possible 100x mod for 10x probe?

[prawncrackers]

i have recently acquired a scope, my first scope, is a Hantek DSO5102P. now it was a pain in the arse to save up for it and i dont want to make a stupid mistake of blowing it up.

now my concern here is not the earthing thing but the input voltage to the scope.

according to the spec sheet max input signal:

CAT I and CAT II: 300VRMS (10×), Installation Category;
CAT III: 150VRMS (1×);
Installation Category II: derate at 20dB/decade above 100kHz to 13V peak AC at 3MHz* and above. For non-sinusoidal waveforms, peak value must be less than 450V. Excursion above 300V should be of less than 100ms duration. RMS signal level including all DC components removed through AC coupling must be limited to 300V. If these values are exceeded, damage to the oscilloscope may occur.

now ive thought that at 10x setting the voltage divider within the probe is 9meg in series with the 1meg input of the scope i can make a 100x "mod" if i parallel a 100k ohm resistor with the input impedance (to make 91k). please note that i will be using this mainly for SMPS primary side repair and occasional valve amps so i guess i dont need very high bandwidth (sub mhz) or "real" HV insulation capabilities. i may have to tape the 10x switch permanently.

is this a good quick'n'dirty idea? do i need to tweak(add/minus) the capacitance of the probe?

thanks!

[Ian.M]

That's an absolutely lousy idea. There are few less forgiving signals than the collector or drain of a chopper transistor in a SMPSU.  The DC bus of a universal voltage mains powered PSU can be as high as 340V, and its not unusual for the spikes due to leakage inductance to reach double the rail voltage.  Also, newer designs of SMPSU have switching frequencies approaching 1MHz, and the switching spikes often have frequency components an order of magnitude greater that the fundamental frequency.   

The *ONLY* way to view this sort of waveform safely is with a high voltage x100 probe.  A bodged x10 probe with an inadequate voltage rating feeding a reduced input resistance would be likely to flash over and could potentially apply the full voltage to the scope input and the person holding the probe.   One can view many of the control circuit waveforms safely with a lesser-rated probe, but extreme care must be taken to avoid probing high voltage nodes.


It is possible to build DIY HV probes, but you have to know exactly what you are doing even for something simple like 9x 10 Meg resistors in series in front of the existing x10 probe.

[danadak]

http://how-to.wikia.com/wiki/How_to_make_a_100X_oscilloscope_probe

http://cromwell-intl.com/radio/probes.html



Regards, Dana

[prawncrackers]

hmm alright, thanks for the responses.

i think ill give the build-up version a go with exaggerated creepage/insulation/sheilding and rigid RG58.   

[tggzzz]

Have a look at the high-quality probing references at https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

[tautech]

hmm alright, thanks for the responses.

i think ill give the build-up version a go with exaggerated creepage/insulation/sheilding and rigid RG58.   

Don't.
Why on earth would you want to put yourself in danger for the sake of $15 
http://www.ebay.com/itm/P4100-High-Voltage-Oscilloscope-Probe-15cm-2KV-100-1-100MHz-Test-Probe-US-NA0C-/141771437118

[krivx]

hmm alright, thanks for the responses.

i think ill give the build-up version a go with exaggerated creepage/insulation/sheilding and rigid RG58.   

Don't.
Why on earth would you want to put yourself in danger for the sake of $15 
http://www.ebay.com/itm/P4100-High-Voltage-Oscilloscope-Probe-15cm-2KV-100-1-100MHz-Test-Probe-US-NA0C-/141771437118

Rated to 2kV but there's a few mm of separation between the probe and ground ring??

[joeqsmith]

hmm alright, thanks for the responses.

i think ill give the build-up version a go with exaggerated creepage/insulation/sheilding and rigid RG58.   

Don't.
Why on earth would you want to put yourself in danger for the sake of $15 
http://www.ebay.com/itm/P4100-High-Voltage-Oscilloscope-Probe-15cm-2KV-100-1-100MHz-Test-Probe-US-NA0C-/141771437118

You get what you pay for.  Here was a review I made for that probe.

[tautech]

hmm alright, thanks for the responses.

i think ill give the build-up version a go with exaggerated creepage/insulation/sheilding and rigid RG58.   

Don't.
Why on earth would you want to put yourself in danger for the sake of $15 
http://www.ebay.com/itm/P4100-High-Voltage-Oscilloscope-Probe-15cm-2KV-100-1-100MHz-Test-Probe-US-NA0C-/141771437118

Rated to 2kV but there's a few mm of separation between the probe and ground ring??

No difference to any other 100:1 probe.
Your point is?

[krivx]

hmm alright, thanks for the responses.

i think ill give the build-up version a go with exaggerated creepage/insulation/sheilding and rigid RG58.   

Don't.
Why on earth would you want to put yourself in danger for the sake of $15 
http://www.ebay.com/itm/P4100-High-Voltage-Oscilloscope-Probe-15cm-2KV-100-1-100MHz-Test-Probe-US-NA0C-/141771437118

Rated to 2kV but there's a few mm of separation between the probe and ground ring??

No difference to any other 100:1 probe.
Your point is?

I have no qualms with building a 100:1 probe like this, but rating it to 2kV seems optimistic for such a small gap between high voltage and mains ground. Teledyne make a similar 1.2 kV probe and it looks like they use pretty similar construction. Maybe there are different requirements for Test & Measurement gear. There is a previous discussion here https://www.eevblog.com/forum/projects/oshw-diy-1kv-100mhz-differential-probe-(dilemma-vs-hope)/msg117052/#msg117052 which seems to suggest a 2kv design like this is pretty marginal 

[joeqsmith]

You could attempt to make your own probes but with only a scope to work with and it being your first scope, I would not recommend it.   It would be better to start saving for a used Tektronix probe. 

If you wanted to make one from scratch,  I have attempted to make my own high voltage wide band probes as well.  Most of them with poor results.  This was the last video I made showing the probe I currently use when testing the multimeters.  It's just a little overkill for that, but I have no concerns about damaging my scope when using it.   There are a few videos showing my build progress and one shows the schematic with the compensation network.  All of the probes I made are similar to what Northstar offers where there is no compensation at the far side of the cable near the scope.    You will note the gas discharge tubes and some TVSs built into the probe in attempt to save the scope if something major goes wrong... 



I also made some smaller ones out of some spent shell casings that are for low voltage (4KV) that work fairly well.  These were the first videos where I show some of my early attempts.

[tautech]

hmm alright, thanks for the responses.

i think ill give the build-up version a go with exaggerated creepage/insulation/sheilding and rigid RG58.   

Don't.
Why on earth would you want to put yourself in danger for the sake of $15 
http://www.ebay.com/itm/P4100-High-Voltage-Oscilloscope-Probe-15cm-2KV-100-1-100MHz-Test-Probe-US-NA0C-/141771437118

Rated to 2kV but there's a few mm of separation between the probe and ground ring??

No difference to any other 100:1 probe.
Your point is?

I have no qualms with building a 100:1 probe like this, but rating it to 2kV seems optimistic for such a small gap between high voltage and mains ground. Teledyne make a similar 1.2 kV probe and it looks like they use pretty similar construction. Maybe there are different requirements for Test & Measurement gear. There is a previous discussion here https://www.eevblog.com/forum/projects/oshw-diy-1kv-100mhz-differential-probe-(dilemma-vs-hope)/msg117052/#msg117052 which seems to suggest a 2kv design like this is pretty marginal 

OK, I get your point however I'll add that the P4100 probe I linked is made by the same company that makes the various infamous P6*** probes that many buy as everyday use cheapies of which we don't hear much bad press.

Also the OP has stated his primary use for 100:1 probes is SMPS primary side for which these P4100 probe will be quite suitable and safe.

(no affiliation with manufacturer or seller)

[joeqsmith]

It's not the "primary use" that gets you in trouble, its the other uses.   

Example, Someone recommended I buy this Vici VC99 multimeter because it was cheap and my primary use was looking at signals from my Arduino.   It worked fine for a few months and I was very happy with it.  Then I decided to attempt to measure the secondary voltage from the microwave oven with it......   

If I painfully saved for my first scope, I would not trust it to this particular probe because of the the components used in it.

[wraper]

Example, Someone recommended I buy this Vici VC99 multimeter because it was cheap and my primary use was looking at signals from my Arduino.   It worked fine for a few months and I was very happy with it.  Then I decided to attempt to measure the secondary voltage from the microwave oven with it......

Frankly, likely you would destroy expensive multimeter either.

[joeqsmith]

I destroy every meter I can, new, old, cheap, expensive.   On the plus side, you get to see the results for free. 

Example, Someone recommended I buy this Vici VC99 multimeter because it was cheap and my primary use was looking at signals from my Arduino.   It worked fine for a few months and I was very happy with it.  Then I decided to attempt to measure the secondary voltage from the microwave oven with it......

Frankly, likely you would destroy expensive multimeter either.

[LHelge]

I received a Hantek T3100 X100 probe from eBay yesterday. I think I paid less than $20 for it including shipping. Haven't used it for anything serious yet, just connected it to the square wave output of the scope and adjusted the compensation. The overall impression though is very good. Based on the look-and-feel the quality is definitely on par with the probes delivered with my Rigol DS1054Z, the only complaint is that the ground clamp is fixed.

[joeqsmith]

I tried to buy that Hantek 100x probe but they swapped out the order for the P4100.  I bought Hantek's 10X probes (2 for $20) and was very impressed with the construction and performance. 

[prawncrackers]

sorry i didnt notice this thread got bigger.  i dont do a lot of HV things so im not really looking into kVs except from maybe a really bad inductive spike.

anyway today on the way home today i thought about using 3/8 inch copper pipe and a bunch of 1meg 1% resistors. then put layers of heatshrink tubing around...
im thinking about 9meg and 100k(+1meg input imp) voltage divider which at 800V gets around 70mW heat.
for the capacitance though, maybe ill try it out with small hv ceramics or a twist-wire cap
maybe put a MOV or series back-to-back zener across the 100k?

i wonder if keeping the 2/3 inner vs outer diameter to keep the 50ohm impedance still critical?

that last video posted by joeqsmith looks promising, the RG-8 inner insulator does fit nicely inside a 3/8" tube then taper it to fit RG-58 which is still very stiff for probe use

[joeqsmith]

The shell cases worked alright.  The problem was getting them tuned.   Putting any sort of capacitance, like the coax or protection would have to be compensated for.   I do not have any protection in the small probes.  The Caddock parts I used are rated more than 4X the voltage I would normally use the probes at and with the Teflon, I am not too concerned about it.   

The end of this video shows how I made those probes.  Looks simple but getting it trimmed to get some sort of useable performance took some time.
https://www.youtube.com/watch?feature=player_detailpage&v=z7rnifjWd_U&list=PLZSS2ajxhiQCAQ6gIp6s-WoKiIEb1gHPD#t=907