Author Topic: Measure high Voltage 500V DC and Current in a floating system  (Read 2115 times)

0 Members and 1 Guest are viewing this topic.

Offline mjoelnerTopic starter

  • Newbie
  • Posts: 7
  • Country: no
Measure high Voltage 500V DC and Current in a floating system
« on: January 27, 2020, 01:35:59 pm »
Hi guys and girls.

I have a power supply of 480V DC that powers some equipment (quite steadily and boring, no signals or exciting stuff) that pulls 0.8 amps. I wanted to do some measurements on this and I thought it was going to be very simple, but oh no, not so much. I would like to measure the voltage and the current the gear is consuming using picoscope, because its nice to see what is going on.

Since 500 volts is quite high I started with reading up on people that have tried the same, and I quickly found that first of all there was a good chance of frying the gear and a differential probe was the way to go, so I bought two of the chinese ones that everyone loves because of the low price.

My idea was to just use the differential probe directly across the 500VDC source, and that worked like a charm. My next idea was to put two resistors in series with the equipment, one for each line, thinking both of them would consume some volts. I would then be able to measure the voltage, and that voltage across one of the 4 ohm resistors should also give me a pretty good idea what the current in the system is doing.

When measuring a handheld scope using one 1x probe, it works pretty good I have to say. But, seeing that I want to save this and use as reference later on, the picoscope is what I really wanted to use.

Using a differential probe actually worked pretty good, but I thought it was a bit to noisy at 50x. One solution then would be to up the resistance a bit, but I cant steal to much of the voltage either. So, I thought about using the a-b differential method using two probes connected together and then measure across the resistor. A calculation in Picoscpoe would then give me the DC voltage across the resistor, and I was hoping to see the same as the handheld oscilloscope. Unfortunately that did not happen. When I connected the probes across the 4 ohm resistor, they both gave me a channel overrange alarm and the numbers made no sense whatsoever. The laptop and the picoscope is not grounded or connected to power anywhere.

Any idea what is causing this, and better, is this what is expected or can I adjust my experiment to give me better values.

I have a picoscope with 4 inputs. Two differential probes 50x/500x. 2x 4 ohm resistors, or even 2x 10 ohm resistors. The goal is just to be able to read the amps in my circuit, and I dont have much fancy equipment to be able to do that.

Thanks for any inputs on this.

 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2135
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #1 on: January 27, 2020, 04:07:15 pm »
If the differential probes are properly designed/built I can't see why this wouldn't work. Might be worth checking their isolation to scope ground.

The noise thing is an issue with the way diff. probes work and their need for high attenuation. Is this a one off measurement or something you need to do on a regular basis?
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 
The following users thanked this post: mjoelner

Offline mjoelnerTopic starter

  • Newbie
  • Posts: 7
  • Country: no
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #2 on: January 27, 2020, 04:19:16 pm »
Hi yes, the differential probes are the chinese 150 USD ones that everyone on a low budget buys nowadays from micsig.

If the measurements give me some usefull results, then I may do this on a regular basis.

I see now that I was a bit unclear. When I used the "A-B differential method" I used the standard probes on the picoscope. It was the standard probes that got overloaded.

Thanks
 

Offline Ice-Tea

  • Super Contributor
  • ***
  • Posts: 3229
  • Country: be
    • Freelance Hardware Engineer
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #3 on: January 27, 2020, 04:37:55 pm »
So, eh, you have two regular probes hooked up to 500VDC power supply? Maybe you should draw exactly what/where you're measuring. Seems to me like you're not taking into account the grounding of the probe or the max voltage of the inputs..

 
The following users thanked this post: mjoelner

Offline mjoelnerTopic starter

  • Newbie
  • Posts: 7
  • Country: no
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #4 on: January 27, 2020, 04:58:11 pm »
hehe, yes, you are right, there might be some issues with this approach.

I have two probes, channel A and Channel B connected to each side of a 4 ohms resistor.  This resistor has 0.8 amps going through it. I would expect to measure a bit less than 4 volts in this setup, but I get channel overrange on both probes.  This resistor is in series with the equipment I want to measure. The power supply is almost 500 volts DC. The ground on these two probes are connected to themselves only, nothing else. The picoscope is powered by a laptop, and the laptop is only running on battery. There is no metal parts touching any other metal parts anywhere. As far as I can understand nothing is grounded. Even if it was grounded, from what I understand there is a 1 Mohm resistor in series in both the two probes, so there would not flow much dangerous current through in any case. If I did connect the ground cables anywhere, I would fry the equipment, yes. 

https://www.picotech.com/images/uploads/library/topics/_med/pseudo-differential.png
This picture resembles pretty much what I am trying to do with two standard probes.

 

Offline Ice-Tea

  • Super Contributor
  • ***
  • Posts: 3229
  • Country: be
    • Freelance Hardware Engineer
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #5 on: January 27, 2020, 05:08:37 pm »
If your crocks are indeed floating and laptop/Pico/everything else is also floating then your measurement will never make any sense. A voltage is always measured between two points, ie the tip of the probe and the crock. If the crock is truly floating, it means nothing.

Comparable: take your DMM and measure the voltage of a battery with only one terminal connected.
 
The following users thanked this post: mjoelner

Offline mjoelnerTopic starter

  • Newbie
  • Posts: 7
  • Country: no
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #6 on: January 27, 2020, 06:03:21 pm »
There is two ordinary probes touching each side of the 4 ohms resistor. I may have misunderstood the whole "pseudo differential" concept.

I tried your experiment, without connecting the crocs to anything, and measuring only using the two probes on a 3.2V battery, and it shows up as 3.2 volts on the picoscope. 

 

Offline Ice-Tea

  • Super Contributor
  • ***
  • Posts: 3229
  • Country: be
    • Freelance Hardware Engineer
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #7 on: January 27, 2020, 08:08:07 pm »
A pseudo-differentlial measurement is just that: pseudo. Pretend. It still measures the actual voltage between tip and crock on both channels and substracts it with simple math.
 
The following users thanked this post: mjoelner

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2135
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #8 on: January 27, 2020, 08:57:24 pm »
If you can truly float your whole picoscope and laptop, then you could measure across a shunt shunt resistor with a standard scope probe.  But I'd be super cautious about this as your whole setup is then at some potential.

If I *had* to do the measurement then I'd probably:
1. insert a shunt resistor in the negative rail
2. measure the voltage across this resistor with the scope ground on the psu -ve side of the resistor, and the ch1 scope probe on the load side of the shunt resistor.
3. build a potential divider off the 480V (3x1M resistor from +ve to ch1. and then 30K from ch2 to scope ground (-ve) gives 100:1) for ch2.

Again, your whole scope/laptop will now float at psu -ve, so could bite! Be ultra-careful or better don't do it.
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 
The following users thanked this post: mjoelner

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7681
  • Country: ca
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #9 on: January 27, 2020, 09:33:29 pm »
If the Picoscope/laptop are not earth-grounded, there is can be a shock hazard or noise problem due to the differential probes inherently having some leakage currents. The Picoscope/laptop can float up to the HV, as most low cost differential probes are simply resistive voltage dividers inside, they don't actually offer full isolation. Never get this mixed up or confused.

Also, if you have two probes connected (one to monitor voltage, other for current), they could clash. Easy way to see how bad it is, measure voltage with a multimeter between the Picoscope/laptop and a nearby real earth-ground. Likely it is high enough to cause the probes to malfunction (too much common mode voltage).
 
The following users thanked this post: mjoelner

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #10 on: January 28, 2020, 02:43:46 am »
Using a differential probe actually worked pretty good, but I thought it was a bit to noisy at 50x. One solution then would be to up the resistance a bit, but I cant steal to much of the voltage either. So, I thought about using the a-b differential method using two probes connected together and then measure across the resistor. A calculation in Picoscpoe would then give me the DC voltage across the resistor, and I was hoping to see the same as the handheld oscilloscope. Unfortunately that did not happen. When I connected the probes across the 4 ohm resistor, they both gave me a channel overrange alarm and the numbers made no sense whatsoever. The laptop and the picoscope is not grounded or connected to power anywhere.

Any idea what is causing this, and better, is this what is expected or can I adjust my experiment to give me better values.

You exceeded the common mode input voltage of your hacked together differential probe.  The difference is the measurement that you want however each probe and input channel sees the total voltage before subtraction.

The high voltage differential probes have enough attenuation on each input to bring any common mode voltage into range and it is this attenuation which also result in a noisy differential signal at high gains.
 
The following users thanked this post: mjoelner

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7681
  • Country: ca
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #11 on: January 28, 2020, 03:19:15 am »
Am I right thinking a differential probe can be useless and even dangerous if it has no ground reference? Then there is no common-mode attenuation possible by the input voltage divider, as seen in this commercial HV differential probe schematic I drew, with the CMRR trimpot center-point floating. Differentially, yes there is attenuation. But common-mode the TVS will just clamp and the return path is to the Picoscope/laptop.
I keep thinking the confusion is these DP's have isolation or work when completely floating. They have a need for a CM return path to give them CMRR.

I would try again with OP's setup but earth-ground the Picoscope. Using (the) three-prong laptop power adapter could suffice, but not a two-prong one.
We don't know much about the noise, CM voltage, the load etc. in the "floating system".
 
The following users thanked this post: mjoelner

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #12 on: January 28, 2020, 04:06:01 am »
Am I right thinking a differential probe can be useless and even dangerous if it has no ground reference? Then there is no common-mode attenuation possible by the input voltage divider, as seen in this commercial HV differential probe schematic I drew, with the CMRR trimpot center-point floating. Differentially, yes there is attenuation. But common-mode the TVS will just clamp and the return path is to the Picoscope/laptop.

That is correct; a high voltage differential probe cannot be safely used in an isolated setup.  If it is, then the isolated side will charge up to the common mode voltage.  Older high voltage differential probes included an explicit ground (really common) connection as part of their high voltage differential input.  On newer probes without this, the ground (common) of the output signal to the oscilloscope is the same thing.

Tektronix includes a warning with their differential probes and oscilloscopes with isolated inputs about using them together.
 
The following users thanked this post: floobydust, mjoelner

Offline mjoelnerTopic starter

  • Newbie
  • Posts: 7
  • Country: no
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #13 on: January 28, 2020, 06:18:51 am »
So, it seems its not that safe to measure the voltage and amps in this setup the way I intended.

Is there any way this could be done safely then, where I and my equipment can safely esacpe without to much harm. Just so it is said, the equipment has very sensitive GFI and trips when its sensing 10mA. So far in my tests it has not tripped at least. I power down and up the gear every time I mess with the cabling.


For a bit more clarity I have attached a file where I draw up both two experiments I have tried this far.
This is the differential probes I have used:
https://www.amazon.com/Micsig-DP10013-Differential-Attenuation-Tektronix/dp/B074K4XPW3
On here they claim: Micsig DP10013 High Voltage Differential Probe is designed for professionals and experts like you to be used with most oscilloscopes in the market.
It allows you to safely measure floating voltages up to 1,300V and has a bandwidth up to 100 MHz.
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7681
  • Country: ca
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #14 on: January 28, 2020, 06:51:02 am »
DP10013 "allows conventional earth-grounded oscilloscopes to be used for floating signal measurements of up to 1300 V of differential voltage and common mode voltage."
The keyword is earth-grounded, so your Picoscope needs to be grounded- not floating along with the laptop. Careful with cheap products with no user manual or safety instructions.

As far as measuring current, the probe attenuates by 50x so a 4R sense resistor 0.8A becomes 64mV plus the probe's 40mV of input-referred noise.  It does not look promising.
 
The following users thanked this post: mjoelner

Offline mjoelnerTopic starter

  • Newbie
  • Posts: 7
  • Country: no
Re: Measure high Voltage 500V DC and Current in a floating system
« Reply #15 on: January 28, 2020, 07:13:07 am »
Yes, floobydust.

Thanks for that input (to all of you really, it has been very helpfull this far). The scope can easily be grounded, so I will definately do that from now on.

I noticed the noise issues when I did the measurement, and I had a feeling it would be very noisy with such a small voltage over the resistor. This is why I decided to try the pseudodifferential method. If I grounded the scope, and used the pseudo differential method, would I also then put myself and the gear in harms way? I am starting to think I should go very low with the voltage and see where it takes me when I slowly turn it up. That might maybe give me some insight, (that some of you probably have told me allready but I did not fully understand)

I could help it a bit by using a bit bigger resistor. I have a couple, so I could potentially put a couple in series. 
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf