Bench DMM isolation

[nedeb]

Hi, I'm considering getting a bench DMM (e.g. for data logging) and until now I've only used handhelds. With handhelds I don't have to worry about ground loops when making a measurement. What about bench DMMs? Are the inputs isolated or do I need to worry the same as with an osciloscope?

[ledtester]

A bench DMM is usually powered by a transformer which allows the probe inputs to float.

See, for instance, the inside of the ET3240 at 5:00 in this video:

https://youtu.be/QXfZaBO2OBg?t=5m

If a DMM has a digital interface (GPIB, RS-232, ethernet, USB, etc.) that interface will be isolated from the probes.

btw, what bench DMM are you considering getting?

[bdunham7]

Unless it is an antique or some specialty instrument, the inputs will be isolated from ground with limitations on how much voltage can be present between each terminal and ground.  This will be clearly indicated on the front face and the maximum voltages can vary quite a bit between models.  If you aren't measuring higher voltages, then this probably isn't an issue.

[nedeb]

Thanks for the informative replies, that sounds good then!

I was thinking of the SDM3045X but I see now that the DCA range is only 600µA-10A. I'm a bit surprised. The 600µA is too high for some applications I have in mind (e.g. logging of parasitic drain). Now I'm thinking of getting the handheld BM869s instead (it advertises a DCA range of 0.01µA-10A) and I hope can still log 5 readings/s (much less than the 150 readings/s of the SDM3045X but at least it should work in all the ranges I need).

[bdunham7]

I was thinking of the SDM3045X but I see now that the DCA range is only 600µA-10A. I'm a bit surprised. The 600µA is too high for some applications I have in mind (e.g. logging of parasitic drain). Now I'm thinking of getting the handheld BM869s instead (it advertises a DCA range of 0.01µA-10A) and I hope can still log 5 readings/s (much less than the 150 readings/s of the SDM3045X but at least it should work in all the ranges I need).

Comparing advertised specifications between models from different manufacturers can lead you astray, as shown here.  The Brymen 0.01µA spec reflects the single-digit resolution of the 500.00µA range, which is the smallest range it has.  The 600.00 µA spec for the SDM3045X is simply listing its lowest range, and in fact it has the exact same single-digit resolution of 0.01µA.  Keep in mind that neither meter can actually measure 0.01µA accurately.

[Doctorandus_P]

The input ranges are probably not far apart. Very likely the 600uA of the Siglent is the full range on the most sensitive range, while the specifications of the Brymen state the resolution.

I have also been looking at bench top DMM's but I find it very annoying that they almost all come in very deep cabinets, and I have no room for that. In the "lower cost" range there are also handheld DMM's disguised as benchtop DMM's, but if it's still a disguised handheld DMM you might as well buy a real handheld meter. And therefore my main interest is also to this Brymen.

The Brymen's have a weird, but documented communciation protocol They basically send the LCD segments over the optically isolated connection in RS232 format.

There are some handheld DMM's with internal Bluetooth. Some Chinese meters, some Agilent meters have it, and I think also the 121GW.

Dave's 121GW also has a uSD socket, so it can log continuously until the bateries die.

=====================
Also, It's always nice to have multiple meters, and it's quite easy to make a simple meter yourself with a uC, some ADC and a handful of passives, and adding a serial connection is also easy. Therefore I've got the idea to make a few simple meters (some that can only measure voltage, and others that can only measure current), then calibrate them with a Brymen (that's the best I'm willing to afford) The homebrew meters can have a much higher sampling rate, mostly dependent on the ADC you use.

[Romualds]

Actually you don't need desktop multimeter for data logging. There are several handheld MMs with possibility to connect to PC using isolated serial interface or wirelessly via Bluetooth.

[nedeb]

Ah thanks for the correction regarding the respective ranges! It makes more sense now.

Yes the 121GW also has Bluetooth LE which could be nice for continuous logging (though maybe trickier to get working reliably than Ethernet as offered by the SDM3045X). Apparently the the lowest DCA range is 1nA ±1.5%+15 which means "+ 15 digits" according to https://www.eevblog.com/wp-content/uploads/2017/11/121GW-Manual-5.pdf. Not sure what this means?

I've also been looking at an Agilent/Keysight U1272A or similar but second hand (given the price) and haven't found a good deal so far.

As for a homebrew meter, that sounds like a nice project but I suppose it would be hard to make one with accuracy in the same ball park as the BM869s?

[Kleinstein]

As for a homebrew meter, that sounds like a nice project but I suppose it would be hard to make one with accuracy in the same ball park as the BM869s?

When limiting the meter to just the range needed and maybe a few small extras that are easy and don't interfere it may not be so difficult to get good accuracy. Especially the lower current ranges can suffer from compromises like the protections for a high current fault or power constraints in a battery powered meter. The BM869 sees to use the same shunt for the 500 µA and 5 mA range - this has a negative effect on accuracy. The current accuracy is not really a strength of the BM869. Using relatively few shunts is a common point with many lower cost or handheld meters so safe costs and switches.

The moderately low current range like 100 µA is amoung the easier ones: not much heat at the resistor (and thus a TIA viable) and still not super critical with leakage in pA range.  A specialized lower current meter (e.g. ranges for 200 µA, 20 µA and 2 µA) would not be a very difficult build. One would still have the problem with a calibration and this would need something considerably more accurate for the calibration.

[wizard69]

Ah thanks for the correction regarding the respective ranges! It makes more sense now.

Yes the 121GW also has Bluetooth LE which could be nice for continuous logging (though maybe trickier to get working reliably than Ethernet as offered by the SDM3045X). Apparently the the lowest DCA range is 1nA ±1.5%+15 which means "+ 15 digits" according to https://www.eevblog.com/wp-content/uploads/2017/11/121GW-Manual-5.pdf. Not sure what this means?

Personally I would not use most handhelds for data collection.  There are just too many limitations.

I've also been looking at an Agilent/Keysight U1272A or similar but second hand (given the price) and haven't found a good deal so far.

As for a homebrew meter, that sounds like a nice project but I suppose it would be hard to make one with accuracy in the same ball park as the BM869s?

I’m not sure what you are trying do but D/A cards/dongles are a wonderful thing for data collection.   Just don’t buy hardware with extreme licensing controls for software.   Two good  reasons to go this route, 1.  storage space as large as you want.  2.  Often the software will be simpler.