EEVblog #85 – High Voltage Oscilloscope Probe DesignPosted on May 8th, 2010 8 comments
Doug Ford shows us how he designed his own 1000:1 high voltage oscilloscope probe.
and the Oscilloscope probe tutorial:
Sweet! I like it very much, particularly the diagnostic info in the video.
You do realise that Ugly Dugly has sex with electrons and that any second hand ones you work with have been stained by him first.
Seriously, well done!
Very nice, thanks for putting that up.
I’m currently looking into designing some over voltage protection and I’ve been looking for a way to test it.
Any chance you could hook that probe up to something with proper input over voltage protection, like a multimeter or an Ethernet surge protector and see what the pulse looks like at the output (that ought to be protected) when firing the surge generator into it?
You can get a used Tek P6015 for not that much on ebay ($100 or so). The freon is usually missing, but even without, it’s rated to at least 10kVp-p or so (derated with frequency). I’d expect the frequency response to be much better, and the circuit loading to be lower. I doubt whether the bandwidth of that probe is really 30MHz, the .35/risetime formula is only valid for a 1-pole -6dB/decade roll-off, which this is clearly not.
A lot of useful information about designing probes, including high-voltage ones (about 40 years old, but still relevant today): http://bama.edebris.com/download/tek/probes/probe%20circuits.djvu (for a djvu viewer, see for example http://windjview.sourceforge.net/ )
Thumbs up, an extremely educating episode, I must say. Thanks to Doug Ford for sharing his expertise.
Very interesting ! But be careful there!
Are you certain that those ceramic capacitors are safe?
1. Mechanical stress can easily rip of the metallized ends.
2. Are they Rated for continuous high voltages (mains and up) with approval for being operating by humans?
EMI filtering caps (which go to the Protective Earth pin of the Mains) need Y1 class grade caps in the Netherlands. If you don’t, it is considered a crime. Also it won’t pass any approval tests.
Im just saying, be safe! Don’t assume 3.5kV rated caps are safe for humans. They probably won’t be blowing up under the specced conditions, but mode their failure may not be safe at all!
Thanks for all your comments – Brief reply below:
The PSU has been returned to my client; No opportunities for further tests until I build another one.
Buying a 2nd-hand probe is expedient, but I wanted to make my own anyway.
Thanks for the link; I’ll get the djvu viewer shortly.
I would have been happy with a few MHz BW for my application; Achieving high BW in this first cut was a bonus. There was no risetime difference between my HV probe and 100MHz x10 probe when viewing 30ns risetime of my 3MHz gen.
Response of my probe is consistent with stray cap of 3pF per 10M resistor; I should get a substantially flat response with compensation cap of 750pF instead of 2nF…
Let’s see how the MkII probe works!
Using 5 x 3kV caps, I’d be reluctant to use this probe at much more than 5kV; The voltage sharing across the caps isn’t equal, due to stray capacitance. And I certainly wouldn’t commercialise this probe – you’re 100% right about safety & approvals!
I love (and miss) watching people draw schematics. I wish there was a web page somewhere of flash animations of common circuits being drawn, with a commentary of WHY the circuit is the way it is.
Take the top 50 circuits out of the “Art of electronics” and animate the form and functionality!
Leave a reply