On 19 August 2010 20:20, Forrest W Christian wrote: > =A0These aren't too bad... the 700V ones (way more than I need) are $332 > for the 25Mhz ones. =A0Little nervous about the 25Mhz rating, since I dea= l > with things like 56VDC square waves where the fall time is in the tens > to hundreds of nanoseconds category. =A0But the 'rise time' spec seems to > be ok that way. > > I will say that this whole 'development exercise' has given me a whole > new appreciation for how painful it can be to validate a design when > your scope is lying to you. =A0Not that I didn't know that before, but > sometimes reinforcement by pain is helpful to drill home a lesson > learned over and over again. > > -forrest > > > On 8/19/2010 12:23 AM, Sean Breheny wrote: >> Hi Forrest, >> >> A couple of years ago at work I bought two $500 25MHz differential >> scope probes from www.probemaster.com. They were one of the best >> purchases I ever made. I often use them now even when ordinary scope >> probes would be acceptable because the signals are so much cleaner >> (not just because of the narrower bandwidth - I mean that the signals >> are a more faithful representation of the actual signal). >> >> These type of probes are NOT cheap - they are very specialized, rugged >> differential amplifiers. You probably cannot just buy a diff amp IC >> with these specs. I suspect that, inside the probe, there are some >> discrete transistor input stages with matched transistors and >> resistors, followed by some high-speed op-amps, all connected in an >> instrumentation amplifier topology. >> >> For home use, I did manage to get an old (1970s vintage) 100MHz Tek >> diff probe for about $100. The downside is that it is bulky (has a >> separate power supply, probe, and amplifier box). The going rate for >> used modern diff probes seems to be about $300. New they are in the >> $500 range - this is for 20 to 30MHz BW. The really high end ones (GHz >> BW) are thousands of $. >> >> Sean >> >> >> On Thu, Aug 19, 2010 at 12:00 AM, Forrest W Christian >> =A0wrote: >>> =A0 My most recent project has created a new challenge for me. =A0 I us= e a >>> TDS2014 4 Channel DSO on the bench, and to date, just some good garden >>> variety 1X/10X scope probes (basically decent quality but not really >>> name-brand probes - maybe pomona or tenma or something like that). >>> >>> This new project involves putting 120us holes in 60VDC. =A0The stuff is >>> basically done at this point, but I think I've spent almost more time >>> dealing with scope grounding issues than actually solving problems. >>> Switching 60VDC off and on that quickly obviously creates some >>> interesting inductive possibilities. =A0For instance, earlier today I w= as >>> seeing some stray spikes on one of the signals being monitored. =A0 Pul= led >>> the scope probe out of it's 'grabber clip thingy (yes technical I >>> know)', and touched the tip to the grounding clip (you know, the one >>> with about 3 inches of wire attached directly to the probe itself). >>> Amazingly, the spikes were *still* there. =A0 Ended up disconnecting al= l >>> the other grounds on the other probes and referencing only to that one >>> to verify that signal was clean. =A0Of course, then all the other signa= ls >>> weren't clean anymore - at least not on the scope. >>> >>> So, I figured to fix this in the future, I'd spend a bit and get some >>> differential scope probes. =A0 So I glanced on ebay and google for cost= s >>> and availability - and in the process about fell out of my chair in >>> shock. =A0 Maybe there's a decent one out there at a reasonable cost th= at >>> I'll find, but looks like they're few and far between. >>> >>> So, before I either just suffer - or go out and buy a few isolation >>> amplifier ICs and build my own, I figured I'd ask on here if there are >>> either some tricks I'm missing (I.E. maybe it's the scope probes I'm >>> using - a $30-50 probe just isn't good enough, OR there is some other >>> way to get a good picture of what a signal looks like referenced to >>> ground near the chip producing it - when you have 4 possible grounding >>> points. >>> >>> Ideas? >>> >>> -forrest >>> >>> >>> -- >>> http://www.piclist.com PIC/SX FAQ& =A0list archive >>> View/change your membership options at >>> http://mailman.mit.edu/mailman/listinfo/piclist >>> Forrest If you have two identical probes, try using them in (A-B) format. Most scopes will let you set this mode (sometime you have to invert the B channel & use (A+B)) and provided the inputs are the same and the probes are the same it can give reasonable results. You may need to check & adjust the frequency compensation on the probes also. It won't beat a true differential probe but will be a lot cheaper. Or - assuming the 60V supply is at lowish impedance you may be able to make one using a suitable amplifier. Many RF amplifier chips have wide bandwidth differential inputs so an amplifier and a handful of matched or close tolerance resistors for a divider may also work out cheaper. RP --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .