OWON XDM1041 the unknown multimeter...

[davebb]

Good afternoon
The flukes 189 spec say 10A continuous and 20a for 30sec
The fluke is measuring well against my load and psu amp meter and all other meters , 
anyway the shunt wire has a lot of solder on it on the shunt/ component side so i wicked some off and moved it a bit and now i am getting a reading of 32.5ma high at 5A, and 65ma high at 10A, I could Play with it some more but i am happy to leave it there,
Thanks Dave

[theHWcave]

Very interesting. I never thought a bit of solder could have that much of an effect, I mean assuming it did not "climb" up the shunt itself. So before the shunt resistance was lower than it should have been and the meter was under-reading. Now it seems pretty much spot on (within tolerances).  Glad that its working now.

At the moment I am struggling with a similar problem for one of those cheap panel meters that measures DC volts & amps. The amp reading is way too high even when adjusting the trim-pot all the way down. I experimentally added a parallel shunt of 0.2Ohm and that got it nearly there. After your experience maybe I should add more solder to the shunt? I am very tempted to try that.

[davebb]

Hi
Yes, after how much difference the solder getting wicked up the legs of the shunt so yes the solder did climb up the shunt legs,i should have taken a photo, there was a lot of solder on the legs ,
i may well try and get it a little bit closer, but it is working well,
try tinning your shunt,you wont need much solder on there, you can always use solder wick to clean it back off
In a way i wish i had sent the meter back before playing with it, because of the temperature problem ,
it does seem strange how this got past there cal and QA, as the current was way out of spec,
i got a owon ow18e h/held meter to use at work as it has bluetooth and that is ok for £45, i work on a car x-ray unit it was drive thru, now we have it on a conveyor system controlled by a plc with 4x 3ph motors each with 500nm of torque,
we run the xray gen at, 250kv at 3ma ,so with bluetooth i dont have to get to close when testing,
Thanks Dave 

[coromonadalix]

Be careful,  i've seen in the past some obscure ways to calibrate a shunt with added solder on it. Not a the pcb / shunt junction but on the "wire itself"   ??

[Kleinstein]

Adjusting shunts by adding solder and removing matrial with a file was common in the old days of analog meters and early DMMs. Adding solder, especially in the center of the shunt can alter the TC as the solder resistance contributes and there is also some alloying (e.g. tin is incorporated into the base material).

Modern meters usually use numerical cal factors and thus calibration by a software constant.

I would be a bit careful in assuming that a used Fluke handheld DMM must be accurate. Usually the factory calibration is no that bad. There can be rare exceptions, like there can be damaged DMMs.

[davebb]

Hi My fluke reads fine the same as all of my good meters, It is the  OWON XDM1041 that was a long way out but better now
i think they have missed some of the cal at the factory as my temp is way out as well,
i should have sent it back to china to banggood but if i do any temp logging i can now use the fluke,
Thanks Dave

[markderlo]

post you email address or email me at ksmail9 AT protonmail.com and I will email you manual in PDF form

[markderlo]

or try this link to that manual
: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwjb-9HZpbvzAhVpkosKHWiqBw0QFnoECAkQAQ&url=http%3A%2F%2Ffiles.owon.com.cn%2Fprobook%2FOWON_XDM1041_Digital%2520Multimeter_USER_MANUAL.pdf&usg=AOvVaw1OUY_rrkgLrCzgycYeAVAM

[markderlo]

or try this link to that manual
: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwjb-9HZpbvzAhVpkosKHWiqBw0QFnoECAkQAQ&url=http%3A%2F%2Ffiles.owon.com.cn%2Fprobook%2FOWON_XDM1041_Digital%2520Multimeter_USER_MANUAL.pdf&usg=AOvVaw1OUY_rrkgLrCzgycYeAVAM

[J_A_F]

Hey Everyone,

I just bought one of the new revisions of the XDM1041 with USB, and I gotta say, not the most intuitive thing to set up as far as the control software goes.

Figured I would share what I did in case anyone has had the same trouble as me, or to save some poor soul in the future...

1. If you get one, ignore the install instructions, they don't configure the driver correctly, and nothing works. do the following:
   1.1 Install NI VISA (got it working on newest release, 21.0)
   1.2 Install CH340 drivers (the serial device used in the XDM1041). Sparkfun has an article pointing to them:https://learn.sparkfun.com/tutorials/how-to-install-ch340-drivers/all
   1.3 Install the DMMEaasyControl software
2. Next, when you open up the control software, click control->connect in the top left hand corner
3. Once it is open, choose "Via COM" instead of USB. Nothing ever pops up for USB
4. It should pop up a choice like "ASRL3(DMM) (mine is configured to be COM port 3)
5. Press ok and you should be good to go

Like other people have mentioned, software isn't the best, but it works, does data logging, ect. so hopefully this helps someone!

Also, if anyone had better luck let me know, I feel like this is a bit ridiculous

[robdejonge]

The form factor of this dmm is what makes it appealing to me. Like some others, I have a small shelf right over my work surface and it's great to be able to just glance up at my dmm and scope. A 'proper' bench dmm would move to a shelf quite a bit higher, which is the main reason I hesitate to get something like that until I can build a deeper bench.

The YouTube review videos theHWcave did were very insightful (thank you!). The voltage range issue is indeed odd, the bigger brother XDM2041 does not seem to have that issue. Other than that, and based on this thread, this seems to be a pretty decent electronics bench meter!

I do have a few questions ...

• Burden voltage: has anybody had a look at this? I mostly play with 0-5V and 0-1A stuff, so it can be a major pain. My current meter drops the voltage by 142mV at 100mA, which is not too bad.
• Software: although a Mac fanboy, I run Debian on my bench computer and so wonder if there is any software available on there which I can use to control / read this mm. Does anybody have any experience? Using software to do data logging is a major reason I'm looking at this.
• Bar graph: I like me a nice bar graph, but I believe this meter does not have one - correct? I've recently been looking at power supplies which are showing a line graph over time, but I guess I would have seen that mode if it was present. I know, could use my scope.

Might sell my current meter and order this on 11/11 if there is a good deal.

Thanks for your comments.

[kwass]

• I measured the current shunts: 500uA/5mA = 101 ohms, 50mA/500mA = 1.2 ohms, 5A/10A = 0.02 ohms.  So on the 500mA range, 100mA would be 120mV drop.
• It's pretty standard SCPI, so I'm sure you could find some software for Debian.
• Unfortunately there's no bar graph.

[robdejonge]

Thanks for checking that. This is an interesting meter indeed, with a relatively low burden voltage like that!

[Kleinstein]

The burden voltage for the 5 and 500 mA range is not low - it is more on the high side. The burden is low for the lower of the 2 ranges. The 20 mOhms for the 5/10A range is low - it this range with a fuse. Quite often a proper fuse has a comparable or even higher resistance. So part of the burden in the 500 mA range is the fuse and with this in mind the value is not that bad.

For a low cost meter it still makes sense to have few shunts and share one of 2 ranges each.

[robdejonge]

Hm, okay.

So the way I look at this is that the maximum burden voltage would be 660mV. My existing meter is a Holdpeak HP-770D and has a burden voltage of 142mV @ 100mA and as a 42,000 count meter would peak at 596mV burden voltage. I'm happy to be corrected, because I've been wondering if this is the right way of calculating that.

In this overview, only the EEVblog meter has a max burden voltage that is impressively low. Not pushing back on your statement, just explaining why this novice thought 500-600mV was low.

[robdejonge]

The Holdpeak HP-770D that I currently use is sort of ok for me. Reviews seemed to indicate it was within spec, and performing quite ok. And I've been happy with it until now.

The reason I'm looking at the XDM1041 is not so much for added accuracy (it does have smaller tolerances than the Holdpeak, and a slightly higher count), but more so because 1) it is a bench meter, I wouldn't be carrying it around; and 2) it supports data logging, which would let me monitor things over longer periods of time.

I'd try to sell my Holdpeak, although the market here has lots of cheap multimeters so I'm not sure I'd be able to.

Generally speaking, I can't seem to make up my mind between "it's worth spending $100 on this new meter" and "you don't need this new meter" 

So if anyone has arguments either way, please do share!

[HKJ]

So the way I look at this is that the maximum burden voltage would be 660mV. My existing meter is a Holdpeak HP-770D and has a burden voltage of 142mV @ 100mA and as a 42,000 count meter would peak at 596mV burden voltage. I'm happy to be corrected, because I've been wondering if this is the right way of calculating that.

Typically handheld DMM's uses 100ohm, 1ohm & 0.01ohm as current shunts with two ranges for each shunt and the high uA & mA range will have a voltage drop across the resistor similar to the high mV range. For the 1ohm & 0.01ohm there is also the fuse to account for, it will about double the resistance for the 1ohm shunt, i.e. the burden voltage will typically be about twice the high mV range (This means about 1.2V for a 6000 count meter at 600mA).
A few meters are different either using lower shunt values or moving the high mA range to the 0.01ohm shunt (GW121)

2) it supports data logging, which would let me monitor things over longer periods of time.

