Open HV Probe 40kV

[beanflying]

Funny I watched Dave's video #85 again this morning as it came up in a search. People are worried about PLA and this one is maybe a touch less safe out there for the masses to watch  I have no qualms about the maker of it being safe after seeing what else he plays with but it lacks for most.

I will grab some dope at some stage but opt for an epoxy clear coat over the hot end of the 40kV probe based on that. Also checked and based on your videos there is plenty of diameter for a secondary thin wall PLA/Kapton Taped inner tube for the Resistors and any above ground terminations to sit in the tip barrel (14mm bore but fitted with a 13.5x12.7 wrapped tube).

Just reprinting some bits including the handle after increasing the thread pitches on the parts so they screw together without needing any post processing.

Shown in the photo is four parts screwed together. Other than removing the Brim (Flashing) from the prints they are as the came off the printer (Ender Pro).

[vk6zgo]

Back in the day, EHT probes were quite expensive, & the users paid due care & attention to the use and storage of them.

For this reason I would trust an old HV probe. (After, of course, checking it thoroughly for physical damage.)
Most of the old style ones are made from hard polystyrene, which, if looked after well, maintains its voltage rating over many years.
I do remember seeing one probe machined out of solid Perspex, but that was an outlier.

Most HV probes were used to look at TV set EHT, which although it has lots of voltage, cannot source enough current to do much harm.
Transmitter HT supplies are another kettle of fish, but most full size TV or Radio Broadcast Tx have their own built-in metering for such supplies, so even at such a site, the HV probes didn't get that much use.

3D printed devices are not as homogenous as cast or extruded plastics, but whether that is a problem, or not , I don't know.

[tautech]

Also checked and based on your videos there is plenty of diameter for a secondary thin wall PLA/Kapton Taped inner tube for the Resistors and any above ground terminations to sit in the tip barrel (14mm bore but fitted with a 13.5x12.7 wrapped tube).

This is key IMHO ^^^
The handle should just be a vessel to enclose the construction of safe innards.

If, and only if the printed materials can be proven to be safe against leakage or flash-over may they be used in any insulation capacity.
We can never be sure someone with the printer files and a printer won't use an unsuitable filament. 

[beanflying]

Also checked and based on your videos there is plenty of diameter for a secondary thin wall PLA/Kapton Taped inner tube for the Resistors and any above ground terminations to sit in the tip barrel (14mm bore but fitted with a 13.5x12.7 wrapped tube).

This is key IMHO ^^^
The handle should just be a vessel to enclose the construction of safe innards.

If, and only if the printed materials can be proven to be safe against leakage or flash-over may they be used in any insulation capacity.
We can never be sure someone with the printer files and a printer won't use an unsuitable filament. 

All things are possible you and can now get fairly easily get Conductive PLA https://www.proto-pasta.com/pages/conductive-pla laced with Carbon. Anyone making an electrical safety probe with it would truly deserve their Darwin Award There is other options out there but Styrene (ABS, HIPS etc) or PLA are clearly the better and easier to use options.

With switching to a secondary tube it is in theory possible to remove the metal tip from having direct contact to the outer shell as it could be mounted onto a stronger inner tube without major changes. A flanged end at the rear of the inner sleeve would hold it in place then carry the probe tip forward. It would only need the first two tip sections modding to make this possible.

BUT The downside to this is it would allow the possible ingress and entrapment of liquids or dirt between the inner and outer unless maybe a teflon encapsulated o-ring or seal of some sort was added. Adding more parts with finer tolerances and complexity would maybe not increase overall safety. Like replicating all the parts used in Joes probes it pushes the design beyond most or you may find people cutting corners

Sample Inner sleeve done and wrapped with Kapton, I will drop the diameter another 0.4mm so a second or third layer can be added easily. New Handle section is on the printer with the PCB mounts and latest revisions. Will be in Clear PLA as I am short of Black at present.

[0culus]

Nice work, bean!

Reading over the conversation in this thread, I gotta agree with bean that open designs are overall better. Idiots will be idiots, so let's stop wringing our hands over what they might do. If it's not playing with large electrical potentials, it will be high powered fireworks, thermite, binary explosives, dumbass lifted pickups, and more. So, put proper disclaimers with it and if someone does something stupid, it's their own damn fault. I'd far rather have open and accessible designs than none at all.

 

[tggzzz]

Also checked and based on your videos there is plenty of diameter for a secondary thin wall PLA/Kapton Taped inner tube for the Resistors and any above ground terminations to sit in the tip barrel (14mm bore but fitted with a 13.5x12.7 wrapped tube).

This is key IMHO ^^^
The handle should just be a vessel to enclose the construction of safe innards.

If, and only if the printed materials can be proven to be safe against leakage or flash-over may they be used in any insulation capacity.
We can never be sure someone with the printer files and a printer won't use an unsuitable filament. 

All things are possible you and can now get fairly easily get Conductive PLA https://www.proto-pasta.com/pages/conductive-pla laced with Carbon. Anyone making an electrical safety probe with it would truly deserve their Darwin Award There is other options out there but Styrene (ABS, HIPS etc) or PLA are clearly the better and easier to use options.

PLA and many thermoplastics are hygroscopic to some extent. Unless proven otherwise I would assume that would change the electric properties over time, e.g. it becomes more conductive.

I have not seen (because I haven't looked for) any measurements indicating whether or not that would be sigificant in this application.

[beanflying]

H₂O (pure) has a far greater dielectric strength than air for example and any amount of moisture absorbed will be more or less pure due to its origin. Discussion of moisture in filament is generally due to the issue of what there is creates steam when used to print something. There is no figures I am aware of re changed electric or physical data over time but is is massively hard to get even very high voltage to pass through pure water due to a lack of ions. Mine for example once a month maybe is the use case my HV scope probe will get over 1kV and I have meters already up to that point so maybe a silica pack and vac bag it.

If anyone is still concerned then styrene options are always there but they then come with some common solvent based issues (acetone in particular). That will include the Vintage ones vk6zgo mentioned a few posts back.

With accelerating/composting PLA for example both high heat (70+C) and high moisture with limited oxygen is needed. UV can effect some pigments. Refound some practical outdoor testing of PLA prints from Angus @ makersmuse in the Australian weather. Bear in mind a lot of these pots are SINGLE LAYER (0.4mm wall) I am printing @ 6 layers with 40% infill where needed.

https://youtu.be/qqNfa_zExRU

[mnementh]

That picture makes me cringe; no less than that video we were just discussing. 

I'm not hand-wringing. I'm pointing out clear and obvious hazards in publishing the stls for this. 

The intended manufacturing process and material simply and clearly are NOT fit for the intended purpose. Add to that the proclivity of dumbasses to just snurch and not read instructions, and the bottom line is that publishing these STLs is ethically the same as leaving a loaded gun lying around on the intardnet.

I'm not the only one here who can clearly see the clear and present danger this project represents; you discount those dangers at your own peril.

I've said my piece; I'm done here.

mnem
"Good luck! And may your god go with you."

[tggzzz]

Yes and yes.

And I too remember Dave Allen fondly

[beanflying]

For those who actually take the time to watch Angus's video in my last post will know the SINGLE LAYER vase mode prints suffered mechanical damage and are not the result of open air environmental, water or UV exposure. PLA needs very specific environmental conditions to breakdown without those conditions it is far from bio-degradable as numerous scientific studies have shown. Google it and do some reading of the scientific papers.

If your doubts are still with PLA then use ABS or HIPS like the commercial probes use and put up with its shortcomings like all materials have. Without having seen one in the flesh Pintek appears to still be using ABS based on it's cleaning warning against solvents http://www.pintek.com.tw/files/pintek/HVP-40.pdf PLA on the other hand is not effected by most solvents.

Part of the idea of this being open was to work on ideas and to improve the safety of the design and in a small way try and redress some of the youtube bs 'designs' that have little to none.

[ArthurDent]

I am somewhat baffled at the endless discussion of possible failure modes of different plastics, methods of construction, or maintenance issues that could cause injury or death while no one seems to be paying any attention to how this safety issue has been solved for decades.

As I said in post #34, the simple safety solution is to have a complete ground potential between you and the HV resistor built into the probe handle. Below is a drawing of the Tektronix 6015 HV probe showing the grounded tube they used around the end of the HV resistor. In the video link it shows the same idea in a Fluke probe of having the closest thing to the HV be a grounded shield and not your hand at minute 3:10.

[tggzzz]

I am somewhat baffled at the endless discussion of possible failure modes of different plastics, methods of construction, or maintenance issues that could cause injury or death while no one seems to be paying any attention to how this safety issue has been solved for decades.

As I said in post #34, the simple safety solution is to have a complete ground potential between you and the HV resistor built into the probe handle. Below is a drawing of the Tektronix 6015 HV probe showing the grounded tube they used around the end of the HV resistor.

Appropriate materials and appropriate construction are not mutually incompatible.

Do you need both or is either sufficient on its own?

Personally, in this case I think both are highly desirable - I wouldn't want to rely on either alone.

[ArthurDent]

tggzzz - "Appropriate materials and appropriate construction are not mutually incompatible."

I don't see anything wrong with any of the plastics used and I didn't mentioned that as a problem. What I am confused about is how the safety issue is being ignored when it has been solved ages ago. Everything can possibly fail, I just don't want that failure to take me with it.   

