### Author Topic: hv probe  (Read 1017 times)

0 Members and 1 Guest are viewing this topic.

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### hv probe
« on: October 30, 2019, 10:27:44 am »
Hi,ive a fluke 85v and want to build a high voltage probe for it,the input inpeadance is 10Mohm on the 1000v dc scale,i want to measure the grid bias on a scope,ie circa -1900v,whats the best way of doing it,looking at building a 100:1 divider anyone know what value resistors to use,also would a 10:1 divider be better,or would that load the dut to much,cheers Paul m3vuv.

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #1 on: October 30, 2019, 10:44:42 am »
Hi,ive a fluke 85v and want to build a high voltage probe for it,the input inpeadance is 10Mohm on the 1000v dc scale,i want to measure the grid bias on a scope,ie circa -1900v,whats the best way of doing it,looking at building a 100:1 divider anyone know what value resistors to use,also would a 10:1 divider be better,or would that load the dut to much,cheers Paul m3vuv.

With anything like this the devil is in the details, and the devil can have a very nasty bite.

If you don't know enough to be able to calculate resistor values for this, then it is highly improbable that you have sufficient knowledge to get the other details correctly implemented.

Do yourself a favour, and buy an HV probe suitable for your meter.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #2 on: October 30, 2019, 02:53:11 pm »
well,thanks for the advice,yet again waffle without answering the question that was asked,im using a string of 10 resisors of 1 megohm in an old hv probe body for the main r1 and a 1meg resistor for r2,hope it should be a 10:1 ratio?,just waiting for some fiberglass to set b4 trying it.

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #3 on: October 30, 2019, 03:11:10 pm »
well,thanks for the advice,yet again waffle without answering the question that was asked,im using a string of 10 resisors of 1 megohm in an old hv probe body for the main r1 and a 1meg resistor for r2,hope it should be a 10:1 ratio?,

QED.

You are wrong, and where EHT is involved you really shouldn't be guessing.

There is no way that I am going to be party to you doing something that might injure you. On my conscience, and all that.

Quote
just waiting for some fiberglass to set b4 trying it.

What tests are you going to do to prove that is a safe construction?

BTW, if you manage to create a voltage divider with the desired ratio, then you should read and understand this bit of the upcoming TAoE X-Chapters book: https://www.dropbox.com/s/dli62uyuldr7z5q/1x.2_Resistors_DRAFT.pdf?dl=1

Sections 1x.2.3 and 1x.2.7 are directly relevant.
« Last Edit: October 30, 2019, 03:19:52 pm by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #4 on: October 30, 2019, 03:28:39 pm »
well,its a 30kv probe body for starters,so insulation should be good for 2ky with ease,the resistors are mounted on a strip of copper clad board  with 4mm gaps between each pad where the resistors are soldered on to it,the resistor string is made up of 8x 1meg 2 watt resistors and 2 500k resistors againe 2 watt rating,had to use the 500k ones to get it bang on 10 meg ohm,its an old fluke hv probe,was duff s the 100g ohm resistor was open,the fiberglass is some p40 around the handle as it had a burn mark from a soldering iron,thats about 8mm thick,so far for a 1v meter reading i have to set my psu to 11.3 volts,i think r2 is slightly off as its the original  resistor.

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #5 on: October 30, 2019, 04:11:26 pm »
well ive replaced r2 with a 1.001 meg resistor,now for 10v in from my psu i read 0.842 vots om my dmm,any ideas on how much to change r2 by to correct this?,my maths is crap!

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #6 on: October 30, 2019, 04:23:26 pm »
well ive replaced r2 with a 1.001 meg resistor,now for 10v in from my psu i read 0.842 vots om my dmm,any ideas on how much to change r2 by to correct this?,my maths is crap!

From the information you have given, I would expect you to measure 0.833V. Since that is only 1% different, I expect that is due to resistor tolerances and meter tolerances.

