Author Topic: High voltage oscilloscope probe  (Read 3421 times)

0 Members and 1 Guest are viewing this topic.

Offline EugenioNTopic starter

  • Contributor
  • Posts: 36
High voltage oscilloscope probe
« on: January 16, 2023, 10:11:01 am »
Hi all!

I'm looking for an oscilloscope probe for high voltage pulse measurements. High voltage means 50kV minimum, with reasonably high probability of 80-90kV excursions. Bandwidth should be around 1MHz, but higher is welcome. This should be good enough for showing high voltage pulses with ~5kV/us rise time, and duration less than 100us.
No need for differential probing, single ended is fine.
Passive probes cannot reach the bandwidth, so I think I need to go for an active one.
I don't like those manufacturer tied gadgets with fancy smart connectors. Just plain probe with its wall wart powered amplifier.

Do you have any suggestion?
 

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3656
  • Country: fr
  • Analog, magnetics, Power, HV, Audio, Cinema
    • IEEE Spectrum
Re: High voltage oscilloscope probe
« Reply #1 on: January 16, 2023, 11:37:02 am »
Bonjour we specialized in HV , EBY, ignitor and transient protection design in 1970s...1990s.

What is the appkication, eg transient research, compliance testing, EMC, EW, radar?

We have used many of these Tektronix HV probes. P6015,   P6015A, see specs. 20..40 kV 75 MHz derating depends on freq.

Current version P6015A: https://w140.com/tekwiki/images/b/b2/51W-8217-2.pdf

P6015 earlier version has larger comp box, requires Freon 114 gas fill, sale of the Freon was  banned long ago. Any used eg epay will have the F114 gas missing or already depleted so derated to 20 kV
https://w140.com/tekwiki/wiki/P6015

Above 40 kV or the derated spec of P6015, you have some probes (see below) but more are  HV dividers:  large terminals with precision vacuum or air dielectric capacitors (AC only) or high meg HV dividers.


Few vendors, like

Ross Engineering

https://www.rossengineeringcorp.com/products/measurement/hv-probes-multimeters/hv-probes.html

https://www.rossengineeringcorp.com/products/measurement/hv-voltage-dividers.html

Stangenes Industries

https://www.electricalproducts.stangenes.com/viewitems/products/standard-voltage-dividers

Expect high cost and long delivery as most of these are made to order.

I can vouch that both my old friends Ross and Stangenes are very experienced high quality manufacturers


Enjoy and be safe!

Jon

An Internet Dinosaur...
 
The following users thanked this post: EugenioN

Offline EugenioNTopic starter

  • Contributor
  • Posts: 36
Re: High voltage oscilloscope probe
« Reply #2 on: January 16, 2023, 01:04:28 pm »
What is the appkication, eg transient research, compliance testing, EMC, EW, radar?

I need to build from the ground up some knowledge in HV generators. The main goal is to design and develop an high voltage pulse generator, that's basically very close to an automotive ignition coil and driver (with the exception of the spark plug), plus some ADCs to measure primary voltage&current and secondary current during discharge.

We have used many of these Tektronix HV probes. P6015,   P6015A, see specs. 20..40 kV 75 MHz derating depends on freq.

Current version P6015A: https://w140.com/tekwiki/images/b/b2/51W-8217-2.pdf

P6015 earlier version has larger comp box, requires Freon 114 gas fill, sale of the Freon was  banned long ago. Any used eg epay will have the F114 gas missing or already depleted so derated to 20 kV
https://w140.com/tekwiki/wiki/P6015

Great! They are also readily available. I'll definitely have a look at that!


Above 40 kV or the derated spec of P6015, you have some probes (see below) but more are  HV dividers:  large terminals with precision vacuum or air dielectric capacitors (AC only) or high meg HV dividers.


Few vendors, like

Ross Engineering

https://www.rossengineeringcorp.com/products/measurement/hv-probes-multimeters/hv-probes.html

https://www.rossengineeringcorp.com/products/measurement/hv-voltage-dividers.html

Stangenes Industries

https://www.electricalproducts.stangenes.com/viewitems/products/standard-voltage-dividers

Expect high cost and long delivery as most of these are made to order.

I can vouch that both my old friends Ross and Stangenes are very experienced high quality manufacturers


Enjoy and be safe!

Jon

Interesting companies! I'll check on them.

Thank you,  :-+
 

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3656
  • Country: fr
  • Analog, magnetics, Power, HV, Audio, Cinema
    • IEEE Spectrum
Re: High voltage oscilloscope probe
« Reply #3 on: January 16, 2023, 01:28:57 pm »
Rebonjour Cher Eugèneion,

The ICE ignition system is over 100 years old , electronic ignition trace back to 1960s.

Not much has changed, so the HV ignition coils and spark voltage ,current, énergie and waveforms are well studied and optimized.  I have never seen a 50..90 kV auto ignition system, spark pluginss, cables, and ignition coils are rated to 20..35 kV not higher, so Tektronix P6015/A should be sufficient.

But, What exactly is your goal? The flyback energy in the coil core creates a 20..40 kV fast rise ring wave, and the load is at first the stray capacité of the coil secondary and ignition wires.After the combustion is ignited, the spark transitions to an arc,  and becomes more resistive.
Suggest you research the many books and papers on gaseous conductors, sparks and arcs.

Finally, beware  the common mode voltage of the ignition can be coupled around the HV probe, masking the true waveforms on the  scope, and can even damage the scope input.

Thus, HV  measurements you propose are difficult and risky to test equipment even with the HV probes.

Bon courage

Jon
An Internet Dinosaur...
 

Offline Stray Electron

  • Super Contributor
  • ***
  • Posts: 2253
Re: High voltage oscilloscope probe
« Reply #4 on: January 16, 2023, 01:39:53 pm »
jonpaul,

   I used to see a lot of those p6015 probes in the surplus since they couldn't get the freon to fill them with any longer. Is there another gas that could be used to fill them with that would allow them to be used to greater than 20 kV?
 


Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: High voltage oscilloscope probe
« Reply #6 on: January 16, 2023, 02:03:09 pm »
I used to see a lot of those p6015 probes in the surplus since they couldn't get the freon to fill them with any longer. Is there another gas that could be used to fill them with that would allow them to be used to greater than 20 kV?
This has been discussed extensively in the TekScopes group. I don't believe anyone has found a suitable replacement that has a high enough breakdown voltage, has a dielectric constant that allows the existing compensation box to compensate the probe and has the right boiling point/vapour pressure at typical ambient conditions.

Offline EugenioNTopic starter

  • Contributor
  • Posts: 36
Re: High voltage oscilloscope probe
« Reply #7 on: January 16, 2023, 02:48:43 pm »
Rebonjour Cher Eugèneion,

The ICE ignition system is over 100 years old , electronic ignition trace back to 1960s.

Not much has changed, so the HV ignition coils and spark voltage ,current, énergie and waveforms are well studied and optimized.  I have never seen a 50..90 kV auto ignition system, spark pluginss, cables, and ignition coils are rated to 20..35 kV not higher, so Tektronix P6015/A should be sufficient.

But, What exactly is your goal? The flyback energy in the coil core creates a 20..40 kV fast rise ring wave, and the load is at first the stray capacité of the coil secondary and ignition wires.After the combustion is ignited, the spark transitions to an arc,  and becomes more resistive.
Suggest you research the many books and papers on gaseous conductors, sparks and arcs.

Whoops, I've been a bit short of details, my fault. It's not an automotive application!
The device I'm willing to design should generate a short (~10-100us) high voltage negative pulse; during the pulse it should measure the discharge current and make some assumptions based on the measured value.
The ICE analogy is limited to the circuit that drives the primary coil: in principle a mosfet from primary to ground to energize the coil with a known current and then open it to release the stored energy into the secondary. I've added a miller clamp from drain to gate to limit the primary voltage to a max value (hundreds of V).

Finally, beware  the common mode voltage of the ignition can be coupled around the HV probe, masking the true waveforms on the  scope, and can even damage the scope input.

Thus, HV  measurements you propose are difficult and risky to test equipment even with the HV probes.

Interesting. Could this be the reason that Tek put a dotted line near the probe tip with a warning of performance degradation (eventually complete probe breakdown, we may say)?
https://www.amplifier.cd/Test_Equipment/Tektronix/Tektronix_other/images/P6015A.jpg
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: High voltage oscilloscope probe
« Reply #8 on: January 16, 2023, 03:07:21 pm »
Interesting. Could this be the reason that Tek put a dotted line near the probe tip with a warning of performance degradation (eventually complete probe breakdown, we may say)?
https://www.amplifier.cd/Test_Equipment/Tektronix/Tektronix_other/images/P6015A.jpg
My guess would be that metal objects near this area may affect parasitic capacitance, and hence frequency response, since the probe is not shielded in this area. Also the creepage distance from the tip would be quite short here.

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3656
  • Country: fr
  • Analog, magnetics, Power, HV, Audio, Cinema
    • IEEE Spectrum
Re: High voltage oscilloscope probe
« Reply #9 on: January 16, 2023, 03:14:39 pm »
Eugene many thanks you are in USA? Canada? EU?

Since 1970s,  have designed many HV pulse transformers for arc lamp and Xenon ignition, single pulse as well as HF ring wave , and radar pulse transformers, from 4 kV to 75 kV.

HV pulse  is an art as  well as engineering and measurement problem.

The essence is understandings the  physics   magnetics and winding design of the transformer and  load. Special core designs and winding structure as well as insulation and low capacitance windings are used.

The technology and innovations for HV pulse generation  and measurement reach back to Nikola Tesla 1880s,  Charles Steinmetz 1880..1935, Lawrence Berkeley and LLNL, LANL, Hansen (SLAC) , etc.

Suggest to  do more research before proceeding with hardware design and experiments.

Please contact me off the forum by PM if you lie to proceed.

Off to work!

Jon

PS: I have some spare P6015...

An Internet Dinosaur...
 

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3656
  • Country: fr
  • Analog, magnetics, Power, HV, Audio, Cinema
    • IEEE Spectrum
Re: High voltage oscilloscope probe
« Reply #10 on: January 16, 2023, 03:32:39 pm »
P6015 requires Freon 114 fill with specific gas/liquid properties. With no freon, it works to 20 kV. \

The Freon usually Leaked out of the seals after some time and needed a recharge. The clear ring at the P56015 tip allowed Freon  liquids level to be checked.

P6015A has a silicon insulation thus no gas.

yes the tip capacitance is just a few pF so any metal near the tip will degrade  the pulse response.

Jon
An Internet Dinosaur...
 

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: High voltage oscilloscope probe
« Reply #11 on: January 16, 2023, 05:59:01 pm »
I recall some people suggested that butane was the closest readily available option to the original P2015 freon (albeit at a somewhat higher vapor pressure) - has anyone here tested it? P2015s are a lot cheaper to come by than P2015As.
Obviously if you do push it to the breakdown point an over-pressure+arc event with butane is worse than with freon, but you're probably in a pickle regardless if you're dumping enough energy to blow up the probe in the first place.

My P2015 actually came with a bit of Freon left in the can (hopefully enough for a charge or two), I've kept it (in the fridge) unused though as I need to replace the probe seal before re-charging it with unobtanium gas.
 

Offline jonpaul

  • Super Contributor
  • ***
  • Posts: 3656
  • Country: fr
  • Analog, magnetics, Power, HV, Audio, Cinema
    • IEEE Spectrum
Re: High voltage oscilloscope probe
« Reply #12 on: January 16, 2023, 07:16:19 pm »
Hello again

I was super lucky to find a mint P6015A in orig case/access plus several probes/comp boxes/clbes lose at a Ham fle I think....

Far superior to P6015.

A  P6015 depends on the dielectric constant of the F114 gas and the boiling point and vapro pressure as well.

The liquid charge of perhaps 12 mm at the tip held dow creates the gas in the proble body.

Other gases/liquids will not have the correc Dielectric constant nor vapor pressure.

Finally Butane is highly flammable and a fire/explosion hazard used areoun high voltage.

The freon cans go deal gradually and you need a certain amount of the freion to properly charge the P6015.

A P6015 withoyut Freon and just air works OK but only to 20 KV.

Stt eh TEK manual on this point.

Finally P6013 is a smaller, older model that has no freon and goes to 15..20 kV.

That is a low-cost alternate.

Enjoy,

Jon

An Internet Dinosaur...
 

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: High voltage oscilloscope probe
« Reply #13 on: January 16, 2023, 09:03:47 pm »
I'm pretty sure that the probe does not depend on the dielectric constant of the freon - only a small amount is in liquid form, and the vapor will not be dense enough (IIRC, something between 2 and 3 atmospheres) to have a dielectric constant appreciably different to 1. I'll admit couldn't find the actual number for CFC-114 specifically, but all other gases I did find had STP dielectric constants values within 1% or so of 1 (as expected for their low density). The gas is in there solely to raise the breakdown voltage (through both having a higher dielectric strength than air, and also by being at an elevated pressure).

The point about it not being ideal to have a flammable gas is a valid one though (note that CFC-114 apparently forms phosgene when exposed to flame though!) - my main concern about using butane would be a leak igniting.
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: High voltage oscilloscope probe
« Reply #14 on: January 16, 2023, 09:54:52 pm »
I'm pretty sure that the probe does not depend on the dielectric constant of the freon - only a small amount is in liquid form, and the vapor will not be dense enough (IIRC, something between 2 and 3 atmospheres) to have a dielectric constant appreciably different to 1.
It depends on the dielectric constant filled with the freon and filled with just air to be approximately the same, so it can compensate to both conditions. So if you fill it with something that produces a dieletric constant substantially different from 1, like silicone as in the P6015A, or a liquid, it will fail to compensate.

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: High voltage oscilloscope probe
« Reply #15 on: January 16, 2023, 10:09:35 pm »
The dielectric constant with freon gas (the liquid is only a small fraction of the volume of the probe so can be ignored), or air, or butane gas for that matter, will all be ~1, no compensation change needed (note the manual doesn't say to re-compensate when used without freon, just that the voltage rating will suffer). You cannot substantially change it from 1 without changing from a gas fill to a liquid (e.g. the silicone in the P6015A, which did need design changes to be able to compensate for) or solid fill.

Using a different dielectric gas will only change:
- Breakdown voltage (gas properties and working pressure dependant)
- Probe internal pressure vs. temperature
- Safety
- Ozone depletion/global warming potential
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: High voltage oscilloscope probe
« Reply #16 on: January 16, 2023, 10:51:07 pm »
According to this paper the breakdown voltage of butane at room temperature and atmospheric pressure is slightly lower than that of air. So what's the benefit of using butane?

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: High voltage oscilloscope probe
« Reply #17 on: January 16, 2023, 11:23:12 pm »
I think people were looking at butane as it's similar in vapour pressure to the original Freon. Unfortunately it's likely that it's not as high a dielectric strength as the original, though the slightly increased vapour pressure should help a little. I didn't find any hard numbers with a quick search, hence my original question - that paper seems to be about butane flames, not sure it's applicable?
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2302
  • Country: us
Re: High voltage oscilloscope probe
« Reply #18 on: January 17, 2023, 04:44:47 am »
...
   I used to see a lot of those p6015 probes in the surplus since they couldn't get the freon to fill them with any longer. Is there another gas that could be used to fill them with that would allow them to be used to greater than 20 kV?
You might find this post interesting:

  https://www.eevblog.com/forum/testgear/high-voltage-oscilloscope-probes-make-or-buy/msg4326361/#msg4326361

In the referenced AN-108, they debunk the butane fill and recommend a special silicone oil.  It's not a Tek app note, but they appear to have done their homework.
 

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: High voltage oscilloscope probe
« Reply #19 on: January 17, 2023, 08:07:28 am »
Thanks for the link - suggests that a) dielectric constant can be entirely ignored for a gas fill (but not if you totally fill it with a liquid) and b) butane sadly doesn't have enough dielectric strength, which is what I feared :(

I think I'll be using butane for pressure testing only - don't want to re-do the seal in mine and fill it with the last of the freon only for it to all leak out again.
 

Offline alm

  • Super Contributor
  • ***
  • Posts: 2903
  • Country: 00
Re: High voltage oscilloscope probe
« Reply #20 on: January 17, 2023, 12:58:13 pm »
Interesting app note, though from a (for me) unknown source. Clearly the parasitic capacitance is increased when filled with silicone oil since they have to more than double the maximum value of those trimmer caps. I wonder how much this will affect the input impedance, transient response / bandwidth and voltage derating.

This thread discusses replacements, and there is some discussion at the end about filling the probe with silicone oil and its effect on compensation and bandwidth, although nothing definitive. Pay extra attention to posts by user ditter2, who was the engineering manager for the redesign project (P6015A).

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 12277
  • Country: us
Re: High voltage oscilloscope probe
« Reply #21 on: January 17, 2023, 02:29:04 pm »
My homemade probe


Offline HighVoltage

  • Super Contributor
  • ***
  • Posts: 5560
  • Country: de
Re: High voltage oscilloscope probe
« Reply #22 on: January 17, 2023, 05:13:37 pm »
For up to 40 kV I also use the Tektronix P6015A
Anything higher, I use self build voltage dividers for up to 100 kV

What accuracy do you need in the probe?


There are 3 kinds of people in this world, those who can count and those who can not.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf