Open HV Probe 40kV

[beanflying]

Reason HV probes are either rubbish from the usual asian suspects or what I consider way OTT on price for what little they contain.

While this is starting out as a 3D Print type of project the options for what goes inside and what you use the enclosure for can be far more wide ranging. The design is optimised to avoid needing supports when printed is the main reason for so many parts.

Currently I am printing the first sample one and tweaking the design as I go but as you see it is a close clone the Pintek and some others on the commercial market.

Version 1: 40kV maximum rated 990Mohm divider against a 10Mohm Meter for a 100:1 divider. Plan is to make up the 990 MOhm Resistor with 500, 400 & 90 MOhm Ohmite Slimmox 1% https://au.mouser.com/datasheet/2/303/res_slimmox-1265542.pdf . Fairly easy to change around from that if you want but the three of them will fit in the long tip barrel easily attached to a 5mm knurl nut in the final tip section to screw the metal probe into. These three cost circa $22 USD from Mouser so add that to a few other bits and for around $40-50 you will have a 1% 40kV HV Probe.

Version 2: Something in a lower kV rating but with far better bandwidth for possible Scope use.? Thinking circa 2-3kV and 1-2MHz? Totally open to circuit options and ideas and I have added a slotted mount into the handle to take a 20x80x1.6mm PCB.

Version 3: Chicken Stick for playing with Neon and MOT Transformers just because it's fun and it should be safe fun lest your internal smoke come out.

Version X: Up to you lot. Possible self contained probe like the older Heathkit one with a built in meter in the handle?

When I have a print that works and screws together I will drop the STL's here for others to try out most likely later today as I am onto the 2nd Red barrel section already and so far so good.

[tggzzz]

What material are you using?

Is it in any way hygroscopic over time?

Can oils/grease/etc lodge on the surface and be difficult to remove?

At 40kV it might be necessary to be careful of those points, but those questions are worth what you pai for them.

[beanflying]

What material are you using?

Is it in any way hygroscopic over time?

Can oils/grease/etc lodge on the surface and be difficult to remove?

At 40kV it might be necessary to be careful of those points, but those questions are worth what you pai for them.

True of fairly much all plastics and especially with HV. Regardless of commercial or home made care with oils and water is a must do. I am planning to use PLA on mine but ABS would be fine too, PETG I have never seen any tests done on it but it is better with moisture then PLA for example but I am uncertain of it's electrical characteristics.

The couple of youtube videos below are worth a look.  There is also some on Lichtenberg Figures and dumbasses with MOT's but playing with salt water and HV is DUMB if you don't know what you are doing.

With Version 1 for example if used properly your hand will be 60+mm away from an Earth referenced point and further away than that to any HV so unless you are being rained on with salt water fairly safe.

Long term keep it clean and dry and if in doubt print another.

https://youtu.be/-_LJXKLgIYc

https://youtu.be/ZqXQRq8As1o

[tggzzz]

What material are you using?

Is it in any way hygroscopic over time?

Can oils/grease/etc lodge on the surface and be difficult to remove?

At 40kV it might be necessary to be careful of those points, but those questions are worth what you pai for them.

True of fairly much all plastics and especially with HV. Regardless of commercial or home made care with oils and water is a must do. I am planning to use PLA on mine but ABS would be fine too, PETG I have never seen any tests done on it but it is better with moisture then PLA for example but I am uncertain of it's electrical characteristics.

The fabrication technique may also be relevant, in particular the body will not be smooth and will not be solid. Both of those may trap undesirable water/oil/etc.

Long term keep it clean and dry and if in doubt print another.

I always have doubt, so that wouldn't help me!

Sorry, I don't feel like spending 40 minutes of my remaining life looking at those videos. (Especially since almost all yootoob vids are talking heads and I can speed read much faster than I can watch videos)

[tggzzz]

For comparison, here's my 40kV meter. Tip to disc is 13cm, disc diameter is 7cm

I picked it up cheaply, and haven't had occasion to use it at more than a too small fraction of its rated value.

[beanflying]

What material are you using?

Is it in any way hygroscopic over time?

Can oils/grease/etc lodge on the surface and be difficult to remove?

At 40kV it might be necessary to be careful of those points, but those questions are worth what you pai for them.

True of fairly much all plastics and especially with HV. Regardless of commercial or home made care with oils and water is a must do. I am planning to use PLA on mine but ABS would be fine too, PETG I have never seen any tests done on it but it is better with moisture then PLA for example but I am uncertain of it's electrical characteristics.

The fabrication technique may also be relevant, in particular the body will not be smooth and will not be solid. Both of those may trap undesirable water/oil/etc.

Long term keep it clean and dry and if in doubt print another.

I always have doubt, so that wouldn't help me!

Sorry, I don't feel like spending 40 minutes of my remaining life looking at those videos. (Especially since almost all yootoob vids are talking heads and I can speed read much faster than I can watch videos)

You wanted information and some details then dismiss without reviewing and then trust in antique secondhand equipment of unknown origin

Also fairly insulting to regard w2aew as a talking head for a start and Kerry is far from one either.

[tggzzz]

What material are you using?

Is it in any way hygroscopic over time?

Can oils/grease/etc lodge on the surface and be difficult to remove?

At 40kV it might be necessary to be careful of those points, but those questions are worth what you pai for them.

True of fairly much all plastics and especially with HV. Regardless of commercial or home made care with oils and water is a must do. I am planning to use PLA on mine but ABS would be fine too, PETG I have never seen any tests done on it but it is better with moisture then PLA for example but I am uncertain of it's electrical characteristics.

The fabrication technique may also be relevant, in particular the body will not be smooth and will not be solid. Both of those may trap undesirable water/oil/etc.

Long term keep it clean and dry and if in doubt print another.

I always have doubt, so that wouldn't help me!

Sorry, I don't feel like spending 40 minutes of my remaining life looking at those videos. (Especially since almost all yootoob vids are talking heads and I can speed read much faster than I can watch videos)

You wanted information and some details then dismiss without reviewing and then trust in antique secondhand equipment of unknown origin

It came directly from an industrial calibration facility, and is visibly in a good condition. I cannot foresee using it on more than 20kV (i.e. CRT voltages).

So yes, I would trust that more than a homebrew fabrication using materials with poorly understood properties.

For amusement, I did make a handle for a Tek HV scope probe (using SLA), but only on the grippable side of the disc; the probe was safe to use without that part of the handle (apart from dropping it!).

Also fairly insulting to regard w2aew as a talking head for a start and Kerry is far from one either.

I didn't dismiss them. I did say I'm not prepared to spend 40 minutes looking at their videos. After all, it may be interesting to you but is only of passing interest to me.

[beanflying]

So in passing with NO evidence to back your claims about 'poorly understood properties' used 3D printing materials when offered some ACTUAL testing of those materials using FDM you still dump on the process and materials?

PLA along with ABS are very well understood materials as is PETG. Please let me know how you arrive at 'poorly understood properties'?

[DaJMasta]

It's a neat project but I would be worried about choice of plastic as well - namely in disclaiming that you really need to verify the exact plastic you're using to print the parts and clean them thoroughly before assembly.  There is a gigantic variation in available PLAs, and especially at these voltages, even the choice of pigment could degrade the dielectric breakdown voltage quite considerably.


Yes, by putting that much plastic between you and the high voltage, a lot of things are covered, but with commercial probes they have the advantage of being able to mold their probe housings around known, tested materials produced in a very controlled environment, which is exactly the opposite of a home 3d printer putting together a neat project.  I have little doubt that with the right choice of materials you can make a safe probe, and I'll go as far as to say that most standard plastic filaments should do the job.... but I know for sure that some will not, and this is dangerous enough that it's not worth the risk unless you can test your specific material used to be sure.


Perhaps there are filament manufacturers that actually have electrical breakdown voltages on their datasheets, and perhaps it's "safe enough" for your own use, but I really think there should be a reminder that emphasizes that not all filaments are the same so that unless you can buy a rated filament and use it within specification or you can test it and verify yourself, that you shouldn't just DIY this one on whatever filament is lying around.


Come to think of it, though, maybe there's a high voltage insulator tubing or something you could cover the divider with that would make much of the worry about breakdown voltage of the plastic away, or even some sort of high voltage rated potting material.

[Zero999]

Also note that the input impedance of most meters isn't that well controlled. I've found it can differ significantly from the stated 10M

A x10 oscilloscope probe should be much better though, as it will be designed with a definite 1M 'scope input impedance in mind.

[beanflying]

Also note that the input impedance of most meters isn't that well controlled. I've found it can differ significantly from the stated 10M

A x10 oscilloscope probe should be much better though, as it will be designed with a definite 1M 'scope input impedance in mind.

Agreed about the impedance all of my handhelds vary off the 10MOhms on lower ranges. It was part of the reason to go with 100:1 and use the highest range setting available. My testing matches what Allan found in the video above on the lower ranges but on the highest ranges they are very close.

For anything above main frequencies I would expect a fairly major error hence looking toward something for lower kV and compensation as a different option.

[beanflying]

It's a neat project but I would be worried about choice of plastic as well - namely in disclaiming that you really need to verify the exact plastic you're using to print the parts and clean them thoroughly before assembly.  There is a gigantic variation in available PLAs, and especially at these voltages, even the choice of pigment could degrade the dielectric breakdown voltage quite considerably.


Yes, by putting that much plastic between you and the high voltage, a lot of things are covered, but with commercial probes they have the advantage of being able to mold their probe housings around known, tested materials produced in a very controlled environment, which is exactly the opposite of a home 3d printer putting together a neat project.  I have little doubt that with the right choice of materials you can make a safe probe, and I'll go as far as to say that most standard plastic filaments should do the job.... but I know for sure that some will not, and this is dangerous enough that it's not worth the risk unless you can test your specific material used to be sure.


Perhaps there are filament manufacturers that actually have electrical breakdown voltages on their datasheets, and perhaps it's "safe enough" for your own use, but I really think there should be a reminder that emphasizes that not all filaments are the same so that unless you can buy a rated filament and use it within specification or you can test it and verify yourself, that you shouldn't just DIY this one on whatever filament is lying around.


Come to think of it, though, maybe there's a high voltage insulator tubing or something you could cover the divider with that would make much of the worry about breakdown voltage of the plastic away, or even some sort of high voltage rated potting material.

There would be room for putting a glass sleeve around the resistors until you get to the Earth and meter tap connection. The Resistors I am looking at will all be contained in the single long section as will the terminations so at that point the highest voltage at a full 40kV in will be 400V with 70mm of plastic between the user and that point.

Edit Avoid BLACK would be a good start re pigments too as it is likely carbon based.

[tggzzz]

So in passing with NO evidence to back your claims about 'poorly understood properties' used 3D printing materials when offered some ACTUAL testing of those materials using FDM you still dump on the process and materials?

I refer you to my first and subsequent posts (with my emphasis), and I am content to let others decide whether I am "dumping on process and materials"...

    What material are you using?
    Is it in any way hygroscopic over time?
    Can oils/grease/etc lodge on the surface and be difficult to remove?
    At 40kV it might be necessary to be careful of those points, but those questions are worth what you pai for them.
    The fabrication technique may also be relevant, in particular the body will not be smooth and will not be solid. Both of those may trap undesirable water/oil/etc.

PLA along with ABS are very well understood materials as is PETG. Please let me know how you arrive at 'poorly understood properties'?

The fabrication technique can be as important as the material, especially for uses like this.

[tautech]

Version 2 suggestion.
1000:1 probe of say 5KV design rating and along the lines of Pintek's 8KV probe.

Far less unwieldy than a 40KV jobbie and offers significant voltage headroom over 100:1 probes.

Example:
http://www.pintek.com.tw/files/pintek/HVP-08-front.pdf

Image for comparison to ordinary probes:
https://youtu.be/W30AqcDQL18?t=18


But hey we need be careful with projects like these and those that follow them must know they do so at their own risk........but hell, if Joe Smith can make a 1000:1 probe for monitoring his 15KV pulse gen then with care and understanding others with the capability could too.
Joe's journey down the HV probe rabbit hole:




There are a few more for those that want to hunt them out.

[beanflying]

And some more 'actual evidence' and a study Study of electrical properties of 3D printed objects

and

https://www.sciencedirect.com/science/article/pii/S1876619616000565

Data on PLA (brand specific) https://ultimaker.com/download/74599/UM180821%20TDS%20PLA%20RB%20V10.pdf

The only downside would appear to be surface contamination and that to a degree is true of any materials regardless of process.

[tggzzz]

And some more 'actual evidence' and a study Study of electrical properties of 3D printed objects

and

https://www.sciencedirect.com/science/article/pii/S1876619616000565

Data on PLA (brand specific) https://ultimaker.com/download/74599/UM180821%20TDS%20PLA%20RB%20V10.pdf

A couple of quotes from the first reference that would concern me w.r.t. this application (my emphasis):

  The dielectric strength measurement was difficult to
  complete due to the surface discharges that occurred during
  the increasing of the voltage. Therefore the relatively high
  observational error distorted the measured data. However,
  the highest observed value of dielectric strength for PLA
  was about 33 kV/mm.

  However, several problems could occur, when using
  PLA in insulating systems. This material is derived from
  corn starch and so it is biodegradable, unstable in time. The
  mechanical and other properties could change very
  markedly during aging. The solution could be also the
  development of proper additive to suitable adjustment of
  the PLA properties.

  Generally, an improvement of the electrical properties of
  these materials is required, for example by adding some
  additives to the base material.

The ultimaker data contains nothing relevant to this discussion; I'm not sure why you referenced it.

The only downside would appear to be surface contamination and that to a degree is true of any materials regardless of process.

Of course. The degree is the key point; professionally produced products wiill have characterised the materials and processes, and won't be constrained to use whatever fits in their machines.

My first boss told me a story... He was using a screwdriver around the neck of a TV CRT, and after a while he got a hell of a belt from it. After looking carefully for the cause, he found there was an almost invisible oil track/film on part of the screwdriver's stem and handle.

[beanflying]

The Ultimaker data does contain electrical properties and goes toward what DaJMasta mentioned about actual datasheets on Filaments.

Far from being an 'poorly understood properties' but perhaps 'needing more study' is appropriate which is the broad NON Conclusions about aging and extended use they made.

If you take there figure of 33kV/mm as a best case the tip of my design is @ 210mm away from the hand gaurd. So 170X my proposed maximum would be needed in theory be needed to get it to breakdown. The PLA would need to be compromised dramatically for 40kV to be an issue.

[tggzzz]

The Ultimaker data does contain electrical properties and goes toward what DaJMasta mentioned about actual datasheets on Filaments.

It only contains dissipation factor and dielectric constant. Those are useful for assessing capacitance and HF losses, but have zero relevance to operation at 40kV.

Hence my statement that "The ultimaker data contains nothing relevant to this discussion".

Far from being an 'poorly understood properties' but perhaps 'needing more study' is appropriate which is the broad NON Conclusions about aging and extended use they made.

If you take there figure of 33kV/mm as a best case the tip of my design is @ 210mm away from the hand gaurd. So 170X my proposed maximum would be needed in theory be needed to get it to breakdown. The PLA would need to be compromised dramatically for 40kV to be an issue.

Engineers base their designs on worst case specs, not best case specs.

I gave an example of how smooth and solid metal and plastic could be dangerously compromised. Hombrew 3D printing is worse in that it has non-smooth surfaces with problems stated in your reference. Add to that any porosity and aging, and who knows what the worst case properties might be.

But hey, it is your heart, and it is up to you to do with as you wish.

[PartialDischarge]

If you take there figure of 33kV/mm as a best case the tip of my design is @ 210mm away from the hand gaurd. So 170X my proposed maximum would be needed in theory be needed to get it to breakdown. The PLA would need to be compromised dramatically for 40kV to be an issue.

Well not really. I'm not going to position myself for or against the use of PLA but let me clarify some terms.
Breakdown is measured across a material and not over the surface. The latter is surface tracking which is highly dependable on humidity and roughness and on the material in question. Some materials are tested for CTI (comparative tracking index) for that reason in some materials used in power equipment.

Now, breakdown can measured at DC or at 50Hz, and it makes a huge difference. That same paper he measured in the video with a dc breakdown of >1kVdc would break down at 200Vac easily.

Most importantly and less known is that for anything high voltage (>10kV) conductivities, permittivities and the distribution of the field lines are of utmost importance yet ignored by mostly everyone.
At pure DC, high voltage only cares about conductivities along its path and will create resistive dividers as a result. At AC material permittivities do matter and capacitive dividers are formed. If a high er material is inserted as an insulator at a certain point near a high voltage conductor for example, field lines will be expelled to the outside, and if that outside is air, the concentration of field lines may create breakdown in this space, something with goes against the idea that inserting a good insulator in a high voltage path is always positive.

Chapter 2 in this interesting book talks about this
https://www.academia.edu/30766171/High_Voltage_Engineering_Practice_and_Theory

[PartialDischarge]

It only contains dissipation factor and dielectric constant. Those are useful for assessing capacitance and HF losses, but have zero relevance to operation at 40kV.

Actually the dielectric constant is the top 1 or 2 most important specification or any material used in high voltage, since it dictates how the field and equipotential lines will be distributed, the capacitances that you mentioned. Since most HV applications are for power line frequency this has been extensively studied, the most typical being an air void inside a high er dielectric (due to a problem in epoxy injection in a bushing for example). This will create arcing inside the void (acting as a very small capacitor) which will create UV light, which will destroy surrounding walls, which will create cracks, conductive paths, etc ,etc

[tggzzz]

It only contains dissipation factor and dielectric constant. Those are useful for assessing capacitance and HF losses, but have zero relevance to operation at 40kV.

Actually the dielectric constant is the top 1 or 2 most important specification or any material used in high voltage, since it dictates how the field and equipotential lines will be distributed, the capacitances that you mentioned. Since most HV applications are for power line frequency this has been extensively studied, the most typical being an air void inside a high er dielectric (due to a problem in epoxy injection in a bushing for example). This will create arcing inside the void (acting as a very small capacitor) which will create UV light, which will destroy surrounding walls, which will create cracks, conductive paths, etc ,etc

I stand corrected; thanks for the clear explanation. I now have to fall back on my statement "...but those questions are worth what you pay for them"

However, since there are so many parameters (e.g. frequency, porosity, surface texture) involved, I don't see how the dielectric constant can give any useful predictive information as to the maximum voltage that such a probe could withstand. I presume measurements would have to be done, and I would be surprised if many amateurs have appropriate equipment.

[bd139]

Just a point on PLA, the actual conductivity varies with temperature and the surface can have pin prick holes in it which are enough for HV to creep out. This is one application where (quality) injection moulding is essential unfortunately. Any failure could escalate pretty quickly!

HV is terribly unpredictable as well. My first proper zap was trying to build an HV probe for TV anode. I did this with some PVC tubing with stacked carbon composition resistors in it and filled with epoxy. The HT decided it was quite happy to go around the probe entirely because it wasn’t clean.

You can probably “megger” a finished probe but I’m not sure I’d trust it hands on ever.

[beanflying]

The Ultimaker data does contain electrical properties and goes toward what DaJMasta mentioned about actual datasheets on Filaments.

It only contains dissipation factor and dielectric constant. Those are useful for assessing capacitance and HF losses, but have zero relevance to operation at 40kV.

Hence my statement that "The ultimaker data contains nothing relevant to this discussion".  As stated it was meant for DaJMasta not you specifically as mentioned but IGNORED by you to score a supposed point about what they don't include

Far from being an 'poorly understood properties' but perhaps 'needing more study' is appropriate which is the broad NON Conclusions about aging and extended use they made.

If you take there figure of 33kV/mm as a best case the tip of my design is @ 210mm away from the hand gaurd. So 170X my proposed maximum would be needed in theory be needed to get it to breakdown. The PLA would need to be compromised dramatically for 40kV to be an issue.

Engineers base their designs on worst case specs, not best case specs.

I gave an example of how smooth and solid metal and plastic could be dangerously compromised. Hombrew 3D printing is worse in that it has non-smooth surfaces with problems stated in your reference. Add to that any porosity and aging, and who knows what the worst case properties might be.

But hey, it is your heart, and it is up to you to do with as you wish.

As the study I added was based on a using a Prusa Mk2 What is the difference in my 'homebrew' printing to their 'Homebrew'?

The fear argument and scenarios you keep repeating without facts and data are a warning  at most (which I am aware of and acknowledge) and I await you to show me some numbers that the design is or will be unsafe and not your imagined remote maybe's?

BTW have you 'tested' that old cord and probe on your HV Tester? Or are you happy to place yourself at risk with an unknown plastic and cable that may or may not have broken down to an unknown degree with age?

[beanflying]

Just a point on PLA, the actual conductivity varies with temperature and the surface can have pin prick holes in it which are enough for HV to creep out. This is one application where (quality) injection moulding is essential unfortunately. Any failure could escalate pretty quickly!

HV is terribly unpredictable as well. My first proper zap was trying to build an HV probe for TV anode. I did this with some PVC tubing with stacked carbon composition resistors in it and filled with epoxy. The HT decided it was quite happy to go around the probe entirely because it wasn’t clean.

You can probably “megger” a finished probe but I’m not sure I’d trust it hands on ever.

Even if the HV found a path out of any flaw in the probe tip as mentioned the ground potential end of the resistors string will be 70mm from the user. It will be about 100 to the end of the first Resistor which at 40kV input will still only be at 4kV. Assuming the probe isn't contaminated 70mm Airgap is a lot of safety even if something does go to poo electrically.

[tggzzz]

The Ultimaker data does contain electrical properties and goes toward what DaJMasta mentioned about actual datasheets on Filaments.

It only contains dissipation factor and dielectric constant. Those are useful for assessing capacitance and HF losses, but have zero relevance to operation at 40kV.

Hence my statement that "The ultimaker data contains nothing relevant to this discussion".  As stated it was meant for DaJMasta not you specifically as mentioned but IGNORED by you to score a supposed point about what they don't include

Far from being an 'poorly understood properties' but perhaps 'needing more study' is appropriate which is the broad NON Conclusions about aging and extended use they made.

If you take there figure of 33kV/mm as a best case the tip of my design is @ 210mm away from the hand gaurd. So 170X my proposed maximum would be needed in theory be needed to get it to breakdown. The PLA would need to be compromised dramatically for 40kV to be an issue.

Engineers base their designs on worst case specs, not best case specs.

I gave an example of how smooth and solid metal and plastic could be dangerously compromised. Hombrew 3D printing is worse in that it has non-smooth surfaces with problems stated in your reference. Add to that any porosity and aging, and who knows what the worst case properties might be.

But hey, it is your heart, and it is up to you to do with as you wish.

As the study I added was based on a using a Prusa Mk2 What is the difference in my 'homebrew' printing to their 'Homebrew'?

How the hell should I know?!

If anything, it is up to you to determine that the prints are comparable.

The fear argument and scenarios you keep repeating without facts and data are a warning  at most (which I am aware of and acknowledge) and I await you to show me some numbers that the design is or will be unsafe and not your imagined remote maybe's?

Please don't advance strawman arguments.

I have asked questions. To me, the responses have not been answers.

BTW have you 'tested' that old cord and probe on your HV Tester? Or are you happy to place yourself at risk with an unknown plastic and cable that may or may not have broken down to an unknown degree with age?

You know the answer to that, because I have already stated "... haven't had occasion to use it at more than a too small fraction of its rated value".

However, given its apparent condition, I would far rather use that than a homebrew probe.

But hey, it is your health, not mine!