[beanflying]

Sigh, the Spaghetti monster paid me a visit overnight 

On the plus side all the threads work well as do the PCB slots. The modded inner tube now takes four layers of spiral wrapped Kapton and would probably take a fifth.

Re the outer shield it is certainly possible to make rolled copper sheet then soldered is the obvious one or you could design around a standard hard drawn Copper pipe size. The downside to it is in terms of usability as the probe diameter will need to be increased making it awkward in more cases. The Scope test point I needed to measure last week was already fairly tight and only 2kV so I could have butchered something up to get it done.

I see no reason not to use the base handle and lower sections and then screw a different tip section to it. It may actually be of benefit to do that in the case of the higher frequency version to add some noise shielding as well?

If we can move on a little:

PLA and ABS are both good insulators - PERIOD !
PLA as some sort of magic water sponge is BS and it doesn't breakdown rapidly even when immersed in water without microbiological activity as well.
PLA as a rapidly Bio-Degradable Plastic in open Air is BS unless someone finds me ACTUAL SCIENTIFIC EVIDENCE to prove otherwise?
Dirt Contaminants and Grime are the enemy so the probe outer needs to be easily cleanable (smooth) so an Epoxy Coating should be added to the 3DP surface.
Operator Safety is paramount as it always was. Lets work to improve it where possible.

[beanflying]

tggzzz - "Appropriate materials and appropriate construction are not mutually incompatible."

I don't see anything wrong with any of the plastics used and I didn't mentioned that as a problem. What I am confused about is how the safety issue is being ignored when it has been solved ages ago. Everything can possibly fail, I just don't want that failure to take me with it.   

Just slowed the video right down you linked and took a screen cap. Some of the commentary about spark gaps and the circuit is wrong and the parts he tried to reference simply hook over the shield on the barrel section as shown. The question if someone that has one can check is how far into the barrel on the Fluke does the shield go? Also the LV end of the Resistors is back right at the operators hand

In the case of the Resistors I am looking at and the layout of my design it sees that point 65-70mm up into the barrel.

[ArthurDent]

I was only using that video to show how the Fluke probe was constructed and not anything else, although his using pill bottles for insulators is a nice touch!

When the metal piece the coax is crimped into is pushed into the front part of the probe, the end of the resistors look like they are almost an inch inside and the two metal tabs on either side of that piece makes good contact with the two metal threaded halves and pretty much enclose any live parts inside the tube. All metal shell parts and the coax shield inside the handle are at ground potential. At 1000:1 the maximum voltage at that end should be less than 40 volts so it should be fairly safe as long as it is connected to a meter. I assume the second (ground) lead going to the resistor assembly is to a safety resistor across the output to limit the maximum voltage if there is no meter connected.

I don't know how far the shield goes inside the front part of the probe but the drawing of the Tektronix probe shows that that one goes about 3/4 of the distance to the tip. The coax connects to the end of that assembly with a BNC connector and that probe is assembly is  sealed. To achieve the full voltage rating with the Tektronix p6015 (non "A" version) you had to open the sealed assembly and pour freon in it. The "A" version used more environmentally friendly silicone oil.  I had used one of the old versions and they worked well but were a pain to maintain. 

[beanflying]

Had a look at the space within the current tip and roughed in the planned Resistors as per the cutaway below and have shown the maximum voltage and distance from the hand. There would be room to allow a thin copper tube perhaps covering up to 1/2 way into the 90MOhm resistor. Inside that tube would still be the multilayer wrapped PLA tube already made. Another option would be some tinned copper sleeve which I already have from some RFI shielding, it could easily be expanded over the last section of inner tube then soldered onto the common Earth/0V wires.

I most likely won't see the Resistors this side of the coming weekend so plenty of time to tweak and remake a section or two.

** I am getting 90, 400 and 500 MOhm resistors so it will be 100:1 not 1000:1

[PartialDischarge]

I don't know how far the shield goes inside the front part of the probe but the drawing of the Tektronix probe shows that that one goes about 3/4 of the distance to the tip. 

The workings of the p6015 series are different, it is a compensated oscilloscope probe and the shield is not there for safety but to 1) isolate the internal high impedance resistor from external fields, and 2) create a internal cylindrical volume where fields are controlled and HF compensation can take place.

I do not think the fluke dmm probe has the resistor covered but theres an easy way to know. Connect the probe to an oscilloscope using a banana-bnc adapter with vert scale set to 10mv/div. Now grab the probe top part with your hand and see if you can see a 50Hz signal in it. Given the thin diameter of this probe I do not think it has this part shielded.

It is mostly forgotten that any high valued resistor is a 50Hz nice pickup.

[vk6zgo]

Please understand... my primary concern here is the "open" part of this project. As soon as you put that label on something, the fuckwits start to come out of the woodworks. Building it for yourself is one thing; sharing the blueprints online is something I'd consider to be roughly akin to all the imbeciles sharing plans for 3D-printable firearms, all of which are far more likely to kill the user than anyone more deserving such fate.

I've committed numerous Darwin-award-worthy acts of stupidity with HV transformers; my grand-dad started me early helping him wind a 3KV transformer for a 160m transmitter when I was 10. 

There's a reason I don't document the details of those adventures publicly; and as anyone in the TEA will tell you, it isn't embarrassment because I have not so much as a single shred of it left in my being.

mnem
*tzzzt*

There is a counter argument to that non discussion of potentially hazardous projects given the mass idiocy on youtube with a simple search on High Voltage. Discussion here in the open is educational, imformative and generally carried out by people with a greater level of understanding than the potential Darwin Award Winners.

The maths in this one is generally ok but the practical is horrifying as are the comments below it in general

https://youtu.be/RAOw0mHQRvk

I also noted he determined the series resistor value on a low voltage range.
I seem to remember that many DMMs have lower internal resistance on such ranges, with the high range resistance around 10/11M .

As far as I know, my old Fluke 77 has the high value on all ranges, but ! may be wrong.

[bd139]

Fluke 7x and 8x meters have variable impedance on each range. It's "around 10-11 meg" and varies from meter to meter and range to range. This is one reason I originally bought a Keysight U1241C as it has constant across all ranges. Shame it otherwise sucked.

[beanflying]

Fluke 7x and 8x meters have variable impedance on each range. It's "around 10-11 meg" and varies from meter to meter and range to range. This is one reason I originally bought a Keysight U1241C as it has constant across all ranges. Shame it otherwise sucked.

It is mentioned and tested in W2AEW (Alan's) Youtube video I put up back on page one closest to 10MOhms at the highest range. Also confirmed on the 121GW it is closest to 10MOhms on the highest range. Partly why I opted for 100:1 over 1000:1

[vk6zgo]

Not  that it matters with commercial probes, because you always use the highest  range.

Many older probes were designed initially for VTVMs which do maintain the same input impedance .
That said, I have even seen ones designed for use with AVO Model 8s.

The most interesting ones came complete with their own display, in the form of an "Electrostatic voltmeter".

[beanflying]

Not  that it matters with commercial probes, because you always use the highest  range.

Many older probes were designed initially for VTVMs which do maintain the same input impedance .
That said, I have even seen ones designed for use with AVO Model 8s.

The most interesting ones came complete with their own display, in the form of an "Electrostatic voltmeter".

Interesting Pintek is still making a version of something similar http://www.pintek.com.tw/product_detail/landersound/index.php?Product_SN=19263&Company_SN=6002&Product_Site_Classify_SN=17071

Making something similar with a modded handle wouldn't take a lot mainly a case of sourcing a good quality movement. The Digital one given cheap DMM's I wonder about even more 

[vk6zgo]

Not  that it matters with commercial probes, because you always use the highest  range.

Many older probes were designed initially for VTVMs which do maintain the same input impedance .
That said, I have even seen ones designed for use with AVO Model 8s.

The most interesting ones came complete with their own display, in the form of an "Electrostatic voltmeter".

Interesting Pintek is still making a version of something similar http://www.pintek.com.tw/product_detail/landersound/index.php?Product_SN=19263&Company_SN=6002&Product_Site_Classify_SN=17071

Making something similar with a modded handle wouldn't take a lot mainly a case of sourcing a good quality movement. The Digital one given cheap DMM's I wonder about even more 

I have used one like in the first picture, but the Electrostatic ones were different again.

The display unit was remote at the end of an insulated lead like the DMM/VTVM with a normal one, but the electrostatic meter had a light source & optical system which projected the meter position onto a wide curved display screen on the front of the unit.
The cases of these units were painted flat silver, & had a strange curved "streamlined "shape,which, together with the probes, gave them very much a 1960s "Doctor Who" appearance.

I don't remember seeing them used, & now, I wish I had taken a photo of them, just for their "weirdness coefficient"!

[beanflying]

There is some odd ones out there.  Having a trawl for Capacitive and higher frequency dividers and came across this (Page 15) http://hparchive.com/Catalogs_Specialized/HP-Catalog-195x-UHF-Equipment.pdf

My 427A Manual also mentions the 11039A (was 452A) Capacitive Divider but I cant seem to find it in any details on it. HP also seem to have had a preference for dropping a 1MOhm resistor across the probe to get to their 1000:1 for their resistive ones which would help with the meter variations.