Fluke Scopemeter - overcautious or safety violation? 2

[Skogsgurra]

I am having an ongoing conversation with Fluke about certain aspects of the Series II scopemeters.

One of the aspects is what voltage limits that apply to the isolated inputs. Especially between what Fluke calls 'Probe reference' which is the same thing as the ground clip (or 'cold' side) of the probe and real (earth) ground. There is a diagram in the manual saying that voltage between probe references and and between probe references and real ground must not exceed 30 V when the signal frequency is more than 25 kHz.

I can understand that specification. That is why I always use an isolation amplifier or differential probe when measuring high-power VFDs with PWM inverters.

Fluke top technicians say that I can use the inputs directly (with the 10x probes), without any extra isolation "because the voltage above 1 kHz will be very low - a few volts maximum".

I do not agree. The derating curve and actual measured data can be seen here:

http://gke.org/pub/files/Fluke Scopemeter Series II maximum voltage between probes 1.pdf

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[itsmoked]

I don't even see the logic of, "only a few volts a xHz".

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

This is a snip from the answer I got: "Og spenningskomponenten ved 1kHz på et signal fra en frekvensomformer er normalt langt under 1kV, bare noen få volt"

It is in Norwegian. Run it through a translation program.

Word list: Bare=only, Noen få=a few

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

What is the basis (explanation) for the published voltage limit that decreases with frequency?

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

I cannot find more than what the picture says. The manual is available for download. I haven't been able to find any explanation.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

The curves are labeled Cat III and Cat IV... has something to do with explosive environments?

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

No Pete. That's quite another thing and does not have much with this question to do.

Cat IV means that you are measuring on transformer secondary or service entrance where you can ecpect transients up to 8 kV while Cat III means that you are measuring inside a building, after disconnects or MCC or VFD and possible/expected transients are lower.

When you measure on an LV motor on a VFD, it is always Cat III.

Read

http://gke.org/pub/files/Fluke Scopemeter Series II maximum voltage between probes 1.pdf

again. The question is in there.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Skogsgurra]

The manual is here:

http://www.tequipment.net/pdf/Fluke/190-204_UsersManual.pdf

Page 112.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

Your measurment shows 270 volts at 31khz. Using the 10:1 probe, would it be 27 volts seen at the instrument (below the 30 volt limit)?

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

No. I used the DP120 differential probe. And I did set the probe scale factor to 200:1 - which is the attenuation you get with the DP120.

As you can see, the PWM waveform (the screen dump) has the normal look with PWM pulses going from 0 V to DC link voltage, which is close to 600 V (400*sqrt(2)). The grid was a bit on the high side.

The idea with the post is to discuss if Fluke are overly prudent using this derating curve, where no more than 30 V RMS is allowed at frequencies above 25 kHz or if they are unaware of the fact that PWM motor voltages are as high as they are. The latter seems to be the case.

Remember that they told me that the voltage components above 1 kHz are only a few volts while the truth is that they are several hundred volts.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

On top of page 103, it lists a voltage specification 300 volts direct connected and 1000 volts for 10:1 and continues "for detailed specs, see page 111"

On page 111 is the figure you mentioned. At low frequency it lists 300 volts, that would seem to correspond to the limit when using direct connection. For 10:1, presumably the curve is somewhat higher? (And even higher for 100:1 attentuation probes?)

I'm just trying to understand what the limits are, if not why. I agree it does not seem to be well explained whatever is going on with this limit.


=====================================
(2B)+(2B)' ?

 

[electricpete]

I imagine that the 10:1 probe is a simple 10:1 resistive voltage divider (using very high resistances to keep high input impedance).

In that case, as a first guess, we'd expect the voltage limits are 10 times higher using 10:1 probes, assuming the limit is based on something indside the box. I can't quite reconcile why they mentioned a 3:1 ratio in limits (1000 volts with 10:1 and 300 voltage direct).... maybe the probe becomes limiting. One would think it would be easier to understand from reading the manual than it is.

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

Pete. Forget about the probes.

The spec is voltage between probe grounds (inputs are isolated) and between probe grounds and 'real' ground. That is where the problem is.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

ok, I missed that part.

So the problem arises because you're measuring phase to phase voltages? I'd think you could measure phase to ground voltages and use that to compute phase to phase voltages (if needed), in which case all references would be ground. Anything wrong with that approach?

=====================================
(2B)+(2B)' ?

 

[electricpete]

Page 30 of the user manual discusses the waveform arithmetic available right in the box, including subtraction.

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

Yes, Pete. Done that, been there. First time back in the sixties (it was thyristor controllers and DC motors then). That is not what I have a problem with. All motor voltage measurements are done phase-phase.

It is the fact that 1) Fluke has the derating curve and that does such measurements impossible and 2) that Fluke doesn't realize that it is a problem because they think that the voltages are 'a few volts' while they are, in fact, several hundred volts, that I try to discuss in this thread.

That waveform arithmetic is of very limited use when rise time is in the 200 ns range. Even Fluke realizes that. Numbers below are from the manual, page 103:

AC or DC coupled, high frequencies:
60 Hz to 20 kHz ......................... ±(2.5 % + 15 counts)
20 kHz to 1 MHz ........................... ±(5 % + 20 counts)
1 MHz to 25 MHz........................ ±(10 % + 20 counts)
For higher frequencies the instrument’s frequency roll
off starts affecting accuracy.



Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

Is that related to subtraction of two channels, or just a limit of the instrument?

I'd think if you are measuring phase to ground voltages of most PWM drives, when Va-g has a steep ramp then Vb-g is relatively constant during that period of time. So I wouldn't think measuring Va-b during that period would be much less accurate than computing Vab=Vag-Vbg from measured Vag, Vbg

=====================================
(2B)+(2B)' ?

 

[Skogsgurra]

We can discuss that in another thread.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Marke]

Gunnar
I agree with you, I would be very loath to use this without a differential input. Just not safe at all.
In some applications with the ringing voltages present, you can expect 1KV+ at high frequencies.
The problem is not that the xxx harmonic has an amplitude of vvv, but the arithmetic sum of all the harmonics present and coincident creates a dv/dt of many hundreds of volts per uS and this is what causes the potential breakdown with major amps energy behind it.
It is enough to blow the insulation in motors and cables if they are rated at 800V.
So much equipment is designed today by engineers and marketing people who have never been in the field.
Tweak it to work in the lab, cook the specs to suit what is easy to achieve and send it to market, then spend years fixing the problems or quoting the manuals. The question needs to be asked before release, "is it fit for purpose" and "what is the purpose?"

Mark Empson
Advanced Motor Control Ltd

 

[Skogsgurra]

Thanks Mark!

That is exactly what I have tried to make clear in this thread.

I agree that the sum (i.e. the square wave and its ringing) is the actual problem. But it is not easy to explain that to the desk officers at Fluke. All I can do is to use their own data. And that derating curve on page 112 is clear enough.

You did check out

http://gke.org/pub/files/Fluke Scopemeter Series II maximum voltage between probes 1.pdf

?



Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.