Calibrating and hacking the UNI-T UT210E clamp meter
Vložit
- čas přidán 28. 06. 2024
- My UT210E clamp meter was serioulsy out of calibration in the current ranges. This video shows how I calibrated it and applied one of the many hacks available for the UT210E, to change it from 2000-count to 6000-count. This means for example that in the 2A range the maximum value that could be measured was 2.2A but with the change it can now go up to 6.2A. Same is true for volts, ohms.. Especially in AC volts measuring mains 240VAC, you now have an extra digit resolution!
BTW, I could not get the CAL menu to work. If anyone has good(!) information how this is supposed to work, I'm happy to give it another try (now that I have a backup of the CAL data)
Videos mentioned in the text:
Making sets of power resistors: • Home Lab Equipment: Ma...
DIY Power Supply repair: • A case of good caps vs...
SCT013 Current Transformer: • Building an adapter fo...
Measuring RMS: • Measuring RMS Voltages...
Ben bdlow? has a brilliant collection on material for hacking the UT210E here: github.com/bdlow/UT210E - Věda a technologie
Beautiful! I need to buy one of these during the next sale. Great work.
Thank you very much for this informative and educational content.
You are too skilled for the average user! Thanks for sharing
Very good explanation.
The Err0 in calibration menu is due to the probes (in your video) EEPROM perform a self-calibration test (compare and rewrite if the modifications are correct) so the probes have to be disconnect (at least open circuit) to entering into calibration mod otherwise the result will return an error !
Thanks for the tip Park80
My understanding is that you need to turn dial to Ohm's to start CAL, instead of Volts.
I have owned this meter for many years but I am happy with my readings - its out by 7% +-
Yes, I agree in its price range, its pretty much unique in having a DC current clamp, with a useful low-amp range and overall hackability
Thanks for the interest. It read .4 something each time it was powered. It zeroed OK but then read .2 to .5 amps when I know the current was greater. I have a Fluke 179 auto R multimeter 40 years second hand and still working although in need of a calibration no doubt but still accurate enough for me at 81. Bad luck used once then useless.
I'll send them an email that took hours to find.
Measuring current directly with a multimeter like the 179 is always going to be more accurate. The selling point of clamp meters is that you don't have to disrupt the circuit you are measuring and there is a price to pay for that convenience. That said, the clamp meter should not be that far off. I believe you have the 210d and its lowest current range is 20A with 2% accuracy and 10mA resolution, so its more suited for larger currents. Also measuring with the clamp is kind of an acquired skill. You need to hold the closed clamp's nose close to the wire, then zero, then open the clamp to get the wire inside and make sure the clamp is again really closed. While doing that, you don't want to change orientation because of earth magnetic and other fields. The 210e has to arrows on the clamp (very faintly) that show how deep the wire should be in the clamp.
@@TheHWcave Thanks for the reply..... 20 A ( i hope you don't mind me mentioning) is the Max amperage that can be read. It was OK when I have used it only once when it correctly read near enough the amps output to a 12v vehicle battery after I had repaired the charger which is quite a big one and has fast as well as normal charge rate. I wanted to see what the max current flow was to the battery on fast charge particularly. As usual claims by the manufacturer fall short of what is expected of a car battery charger !
The charger has always worked but not to my expectations although the fast setting charges faster than the normal (trickle charge). The calcs re the time charging proved the clamp meter at the second use was a waste of money. Hence my quip - any electromechanical can go faulty from day 1 .
Having just bought one of these meters I found this video. A little bit over my head though. Looking at an online Hex converter should 40330 not be 9D8A? Anyway I compared mine to a Fluke and was within 3%. Would be nice to have it a bit more accurate but at least I now know to deduct about 3% from any reading I get
Great cal info. I think mine is closer to its spec in current ranges, so I won't go in and change cal values just yet. You got away without pulling uP rst low - I see from the github link that this may or may not be necessary - I did pull it low with a jumper, as I couldn't get it to talk properly without this. I have changed some eeprom values a few years back, when I read the EEVBLOG thread - swapping AC for DC on select button, longer light t/o, raising the 100A limit to, um, 300A I think - thought it _does_ under read above about 120A, (jaw saturation ?) so I set the warning beep at 120A to remind me that readings are not accurate from here on up - nice to be able to see whether you have 120A or 250A approx. (for eg) though. This doesn't come up often mind you.
I didn't do the 6000 counts mod, as it screws up when using 2A mode, as you show here - and I didn't like the sound of adding the full 'dotless' 6000mA mode. Maybe one day. I used my PICKIT 2 pic programmer and the Microchip s/w to read/write by the way.
Hi noakeswalker, thanks for the comments and the info on what you did with your UT210E. You are correct in that I did not had to hold the processor in reset. I always read consistent values without that and programming worked fine too.
On the 100A vs 200A issue and the increasing error, I looked at the schematics (I found it online here:
elektrotanya.com/uni-t_ut210e_schematic.pdf/download.html
and the amplifier following the two hall effect sensors is switched between x64 for 2A, x6.4 for 20A and x1.4 for 100A. It seems to me the last value should have been x0.64 if this meter was really meant for 200A. With the chosen amplification, I guess currents >100A simply drive it into saturation.
@@TheHWcave Thanks for cct diag link - I had not managed to find one before, and had reverse - engineered some of the front end analogue stuff, so it's nice to see the real circuit. I note the range gain values you mention, also the Hall sensor current is adjusted slightly, but only in the '200A' range. There are references to both '100A' and '200A' ranges - I wonder what caused the end product to only make it to 100A ? Suggests there is something else limiting the max current....I guess we'll never know !
The UniT 210D uses the same physical case by the looks of it, yet has a 200A dc range. (ie the clamp jaws are likely the same inside). So maybe this is just 'marketing'.
More UT210E tweaks: Current offsets can be zeroed, but you are still stuck with the variation in that 'zero' with meter position - I have now managed to tweak this variation to a much better value, (I could not find ANY useful info on this online) and only properly after getting the schematic confirmed. It involves powering the meter with the back off, and tweaking VR4 (the null pot in the Hall bridge) carefully, then moving the meter through a repeatable pattern of motions, and noting the min/max/difference readings. When I had a few of these, I could see a distinct 'low point', I won't call it a null., it's not sharp. Once set in that low point, you have to tweak either (not both) of the two pots VR+/- to get a proper power-on zero offset back. My meter is now much less sensitive to position, and this process doesn't seem to have affected either cal. or the slight variation in readings you get with wire position in the jaws. Nice ! Let me know if you want the full procedure - it'd make a good subject for a video I think...though that is optional of course...it's tricky enough without messing with filming it.
Atb, Dave
Hi noakeswalker, that sounds great! Yes please get me the procedure you followed. For example, did you do the "rotating of the meter" 3D (sphere?) or 2D (e.g horizontal circle)?
I already made myself a piece of wood with 2 pogo-pins to power the UT210E with the back off. Its held on with a rubber band. Crude but it works. In the end I did not need it when programming the EEPROM. Sounds it will come in handy after-all.
Brilliant piece of research. If I do make a video about it, you will get full credit!
@@TheHWcave Hi mr. HWCave - meter rotation I did in two circular 2D ways - one H and one V. (so yes, sort of 3D) I _think_ you could actually miss out one and just rotate in a vertical plane in fact, since this seems to give by far the largest offset values in each direction. I did it two ways in case I missed some vital and annoying dependency. Weirdly, with the meter rotated in the 'upright' plane, ie plane of display is vertical, I didn't get any variation of readings with 'compass' direction - which is frankly amazing. There's more going on there than I care to imagine I think...
Your meter (from the vid) seems to have about 100mA offset even after degauss - the q. is - does it vary with position much ? If it doesn't, then you might well not gain anything by doing this tweak - fixed offsets we don't care much about with a zero button, it's that varying offset that is a pain, isn't it ?
I have emailed you the text file with my method. Hope it makes sense. Any questions just ping me.
Thank you for this amazing content. Is it possible to share CAL values for other ranges as well? I don't have proper equipment to calculate CAL values for other ranges.
