EEVblog #933 - Keysight U1272A EMC Issue

[EEVblog]

A viewer found an EMC issue with the Keysight U1272A multimeter
Dave confirms both conducted and coupled issues and finds some spooky action at a distance!
Comparisons with other meters confirm they don't have the same issue.

[Barny]

That explains the funny readings at my bench, when my mill is running or the LED - benchlight is running.
I always thought some of the EM-crap is inducting into the unshielded wires.

Thanks for the information.

[riyadh144]

Okay, I am probably wrong but from the proximity when you were moving the wire around and from the data sheet the frequency response and the THD, I think MAX4611 is the problem, and there is probably an antenna below it.

https://datasheets.maximintegrated.com/en/ds/MAX4610-MAX4612.pdf

[station240]

Disappointed Dave didn't whip the cover off to figure out which component is most sensitive to EMC.

[EEVblog]

That explains the funny readings at my bench, when my mill is running or the LED - benchlight is running.
I always thought some of the EM-crap is inducting into the unshielded wires.

Interesting, can you reproduce it?

[EEVblog]

Disappointed Dave didn't whip the cover off to figure out which component is most sensitive to EMC.

That's a whole different video. You can get lost down a rabbit hole with that.

[PartialDischarge]

I thought this kind of thing only happened to siglent/rigol (being sarcastic...). My Extech MG302 its not affected either

And this is why nowadays you can't say brand X is the best/worst. Instead say subsystem Z of the model Y from brand X is really good/bad. This applies especially to cars

[EEVblog]

I thought this kind of thing only happened to siglent/rigol (beign sarcastic...). My Extech MG302 its not affected either

The GSM flaw was in the Fluke 87V for a decade before someone found it.

[Barny]

That explains the funny readings at my bench, when my mill is running or the LED - benchlight is running.
I always thought some of the EM-crap is inducting into the unshielded wires.

Interesting, can you reproduce it?

My DMM don't show that stable value like yours when influenced by EM.
The value is dancing around by several 100mA in a random pattern.

The LED light is a simple 12V LED-strip glued on the shelve, powered by an cheap switching suply.
The distance to the DMM is ca. 1m.

The mill is powered by a 2 KW HF spindle at 400VAC at maximal 400Hz Sinus
(I don't knew the internal switching frequency of the U/f converter.)

The axes are driven with 46V stepper motors.
The distance to the DMM is ca. 3m to 4m.
I thought this readings are normal in this dirty environment.



At the moment I sadly have no other DMM to double check.
My cheap arse fluke branded one is slowly dying on me after 15 years. (Its build quality is quite 10€ china-like)

I thought about a new EEVblog-Brymen one as a replacement, but sadly I can't get one on Amazon (Europe) and the shipping costs from Australia are quite high.

[FrankBuss]

I guess the engineer who signed the EMC test report is getting nervous now. This is the standard:

http://www.etsemc.co.uk/wp-content/uploads/EN61326-1.pdf

For the basic enclosure immunity requirements it says for EM radiated field: 80 MHz - 1 GHz, 3 V/m (I guess lower frequencies are not that important, because e.g. at 5 MHz the wavelength is already 60 m?). Industrial immunity requirement is even 10 V/m. I have no idea, is this high for an EM field? Can anyone do a back-of-the-envelope calculation how strong is the field of the harmonics at these frequencies with Dave's 5 MHz signal, using just a wire as an antenna?

[HighVoltage]

Nice video!
Confirmed on my U1272A and the problem persist when you also insert the ground lead from the generator to the COM connector.

[PA0PBZ]

My 34461A is somewhat nervous, but like you said it's not much
My U1241B doesn't care at all.

[madires]

This is disturbing. A while ago the grey beards told the young engineers that you can't go wrong with HP. And now we have to wait a few years for the early buyers to do the beta testing?

[Faith]

Wow, was just about to get one of these a couple of days ago. I guess that plan is out of the window now >,<"...

And to think that the U1272A has been released for almost five years already you'd imagine that such a big issue would have been resolved by now.

Wonder how many people have been affected by this without realising it.

Would love to see what Keysight has to say about this.

That'll be the real test!~

[isb]

I can reproduce this on U1242C but only when sig gen is connected. Couldn't find any orientation where radiated emi would mess with the current measurements.

Using 5 MHz, 10 Vpp meter shows -24 mA on mA range and -90 mA on A range. 

[flodins]

Hi

I can test this meter in my EMC lab, in GTEM cell https://en.wikipedia.org/wiki/GTEM_cell for radiated immunity (to 10V/m)and also conducted immunity on the testing bench with LISN couplers. I can provide precise graphs showing where meter is doing weird stuff. From 150k to 3GHz with AM and FM modulation.  Problem is I don’t have one, so anyone interested can send it to me for testing. I have only Brymens in my company.

Also I can perform all tests from this list http://www.etsemc.co.uk/wp-content/uploads/EN61326-1.pdf

[128er]

My U1273A (S/N MY531100XX) behaves the same way. Tested with the function gen on my DSO-X2022A.

And I saw that my warranty expires in 3 days.

[electr_peter]

I confirmed issue with U1272A. DC current measurement is only slightly changing when HF source is near it, so I do not see as very significant problem. When current jack is directly connected, currents of 1-5A were shown on LCD. Tested with 2.5V peak-to-peak, 1-20MHz, sine and square wave.

Problem has everything to do with common mode currents and parasitic capacitances. Thus it may or may not manifest itself depending how well everything around meter is grounded.

Such problem should be fixed nevertheless - it can influence readings without one knowing. Keysight, we are looking for a response.

[splin]

I see that the input impedance in DC modes is not actually specified in the datasheet or user guide - sloppy work Keysight. It's only specified as 10Mohms for AC and AC+DC modes, low impedance mode and powered off state - both 1.67Kohms. I see that electr_peter here did measure it at 11M ohms which is to be expected.

I wonder how many users have been bitten by the low impedance when the meter is switched off; it appears to me to be another design error.

[serby]

I tested my good old Agilent U1241B and it is rock solid. No sign of the problem

[Berni]

Tested on my U1273A and yes it happens.

One thing i noticed it that its worse if i touch it as shown in the photo.

[electr_peter]

I also noticed selective susceptibility around DMM (near sides and . Touching DMM with hand changes value a bit as well, as others mentioned (due to parasitic capacitance).

I guess that teardown and live testing is in order. Pretty tricky to do, because LCD is on front side of PCB while most of the circuitry is on the backside.

[calexanian]

I thought this kind of thing only happened to siglent/rigol (beign sarcastic...). My Extech MG302 its not affected either

The GSM flaw was in the Fluke 87V for a decade before someone found it.

When the 87's were designed only stock brokers had cell phones! GSM was not really even a thing yet as that did not hit the US until the mid 90's. It was still a euro thing until then. a quick google search reveals that GSM was introduced in europe in 91.  We got our first 87 true RMS meter here in about 89 for a special project that required such a fancy meter. It was brand new at the time and cost more than what we sold our Oldsmobile for.

(Edit) it may have been an 85 we got. cant remember

[amitchell]

Any results with the 1250 series?

[Kleinstein]

The wire on top of the meter is already quite a strong field: much of the field should be in the few cm to the meter. So this is more like 1000 V/m compared to the 3 or 10 V/m test requirements. Still even the lower field showed some influence.

The EMI Problem is no unique to this DMM. I have an old analog meter (batteries just for ohms), that also gets crazy with RF voltages. Something like 5 V at 100 kHz gave a full scale reading on essentially all ranges, including 1200 V DC. Checking a switched mode regulator can get really scary with this.