Calibration on common multimeters?

[MBY]

I was inspired by this thread: https://www.eevblog.com/forum/index.php?topic=3020.0

Many medium to high range handheld DMMs today used closed case calibration. The calibration is not done via trimpots but by feeding a reference voltage/current/resistance/freq/etc to the meter when put in "calibration mode". This is demanding for the hobbyist in that you cannot calibrate to an arbitrary voltage as you need fixed stable voltages at, say, 19.000 or 39.000 VDC, eg.

I have plenty of top quality, stable instrument to do calibration and and lots of chepo, mid range and high end DMMs. My problem? I don't know how to enter calibration mode (if even present).

I thought that this thread could be a list of known methods to enter cal mode in common DMMs. This procedure are typical (and for good reasons) left out of the user manual and service manuals are seldom to be found for lower brands of DMMs.

So, a bunch of meters that I know calibration can be done on, but no other clues.

If you know some DMMs calibration procedure, especially if it is a "soft" closed case cal, please share with us!

More specific requests: BK Precision 2709B, Amprobe/Meterman 37XR-A (I'm sure that at least the 2709B has a calibration procedure and no adjustable components on PCB, so it must be a software one). Do anyone know?

(I have emailed a bunch of manufacturers but no luck so far).

Edit: I lied about the BK 2709B - it really have at least three really small pots on the PCB, named VR1-3. Anyhow, without the cal procedure I don't dare to adjust them.

Edit: The list (only on entry so far ):
37XR-A: 37XR Calibration Procedure (soft + HW adjust, no closed case cal)

Edit: Changed topic name to better reflect that the thread is intended to list calibration procedures in general. Sorry, I was not very clear in my first post.

[alm]

This is demanding for the hobbyist in that you cannot calibrate to an arbitrary voltage as you need fixed stable voltages at, say, 19.000 or 39.000 VDC, eg.

Some (high-end) meters allow you to specify the voltage you feed it, so you can calibrate it to 13.783V.

I have plenty of top quality, stable instrument to do calibration and and lots of chepo, mid range and high end DMMs. My problem? I don't know how to enter calibration mode (if even present).

The common method used to be a recessed button or switch. The use of a calibration seal requires some sort of mechanical button under the seal, but I'm not sure if the cheapies even use a calibration seal? The only cheap DMMs I've seen use old-fashioned trimmers.

[IanB]

Let me add the Protek 6500 to the list. This has a calibration seal on the back (actually a removable sticker) and under it a recessed button to press. However, if one presses the calibration button, the question becomes what happens next? Could one totally screw up the meter by pressing it and not having instructions on how to exit calibration mode without saving changes?

[EEVblog]

I had to re-cal my Gossen Energy recently when a firmware update trashed the cal values
I have some of it on video, I needed to use USB cable and a terminal program, and the commands and procedure that Gossen sent me.
Pretty annoying.
It appears as though the DC calibration (which is all I've done so far, to 10V) set the resistance as well.

Dave.

[alm]

With models not using a computer interface, it often involves setting the meter to the function you want to adjust, and then use some of the buttons to cycle through the various ranges. The actual buttons, order and expected input should be in the calibration manual, unless you have some free time and don't mind corrupting the current cal constants.

Most of the cheap meters are unlikely to ever see a cal lab since the cal costs exceeds the costs of the meter, so I doubt that they spend much time documenting it.

It appears as though the DC calibration (which is all I've done so far, to 10V) set the resistance as well.

I'm surprised, how do they calibrate the internal current source with just a voltage source? Even artifact calibration usually requires a voltage reference and a reference resistor.

[EEVblog]

I'm surprised, how do they calibrate the internal current source with just a voltage source? Even artifact calibration usually requires a voltage reference and a reference resistor.

I don't know, I have to clarify with them.
All I know is that all the readings were screwed up before the cal, and now resistance is spot on.

Dave.

[alm]

Of course the resistance measurement feature consists of a voltmeter and a current source. Maybe they calibrate the ADC/voltage reference separately from the current source, although it seems unnecessarily complex.

[Kiriakos-GR]

My problem? I don't know how to enter calibration mode (if even present).

At this point is where the proper documentation does miracles. 

Fluke hides the 70% of the process.
Agilent looks more big spender about sharing the info.
And I have no awareness about cheap meters with close case calibration process, this sport belongs to the expensive range.
 

[saturation]

As a typical cal is about $100, its important to do some checks and adjustments on your own otherwise the cost of maintenance will easily exceed the cost of the DMM.

If the cal procedures aren't published, that model DMM is in a dubious league.  One shouldn't consider owning such a DMM, or at worse case can't be guessed, particularly the close case types.  Many lower cost DMMs used a few trim pots, and you can easily experiment to see what it does and these can be better value than intermediate priced close cased ones without any support documentation from the maker.  Lower cost ones tend to drift sooner, but it could also be easier to adjust.

If the cal procedures aren't complicated, that is one or two adjustments whether closed or trim pot, you can adjust the meter by comparing it against another calibrated meter using a reference that is stable long enough to 'transfer' a reading from the calibrated meter to the DUT.  We discussed this on the link from the OP.

[MBY]

Yes, even a close cal can be simple. If you can allow to experiment and destroy the cal, you can probably figure it out. But the problem is not the calibration in it self, it's enter the calibration mode! With around 8-10 positions on the range switch and typical 4-8 buttons, there is a lot of combinations to test. I haven't succeeded to enter calibration mode on a single meter that I know have closed case calibration. Some may have "semi closed" by removing a jumper or pressed a hidden button, but some models is hard to understand. The link I provide in the OP talks about the UNI-T 71 meter. I have the Conrad "VC940" rebanding and as the video shows it has a software calibration procedure. I know it has a jumper on the PCB (and at least a trimpot), but scanning through the thread does not reveal how to actually enter the cal mode (maybe I missed it?).

Edit: When I say I "know" that a meter has closed case or soft calibration, I check if the meter is listed on calibration services or the meters manufacturer homepage as calibratable and check the PCB for adjustable components. The trimmer can be very small, like the size of 0805 or even 0603, but when I do not found any of this and the meter is told to be adjustable I conclude that it should be a soft process. Well, off cause some models may not be adjustable at all and the calibration process only checks the meter to specs with no adjustment.

Edit2: I found the cal procedure for Meterman/Amprobe 37XR-A. Link in OP.

[Lightages]

I have the Conrad "VC940" rebanding and as the video shows it has a software calibration procedure. I know it has a jumper on the PCB (and at least a trimpot), but scanning through the thread does not reveal how to actually enter the cal mode (maybe I missed it?).

To enter calibration mode on the UNI-Ts and that Conrad, hold the blue button and exit and then turn on the meter to the desired function to calibrate. You can see in the videos linked in the thread I started on the UT71E this procedure.

[firewalker]

I had to re-cal my Gossen Energy recently when a firmware update trashed the cal values
I have some of it on video, I needed to use USB cable and a terminal program, and the commands and procedure that Gossen sent me.
Pretty annoying.
It appears as though the DC calibration (which is all I've done so far, to 10V) set the resistance as well.

Dave.

The cheap one? 

What if you hadn't bought the cable? Would they accept the device in order to do a free of charge calibration?

Alexander.

[EEVblog]

The cheap one? 

What if you hadn't bought the cable? Would they accept the device in order to do a free of charge calibration?

No idea.
I happened to have the super expensive cable.

Dave.

[Wartex]

I had to re-cal my Gossen Energy

This is after beating yourself in the chest that "calibration is not adjustment". Sheesh man, be consistent at least.

[nukie]

Goodluck with closed case calibration with less popular meters, the manufacturer would just tell you to send it back for 'repair'. What you can do is, you get a fairly accurate meter and compare the results side by side then take note of the differences. Also, do it at room temperature. You shouldnt be messing with calibration if you have a high spec equipment chances are you dont have a reference standard that is good enough for calibrating your target.

[RCMR]

I wanted to get a baseline for some of my DMMs so I ordered a TL4050 voltage ref from Element 14.  They sent me a PN2907A in a plastic bag carrying a TL4050 label.  Fortunately it was a TO92 package rather than a SOT23-3 otherwise I might not have noticed and then spent a long, long time scratching my head as to why it wasn't working properly.

Looks like they use electronics neophytes to do their bagging at Element 14 (ie: those who don't know the difference between a SOT23-3 and a TO92 for a start).

[Jidis]

Edit: I lied about the BK 2709B - it really have at least three really small pots on the PCB, named VR1-3. Anyhow, without the cal procedure I don't dare to adjust them.

Dumb question, but I just looked inside a cheap Greenlee (DM-55) here and found the same thing. The center pot turned out to be able to adjust DC voltage readings, but the other two seemed to have no effect on anything, including the LCD. I was using a reference board which includes 5V AC/DC,1mA current, and resistances of 100 ohms to 100k.

If any of you guys had to guess at it, what would you figure those other two trimmers are for?

Thanks!

PS- FWIW- I checked the resistance settings of all three beforehand, so I could put them back where they were, but I left the newly aligned voltage pot where it was (probably not a wise idea if they interact somehow).

[retiredcaps]

Dumb question, but I just looked inside a cheap Greenlee (DM-55) here and found the same thing.

Post a clear focused picture of it so we can see.

[Jidis]

Post a clear focused picture of it so we can see.

Thanks retiredcaps!   I'll have to rip into it again tomorrow, but it should be no problem.

Take Care

[Jidis]

Back again retiredcaps,

Not sure what you can tell from this as most of the traces near the pots are obscured by parts, but here they are. I'll try to keep it apart for a bit if I can.

I guess I mainly was wondering if anyone familiar with DMM guts has noticed any common adjustments they typically give access to on the ones with more than one trimmer.

Thanks Again!

[devanno]

I had to re-cal my Gossen Energy

This is after beating yourself in the chest that "calibration is not adjustment". Sheesh man, be consistent at least.

Dave is not being inconsistent here... he already indicated that a firmware upgrade trashed the calibration data on the meter, and I assume from that statement that it was essentially bricked.

As to "would they take the meter back for a free recal if you hadn't bought the cable" ... well, if you hadn't purchased the cable, you most likely wouldn't have been able to do the firmware load that trashed it in the first place.  Just sayin' ... 

[saturation]

DMMs have specific calibration requirements that are rarely identical for each model, even with similar key ICs used.

This low cost design looks like its powered by an ICL7106, based on the length of the chip and suspiciously, were all the pots are clustering, its very much like a modified panel meter design to act as a DMM.  My guess is those pots are for calibrating other volt ranges if adjusting them cause nothing to mA or ohms. 

Back again retiredcaps,

Not sure what you can tell from this as most of the traces near the pots are obscured by parts, but here they are. I'll try to keep it apart for a bit if I can.

I guess I mainly was wondering if anyone familiar with DMM guts has noticed any common adjustments they typically give access to on the ones with more than one trimmer.

Thanks Again!

[retiredcaps]

The reason I asked for pictures are:

1) Everyone likes the "tear it apart" aspect and wants to see the insides including me.  Those are excellent clearly focused pictures.

2) Seeing the PCB itself can give clues as to who the original manufacturer is because Greenlee doesn't make multimeters.

3) Seeing the main IC can give clues as to what the pots do.  As saturation mentioned, the 40 pin IC looks like a 7106 or variation.

4) What are the part numbers of IC2 and IC3?

5) If you can take off the lcd without breaking anything (4 clips), can you shine a bright light from behind and take a picture so we can see the traces in the pcb?  Maybe we can see where VR2 and VR3 connect to?

Right now, I have no ideas what VR2 and VR3 (a guess would be temperature for one of them) do, but I will keep thinking about it.

[Jidis]

Thanks Guys!

I went ahead and pealed off that buzzer. Looks like Saturation's guess was spot-on. The main chip is this Cyrustek ES5106, which appears to be something compatible to an ICL7106 and is an A/D chip used for panel meters.

@retiredcaps- Yeah, I got lucky with a nice sunny day today for snapping pictures. I'll try to see if I can work out a "see-thru" view of the traces. IC2 & 3 are just a 358 opamp and a 4011 gate chip.

I'm not in love with this meter or anything. It was bought many years back in a pinch when a nicer one broke. Don't know who decided on hardwired test leads, but it looks like the case has holes where banana jacks should have been (they don't line up with anything on the PCB). If the pots turn out to be temperature or volts/amps outside the range of my reference board it's no biggie, I'll just leave them put and hope I got them back where they belong.

Take Care,

George

[Jidis]

Sorry it took me a bit to get these.

Here's two shots of the bottom and top. - Note: I "mirrored" that top shot, so all the pins would be in the right order

Take Care