Hi Ibrahim, I am not sure I understand what you are after? About 9 mins into the video, I share the interpreted dump of my EEPROM. This shows all CAL values for all ranges in hex. The video then explains to convert from hex to decimal and then to take the decimal and divide by 32768 to get the calibration factor (CF). If the conversion from hex to decimal is the problem, I think the windows calculator app has that (you may need to switch to programmer mode or something like it). Another way is to use Excel and the built-in hex2dec() function.
@@TheHWcave No, I mean you changed those values for better accuracy. If I'm not wrong, you changed CAL Value for 2A(6A) range from 36505 to 39447 for better accuracy. I was asking those CAL values for other ranges as well.(20A and 200A)
@@Ibrahim-mk4qh Oh, I see, sorry my mistake. My 20A (60A) CF was 1.153869629 or hex 93B2. For 100A it was 1.069885254 or hex 88F2
The current calibration was the only things I changed. Of course these values are meter-dependent so I don't recommend just using mine for your meter. If you managed to do the 2A calibration, doing 20A should be easy. Just repeat the process but this time wind the wire carrying the current exactly 10 times through the jaws of the UT210E.
@@TheHWcave Ok, I didn't know it's meter dependent. I will try it for mine. Many thanks.
EXCELENTE!!!, La mia está descalibrada.
hola en cuanto tiempo se descalibro?
HolaJMV. Hace mas de 2 años. Ya lo deje por imposible.
Yyyy best ripley please
Very interesting and informative video! I have a Uni-t multimeter that uses the same basic chip set and have wondered about tweaking calibration of the various voltage ranges. It seems logical that the 'cal adj. ratio' data stored for the DC 6V, 60V, 600V ranges would follow the same coding methodology as the DC currents, and I see in your video that the default values are 8000 hex, just like the currents. However, the DC voltage cal numbers you retrieved from your eeprom (as seen at the bottom of the table presented at 8:56) are quite low, suggesting some other coding method is used here. Any idea as to how these particular numbers are coded?
Unfortunately no. You are correct in that some of the DC V factors are pretty low which means a rather unlikely ratio between the reference and measured value. Strangely the 600mv seems to use a factor of 1, same as for 600A and 6000A where I strongly suspect that UNI-T simply did not bother calibrating.... But I tested the DC mV and V ranges and the meter is in spec. I did not explore the coding myself but used that Russian forum's finding and yes, I was really only looking at the amps.... One way to find out would be to override say the 6VDC with 0x8000 which should be a factor of 1 and then try some other values to see how the reading changes. As I said, at the moment it seems spot on. I am too busy to do it but if you have the gear and do it for your meter I would be very interested in your findings.
@@TheHWcave I bought one of those little CH341A programmer modules on Amazon and was able to read out the eeprom on my meter. For the most part, contents follow the exact format outlined in the DTM0660 document and your video. However, the data fields designated for 'DC volts' correction factors (addresses EA~F1) are all written 'FF' on my meter. Out of curiosity, I reprogrammed a couple of these data fields with values similar to what one would expect to see, but it had no effect on DC Volts readings on any of the ranges.
Interestingly, what does affect the calibration on apparently all of the DC Volts ranges is the '500mV correction rate' parameter at addresses (22,23). Tweaking data values here will cause displayed DC volts readings to scale accordingly across all DC volts ranges. It appears to be a 'master' base correction that all DC Volts ranges utilize. Still don't know how the individual ranges are adjusted (maybe they're not?), but it's not as outlined in the DTM0660 document EEPROM table.
Wow, that is really very useful information. Many thanks for doing that test. One should probably do a deeper investigation since the meaning of some of the EEPROM fields is obviously different from what everybody seems to believe. Unfortunately high demands from my day job means I can't afford the time, so if I get to it at all it would be not very soon.
Those locations are not correct for DC Volts. The calibration values for those start at E0: millivolts, volts, 10's, 100's.
@@bobwhite137 Thanks much for that bit of info. A readout of my meter's (UNI-T UT61D+) EEPROM does show locations E0~E6 contain data consistent with default calibration values (i.e., 8000H). I'll try changing these values slightly to see if the various DC Volts ranges are affected.
Do you know what memory locations (if any) are used for calibration of individual Resistance ranges? There's no such designation in the DTM0660 document. My guess is it may share the DC Volts range calibration values, since the same internal resistor divider network is used for both functions.
If would buy a Clamp meter from AliExpress that i'm using for measuring DC amps up to 30A DC, would this model be any good. Are there any better ones that do not need to be zeroed calibrated before use.
The UT210E can measure DC 30 amps easily. Moreover, with that kind of currents, a few mA offset are pretty irrelevant. The problem of offsets really only matters in the range to 2A. As for other clamp meters, you need to look very carefully at the specs. Most budget clamp meters only do AC amps because they use a much cheaper current transformer and not Hall-effect sensors. As far as I know the problem of needing to zero offsets caused by the earth magnetic field (and stray fields from wires near by) is common to all clamp meters that can do DC amps. I just checked the manual on the Fluke 381, a >$600 clamp meter, and yes, you still need to zero the meter before measuring "to remove any dc offset that could affect the accuracy of dc readings".
Can you use a ut-t10k probe for temp measure with this multimeter? I suppose that it can measure drop voltage from temp probe
I tried to find data on the ut-t10k and UNI-T are not very clear. If it is a thermocouple (two dissimilar metals) I don't think it will work very well. The voltage is tiny. I tried it with one of my K-type thermocouples in a UT210 and heating it in the hand produced maybe one or 2 digits. If the probe is using a PTC instead, you can use the OHM mode to read its resistance. In either case you would need to create your own conversion table of temperatures vs ohm/volts by using a few known temperatures and checking the readings on the UT210. I wonder if it worth the effort compared to using either a dedicated electronic thermometer or a multimeter with a real temp range?
By the way the chip inside the UT210E can actually do temp measurement but hacks to enable it seemed not to have worked. At least I could not find one.
Thanks for information. Have both same ut and bm and tl866a programmer, and psu 32V@5A. Going to do some tweaking do i need to buy ch341, or just clip only in your case it doesn’t helped? Could i same way calibrate dc voltage range because it is few counts out?
There are so many clips out there in different price ranges and features. Some people have used the clips successfully to read and program the EEPROM, Maybe I was only unlucky in my choice of clip. Its more a mechanical thing with the clip not making proper contact. Main thing is of course you need to somehow attach it correctly to the EEPROM and the programmer. It will probably work but I don't know how to do do that for a tl866a
@@TheHWcave Thanks, you are correct with pour clip contact issue, a saw same problems at another channel video, author just cut some plastic from clip nose and it fit to IC. You can do same, and check continuity after clipping with multimeter. What about voltage readings? Is it possible to do them more precise by flashing?
Yes, I considered trimming the clip but in the end the connector I used worked so well that it was not necessary.
Once you have access to the EEPROM you can calibrate the DC(!) voltage readings as well. It works the same way. You do a bunch of readings with a trusted multi-meter and the UT210E and calculate the correction factor for each range. Note that for space reasons I did not show all EEPROM locations in my video. For AC volts the calibration is unclear (to me) because there are a bunch of "noise deduction" (whatever that means) factors that do not follow the CF method. This means that if you want to do AC volt calibration, you will have to do more research yourself. I don't have that information, but would be grateful if you can share whatever you found out.
@@TheHWcave Thanks for reply. Glad to hear what dc voltage calibration is possible. Will by programmer then. Unfortunately i can't help you with AC calibration info because i do not have needed precision ac source with out those spikes you ve mentioned and even oscilloscope so my way as yours dc. And of course i should do backup before starting.
Wow..