Apart from that, I refer you to my previous responses.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #7 on: October 30, 2019, 04:41:42 pm »
i assume then by the time i measure 1ky the error will be 10%?any idea what value resistors i need for 1v reading for 10v in,im struggling with the maths!,cheers,btw the pcb i made is fitted into the nose of the fluke probe and is sprung loaded to stop any arcs,im sure if the original body and leads were good for measuring 20 kv then 2 kv should be well in limits,also if we say 350v over each resistor then the string should be good for 4kv,thats with a 4mm air gab between the resistors as well,how do you get !%?,i just measured the mains here just with the dmm it was 231,with my probe i get 19.3vX10 +193v,thats a difference of 17.93%,is that due to the fact im measuring ac? without any compensation caps?,not that it matters as i need this just for dc.,cheers m3 vuv

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #8 on: October 30, 2019, 05:09:35 pm »
i assume then by the time i measure 1ky the error will be 10%?any idea what value resistors i need for 1v reading for 10v in,im struggling with the maths!,cheers,btw the pcb i made is fitted into the nose of the fluke probe and is sprung loaded to stop any arcs,im sure if the original body and leads were good for measuring 20 kv then 2 kv should be well in limits,also if we say 350v over each resistor then the string should be good for 4kv,thats with a 4mm air gab between the resistors as well,how do you get !%?,i just measured the mains here just with the dmm it was 231,with my probe i get 19.3vX10 +193v,thats a difference of 17.93%,is that due to the fact im measuring ac? without any compensation caps?,not that it matters as i need this just for dc.,cheers m3 vuv

imsorryicantreadthat butimnotgoingtohelpsomeone  attemptsomethingibelieveis ,,dangerousandbeyonsd their capabilities,.;
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

The following users thanked this post: Ysjoelfir, schmitt trigger

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #9 on: October 30, 2019, 05:24:52 pm »
dont then,ill sort it myself,my text has commas and spaces,is your pc screwed?

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #10 on: October 30, 2019, 05:35:52 pm »
dont then,ill sort it myself,my text has commas and spaces,is your pc screwed?

As you can appreciate, using commas and spaces helps the reader understand what you are trying to get across.

Does your keyboard have apostrophes and upper case as well, or are you "iil"?
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #11 on: October 30, 2019, 07:08:14 pm »
Its fine on my screen smartass!,! recomend trying specsavers!!

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #12 on: October 30, 2019, 08:34:37 pm »
Its fine on my screen smartass!,! recomend trying specsavers!!

If you want to have readers that feel it is worth responding, it is wise to not put impediments in their way.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #13 on: October 30, 2019, 09:15:52 pm »
Well,if its all fine on my screen i see no reason why everyone elses isnt the same!.,maybe as im using a lappy tethered to my phone for the internet?

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #14 on: October 30, 2019, 09:47:43 pm »
Well,if its all fine on my screen i see no reason why everyone elses isnt the same!.,maybe as im using a lappy tethered to my phone for the internet?

There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #15 on: October 30, 2019, 10:39:07 pm »
I dont see the problem all the punctuation is there ,If the website fails to display it correctly,then i dont see what i can do about it!!.

#### TimNJ

• Frequent Contributor
• Posts: 810
• Country:
##### Re: hv probe
« Reply #16 on: October 30, 2019, 10:56:37 pm »
I am really not trying to be an ass, but I agree that is is difficult to read your writing. This is meant to be constructive, so I hope I don't come off the wrong way. Periods and capital letters would help a lot!

i assume then by the time i measure 1ky the error will be 10%?any idea what value resistors i need for 1v reading for 10v in,im struggling with the maths!,cheers,btw the pcb i made is fitted into the nose of the fluke probe and is sprung loaded to stop any arcs,im sure if the original body and leads were good for measuring 20 kv then 2 kv should be well in limits...

---

I assume, then, by the time I measure 1K the error will be 10%? Any idea what value resistors I need for 1V reading for 10V in? I'm struggling with the maths! Cheers. BTW, the PCB I made is fitted into the nose of the Fluke probe and is spring loaded to stop any arcs. I'm sure that if the original body and leads were good for measuring 20KV, then 2KV should be well in limits....

The following users thanked this post: schmitt trigger

#### Ysjoelfir

• Frequent Contributor
• Posts: 474
• Country:
##### Re: hv probe
« Reply #17 on: October 30, 2019, 11:23:11 pm »
One thing you could do is... well.. maybe stick to the standards used everywhere? like, writing the space AFTER the comma, ending your sentences with a fullstop, ending your sentences at all in a somehow understandable manner?
I mean, I can't be the only one who wants to rip out his eyes while trying to read what you tried to express?
Please note also that it is not really appealing to help someone who calls you a smartass and tells you you're waffling without being helpfull...

Your attitude seems to be like "Hey, you, yes you! Help me, I can't get this solved by myself. I mean, I actually can, since I am one of the smartest of all, but you know, you dumb uncle bumblefuck could do this work for me, for free, in your spare time, when I ask you to do so and the way I want you to do it. What, you're thinking this topic is over my capabilities? How dare you! I know exactly how this works!!!1", at least that is the impression I get from reading this topic and your older posts.

Please consider if this is the impression you want other people to have from you. It could be that this isn't the way you are in person, but maybe it's the way people see you because of the way you express yourself.
Greetings, Kai \ Ysjoelfir

The following users thanked this post: schmitt trigger

#### tggzzz

• Super Contributor
• Posts: 10703
• Country:
##### Re: hv probe
« Reply #18 on: October 31, 2019, 12:18:49 am »
I dont see the problem all the punctuation is there ,If the website fails to display it correctly,then i dont see what i can do about it!!.

Ah. You are channelling Eric Morecambe's famous joke, still remembered after 48 years: āIām playing all the right notes, just not necessarily in the right order.ā

« Last Edit: October 31, 2019, 12:22:47 am by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less

#### sourcecharge

• Regular Contributor
• Posts: 181
• Country:
##### Re: hv probe
« Reply #19 on: October 31, 2019, 06:30:33 am »
Hi,ive a fluke 85v and want to build a high voltage probe for it,the input inpeadance is 10Mohm on the 1000v dc scale,i want to measure the grid bias on a scope,ie circa -1900v,whats the best way of doing it,looking at building a 100:1 divider anyone know what value resistors to use,also would a 10:1 divider be better,or would that load the dut to much,cheers Paul m3vuv.

Why would you want to build a probe for that low of voltage?

They readily make probes for about 100 bucks that go up to 2000v at a 100:1 ratio for oscopes and you can use an adaptor for the banana plug to CNC connection.

That way, you can use it for both an oscope (if you have one or are planning on getting one) and your Fluke 75v dimm.

Here's an example:
https://www.digikey.com/product-detail/en/b-k-precision/PR2000B/BKPR2000B-ND/3783011

PR2000B
Datasheet
Digi-Key Part Number    BKPR2000B-ND
Manufacturer    B&K Precision
Manufacturer Part Number    PR2000B
Description    OSCOPE PROBE X100 200MHZ 100M
Manufacturer Standard Lead Time    8 Weeks
Detailed Description    100:1 Voltage Probe 200MHz 48.000" (1219.20mm) Black and Red Oscilloscope Probe Test Lead 100M Input Resistance 2000V

This one has a 200 Mhz bandwidth!

« Last Edit: October 31, 2019, 06:33:55 am by sourcecharge »

#### vk6zgo

• Super Contributor
• Posts: 5162
• Country:
##### Re: hv probe
« Reply #20 on: October 31, 2019, 08:37:09 am »
well,thanks for the advice,yet again waffle without answering the question that was asked,im using a string of 10 resisors of 1megohm in an old hv probe body for the main r1 and a 1meg resistor for r2,hope it should be a 10:1 ratio?,just waiting for some fiberglass to set b4 trying it.
https://en.m.wikipedia.org/wiki/Voltage_divider

What you have is an 11:1 ratio, so for an ideal DMM with infinite input resistance, you will read, (for 2kV),  181.182 volts.

But wait! there's more!

You don't have an ideal meter, so your DMM looks like 10 Megohms in parallel with R2, reducing the resistance of that part of the voltage divider to 0.9091Megohms.

The voltage divider is now a 10.9091: 0.9091 ratio, so 2kV will read as 166.668 volts.
That is 0.0833 times the voltage you are probing, fitting well with the figure of 0.833 volts tggzzz predicted for a 10 volt source.

There are two ways to adress this problem;-
(1) Determine the correct values required for R1 & R2 to give you a 10:1 ratio.

(2) Just go with what you've got, & apply a correction factor to the result, determined by your tests on an accurate voltage source.

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #21 on: October 31, 2019, 08:40:21 am »
Hi,ive a fluke 85v and want to build a high voltage probe for it,the input inpeadance is 10Mohm on the 1000v dc scale,i want to measure the grid bias on a scope,ie circa -1900v,whats the best way of doing it,looking at building a 100:1 divider anyone know what value resistors to use,also would a 10:1 divider be better,or would that load the dut to much,cheers Paul m3vuv.

Why would you want to build a probe for that low of voltage?

They readily make probes for about 100 bucks that go up to 2000v at a 100:1 ratio for oscopes and you can use an adaptor for the banana plug to CNC connection.

That way, you can use it for both an oscope (if you have one or are planning on getting one) and your Fluke 75v dimm.

Here's an example:
https://www.digikey.com/product-detail/en/b-k-precision/PR2000B/BKPR2000B-ND/3783011

PR2000B
Datasheet
Digi-Key Part Number    BKPR2000B-ND
Manufacturer    B&K Precision
Manufacturer Part Number    PR2000B
Description    OSCOPE PROBE X100 200MHZ 100M
Manufacturer Standard Lead Time    8 Weeks
Detailed Description    100:1 Voltage Probe 200MHz 48.000" (1219.20mm) Black and Red Oscilloscope Probe Test Lead 100M Input Resistance 2000V

This one has a 200 Mhz bandwidth!

Reason :A,im out of work and cant afford to buy one. Reason B,i already have a 1000:1 Probe,was rated at 30kv but the hv resistor is faulty,im basicaly building a new guts to fit inside the body of it!.

#### sourcecharge

• Regular Contributor
• Posts: 181
• Country:
##### Re: hv probe
« Reply #22 on: October 31, 2019, 09:34:53 am »
Get a job, da na nah na, da na nah na nah, get a job...

That sucks, hope you find some luck soon.

#### m3vuv

• Frequent Contributor
• Posts: 504
• Country:
##### Re: hv probe
« Reply #23 on: October 31, 2019, 01:57:20 pm »
Well,looks like i have sorted it as near as  dam it,swapped and experimented with resistor r2,after a few trys i finished up with r2 at 1.235mohm,on the 240v ac mains  measured without the probe i read 241.8v on my meter ac,with the probe attached i get a reading of 239.7v ac,measuring the anode voltage of my yaseau fl2100z i read 245v,it should read 240 in an ideal world ie x10,its within 50v at 2.4kv,thats near enough for what i need,73 m3vuv.

#### ArthurDent

• Frequent Contributor
• Posts: 973
• Country:
##### Re: hv probe
« Reply #24 on: October 31, 2019, 02:24:18 pm »
Well the old saying "consider the source" applies here as well. There is a very good reason why HV probes designed to measure CRT voltages use divider resistors 1000 time greater than what you have chosen. The testing you have done so far is meaningless in this application. You should follow the advice that other posters have given you and wait until you understand what you are trying to do.

Smf