If you want to log on a computer, try checking TestController, it support a lot of different meters (Not XDM1041 yet). There is a long thread about it here: https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/

[Kleinstein]

The burden drop is often the same as the lowest mV range, as the mV range is used to measure the votlage at the shunt. With the older 2000/20000 count meters that was 200 mV. The meters with a 500/600 mV  range as there lowest have slight problem here. The burden would be quite hight an in addition the heat produces with 600 mA and 600 mV is already quite a lot.
In analog times the internal meter movement often got away with even less and a burden voltage of some 100 mV was more typical.

[HKJ]

The burden drop is often the same as the lowest mV range, as the mV range is used to measure the votlage at the shunt.ore typical.

No, it is the highest mV range (200-600mV) for high uA and mA, if the meter has a low mV range (20-60mV) it is used for the low uA and mA range. If the low mv range is missing the internal x10 amplifier if not used for mV, only for current.

[theHWcave]

Have a look at my 4 part detailed review of the OWON XDM1041 . I also wrote some python software for logging (works on Linux and Windows) you can download from my Github page https://github.com/TheHWcave/OWON-XDM1041. This link also contains my documentation of the SCPI remote control commands implemented in the XDM1041 (quite comprehensive for what you pay).
The XDM1041 has its flaws. The main one is that by default its slow (but very accurate). It allows you to select faster speeds (sacrificing some accuracy) but you have to do that every time it is powered on (it doesn't remember that setting). Oh and the auto-ranging on capacitor measurement is not working well. If the XDM1041 is your only way to measure capacitance, buy another meter.  And did I mention that supporting PT100 temperature probes is well intended but useless because of the way REL mode is implemented for temperature. (K-type works better but has its own interesting quirks, watch the video or read my earlier posts on the XDM1041)...
Besides the XDM1041, I have an Agilent 34401A, Solartron 7150+ , HP3478A, BM869S, BM235 and a couple of other handhelds but the XDM1041 has become my most often used meter for repair work on the bench because its so convenient, small, and always on.

[robdejonge]

Besides the XDM1041, I have an Agilent 34401A, Solartron 7150+ , HP3478A, BM869S, BM235 and a couple of other handhelds but the XDM1041 has become my most often used meter for repair work on the bench because its so convenient, small, and always on.

That says a lot!

[robdejonge]

Why does the meter require a function switch between mV and V for the VDC and VAC measurements? 

I recently saw a comparison of the XDM1041 and XDM2041 by Tony Albus
https://youtu.be/ZVDJSa3pxrY

The big brother of the two does not have the weird "function switch" requirement. But it seems to be using the same software or at least software framework. I'm curious what would necessitate the requirement of such a switch in the software. Is there a hardware reason? Or could we (let's not hold our breath!) eventually see Owon correct this issue in a firmware update?

[theHWcave]

You are right, there isn't a hardware issue for that as far as I can see. At least for volts... It needs to do the same for current but there the reason is that A and mA/uA use different input sockets.  Could be something with the HY3131 front end? But nothing that the firmware could not fix.  Does Dave's 121GW do the same thing? It has the same front end chip.  OWON is sluggish at best responding to bug reports. What I hope is someone writing completely new software. This meter looks to be very hackable, since the front end used just an SPI interface and I think 2 discrete signals to drive relays

[Kleinstein]

The separate mV and V ranges are a bit odd. It could be just avoiding the very long list of voltage ranges, that may not fit the screen. This should be a SW only issue.

It is interesting that the reading rate via the PC interface is quite a bit higher than with the updates on screen.

For the K-type thermocouple it does not really mater if the temperature is ITS90 or a different curve - the differents are so much in the details not even the 34401 with a PT100 may notice.

[ciccio]

I received my meter form Banggood yesterday.
First impression was positive, but when I checked  temperature measurement, I discovered that readings from a know good PT100 vere off, sometimes very off.   Reading increased slowly from 21 to 160 °C
My resistance box was not available,  so I checked at three resistance values,  and it was almost correct at 0, 100, 200 ° C.
Will check later with a better decade box.
As for thermocouple, the data sheet says K type, but the meter allows to select  "LOAD"  (why Load and not Source? ) as PT100 and KIT90 TC,  which I discovered being a new designation for K type.
With a known good TC, it seemed to read correct values.
I will examine cold junction compensation as soon as I'l  have a free evening.

The software installation was a nightmare,    I will try next week on a  Windows 10 PC.
My Windows 7 work machine seems to be allergic to   N.I. VISA32 installation.
I had the same problem with the software for a  JUNTEK function generator.
Best regards