Hi, I am having a bit of a strange problem here. I bought an UT216C about 3 years ago. Sometimes it would not zero in DC-Mode. Was about to write it off and bought an UT210. Wanted to test my buzzbox-welder AC current and now it zeros in dc-mode. I think the current reading was about 1.7 A. Now (after the welder-measurements) it shows about 0.75 A. Now I know - it wouldn't want to zero because it was off too much (1 A seems to be the limit). Yes I bought that demagnetizer and tried to fix it that way - unfortunately didn't change anything
Is it even worth trying to calibrate that thing, if it is wandering all over the place. I can't imagine, that behavior can be considered normal? Any ideas? :)
I am a bit confused. So, first we are talking about a 210E not the 216C? 2nd, the 210E did not zero in DC mode, then you measured an AC current and now it zeros? This would point to residual magnetism in the clamp and the AC field managed to at least reduce it. It now shows 0.75A in DC mode when no current is measured? The demagnetizer needs to be used with a decaying field to work (either add a PTC or drag the UT210E away while the demagnetizer is working). It often takes several passes and with different orientation of the clamp. Does the meter wander when it is lying steady on the bench? It should be stable when not moved. It is normal to change readings when the meter orientation is changed because of the earth magnetic field (or any other stray field, e.g. from magnetized table frames or legs, metal racks etc, so maybe try a different location to see if that makes a difference.
elo sir, can u make for 210d?
Interesting testing and reprogramming calibration of the EEPROM! Nice job. Too bad the auto calibration menu function produced ERR0.I bought the pro version of this meter that has duty cycle also and some extra range in capacitance. I wonder if it will suffer from the same issues yours had calibration wise.
I am pretty sure the hardware is essentially the same. But let us know what your experience with the pro-version is.
NVC DIY FIX: use Bounce sheet (the sheet you put in your drying machine) and rub it on the meter and the test lead and the housing... it will prevent the meter from getting statics and beep for no reasons...
and ut61e + can you upgrade to 30,000 counts?
Sorry, no idea. I don't have one of those
Which solution for display contrast increasing?
Unfortunately, I don't have one. The only display visibility enhancement I am aware of is that you can turn the back-light on for longer using a patch in the EEPROM
Hello, I own a (ut216c) (it gave me 5 amps on the DC A reading) without measuring
Yes, totally possible for many reasons. Assuming that nobody messed with the trimpots already check first that there are no stray magnets in the area and if the clamp closes properly . Then demagnetize . These would be the first things to try (you may have to demagnetize multiple times if it is really bad. Only start adjusting the trimpots at the very last resort. because the adjustment process is exceedingly time consuming and delicate
Hola amigo excelente video!
Tengo una pinza amperimetrica Uni-t UT216A que me da mediciones erróneas, será posible que esté descalibrada, como puedo calibrarla?
Saludos desde Costa Rica...
Thanks for the comment. Unfortunately I know nothing about the UT216A and the way it does calibration. I guess it might be similar to other UNI-T meters in that you hold some buttons while turning it on.
@@TheHWcave
Muchas gracias por tu respuesta, no e visto mucha información sobre esta marca Uni-t y realmente no sé el procedimiento para realizar esta calibración...
Hello did you find the solution?
I have one UT210E, whan press hold and select and power on the "CAL" menu show. How to use this menú?
The only description of sorts I have is in this document (starts on page 13) caxapa.ru/thumbs/794708/DTM0660---.zh-CN.en.pdf
Tried the link but it just times out.... possibly a ruzzian site that has been taken down?@@TheHWcave
I tried to degauss with the meter turned on and set to current DC as it is showing 1.5A but it didn't degauss.
What degaussing methods did you use? My experience is that it will take a couple of attempts and different orientations of the meter to reduce the permanent magnetism,
waw great hack,i own the ut210d and after entering to calibration mode by pressing select button and hold button doesnt work well,do you know if there is any way to hard reset or something? thank you
Sorry, I personally don't but hacking the UT210's seems to be popular, maybe a search on the wider internet? When I was doing my research then, some Russians seem to be very much into it.
@@TheHWcave thank you,there must be a procedure because at 9v was showing 500mA and after trying some buttons sequence at 9v has rised at 7v but stuck there,i have talk with the seller and they will replace the unit in warranty
@@radu_mol2935 That's great service, so rare these days. Better luck with the replacement
What the value of the thermistor ptc1?
ptc1 (and ptc2) are shown as 1.5K in the schematics
@@TheHWcave Do you know the value of the ptc1 e ptc2 of the ut-216c?
No, unfortunately not.
i have 216c (Is it useful if you measure the resistance value?)
my multimeter uni-t, start problem😭😭😭 im so frustrated!
What kind of start problem are you having?
@TheHWcave, sprichst du normalerweise deutsch?
not anymore, since I am living in the UK for 20+ years
@@TheHWcaveoh, OK. Du machst sehr interessante Videos aber die gibt es dann sicher nicht auf deutsch oder? 😂
@@Chris_Dr Leider nicht, fuer 2 Versionen habe ich keine Zeit. Ich bin naemlich kein professioneller CZcamsr
in my opinion, with readings 8% off that meter is simply broken and cant measure properly. If Uni-t is too lazy to calibrate it properly, or simply make calibration potentiometer then everyone should send it for warranty. If they lose money maybe they will learn. My 20 years old multimeter have easy calibration and thats why i can still use it.
I agree that UNI-T should have paid more attention into calibrating this meter more accurately in the factory. I guess it comes down to price. It used to be that the UT210E with its DC current range down to 2A was an unbeatable bargain compared to other clamp meters (the DC current capability has its own problems, see my other videos on the UT210E). Measuring accurate DC current on a clamp meter is very difficult. I have briefly played with some expensive professional meters and they get around the stray zero problem by simply having a very large lowest range, e.g. 60A. A few stray 100mA are hardly noticed in that range, but its tough in a 2A range like the UT210A. These days, there seems more competition, mostly from other Chinese manufacturers but I am wondering if these are much better. I am very curious when Dave Jones (EEVblog) clamp meter made by Brymen comes out
@@TheHWcave UT211B can measure 0.1 mA and according to some forum posts doesn't need to be reset to zero when measuring low currents so frequently as UT210E. Unfortunately it costs more than double compared to UT210E.
@@ChupoCro Thanks for the update. It seems (to me) more expensive DC-capable clamp meters do an automatic zero, at least one (rather old) LEM Heme meter I briefly played with did that. I guess its the old "you get what you pay for"
In a nutshell the Uni-T210D AC DC clamp meter is absolutely crap. Used once last year. got it out of the always 70F dry cupboard and it does not read DC amps which is exactly why I bought it. A complete waste of money.
That's of course no good. What does it say when you try DC amps? Do the other functions still work?
P.S have not tried the other functions. I'll give the a go.
@@TheHWcave will it work on uni-T210D ?
@@5iCn4rF Not having one I can't tell but the meters look very similar. It seems the only difference is a higher amp range in the 210D which is maybe just a different gain setting. It is very likely that the 210D and E are basically identical in particular on firmware.
@@TheHWcave I just watched your video a bit in awe, i have no real idea how that is done; meaning i heard the bird sing, but i don't know where from...lol
Well, i have a UT210D ,for 3 months now, and the thing is really awesome in ALL ranges,all readings; only 2000 display counts though; but readings are accurate and rock steady; compared to my 30 yo 20000 count manual fluke,which i no longer need,and don't, use much,since it is not automatic, the UNI spoiled me, so convenient; i bought almost only for the DC Amps on my solar setup,but now use for mostly everything; i only reach for the fluke when in doubt,or for more precision ; also another thing i love is that it uses 2xAAA ,not the expensive 9Volt bat.; and in 3 months,used daily,the batteries are still the same.
Too technical and too much work for this calibration procedure. I'd rather buy a new one, and very possible, the new one would be a different brand.
Changing a few bytes in EEPROM takes about 60 seconds if the case is already opened and the values are already calculated. To calculate the values you don't have to use many measuring points - it is enough to use just one measuring point at the current at which you need most of the precision and it takes 30 seconds to calculate two bytes. *Which exactly clamp meter would you buy instead of UT210E that doesn't cost at least twice and can measure 1 mA DC currents?* BTW, there is UT211B which can measure 0.1 mA and reset to zero when measuring low currents doesn't need to be used frequently.
Mine now switches to ac current when clamp is opened. But its ransom lol. If I switch to dc it changes when its opened. Also if the meter slams shut or opens too fast it quickly switches to resistance then back current lol
Maybe a broken track or something so the meter resets when experiencing mechanical stress? Or it could be bad contacts in the range switch. Cleaning the contacts may help and examining the PCB solder joints for cracks
@@TheHWcave ah . Ok yeah that's a great idea
@@TheHWcave thank you very much