a you know, a .00 you know, 1% Tempco or something a very low Tempco and it' be very expensive. So it's not all about the absolute accuracy, it's about that temperature coefficient and how much it drifts with time and temperature. What I've done here is I've really cranked up the air. Con: I've got it down to 19.1 De haven't really let anything, uh, settle at all.

so it's been ramping down and we're only too significant digits under, so it's not that bad if I switch it down to 100 MTS we're still sorry 10 molts, 100 molt range. We're still spot on in terms of 10 molts there, but on the 10vt Range we're using before, we have dropped a couple of least significant digits, so you can clearly see that one of these or both of these has drifted a couple of least significant digits there. It's most likely to be the 347 8 A actually, rather than the uh, the voltage standard I've got here, but it's only two least significant digits. If we let it stable down more, we'll probably find that it might settle back down to where we were before, but it's not a big deal.

I mean that's a 5 C a pretty rapid 5 C temperature drop and in terms of uh, calibration, that is 5 C is a pretty drastic change. but even then it, uh, still still really? Um, I'm pretty darn close to spot on. So there you go. That's just a quick look at, uh, the gear I've got here in the EV blog lab to allow me to do uh, basic calibration checks, or you know, transfer standard test and you can essentially do the same thing yourself.

If you've got a bunch of multimeters, you got two or three multimeters, you can continually cross reference them with each other against, you know, basic resistance standards. or you can get a voltage, uh, standard chip pretty cheap and build one up. or you can actually buy uh, off the shelf ones for like $50 or something like that. And really, um, there's no reason for your meters to be out of calibration in quot marks.

And when it comes down to it, if you really want to go, Bare Bones You don't even really need resistance or DC voltage standards like this. If you buy at least have one quality meter like a fluke or an agilant, then you can be pretty confident that that meter is going to be absolutely calibrated and mean it specs from the factory when you buy it brand new. So if you got two or three multimeters, you can then cross reference them all against one another and you can be pretty confident that they haven't drifted at all. So I've effectively got myself a little do-it-yourself cow lab here.

and I'm pretty darn confident in the setup. and I can continue to monitor this thing over the years. Check the uh, uh, you know the drift of it against other traceably calibrated stuff, so you don't necessarily have to send your gear away for calibration, especially. um, if you've got multiple multimeters and you know you measure them all at the start and they're all the same.

You measure them in a year's time 6 months time, and they're all the same again, then you can be pretty darn confident that they they haven't all drifted identically. So if you like the video, please give it a big thumbs up. And if you want to discuss it, jump on over to the Eev blog forum and just after this, I'll link in a couple of videos for the MV 106 here and some other calibration. uh, stuff I've mentioned.

So go watch those videos if you haven't already. catch you next time. Yeah.