Differential probe for offline PFC'd PSU

[Faringdon]

Hi,
The following diff probe used to be some £180 just  4 years ago....

https://uk.farnell.com/pico-technology/ta041/probe-active-differential-powered/dp/1667343?st=differential%20probe

....now £319 !!!

As far as i am concerned, its now time to take my Rigol 40MHz scope, and simply and cheaply  snip the earth lead from the  mains cable that supplies it...and just be jolly careful i dont touch the scope ground clip or scope BNC whilst the scope is powered. Since AYK, when earth lead is snipped....the  scope ground is then connected to the capacitive divider point of the input filters' Y caps...which is 2 Y caps in series, across Line-Neutral.

I know people say the scope can be damaged like this....but i worked at a big electronics company where they routinely did this and had no problems.
People say the Y caps can get overvoltaged if a transient occurs when the scope earth lead is snipped...., but i havent seen it happen.

[BrokenYugo]

Why do you think the rules don't apply to you?

[tautech]

Calling Darwin.....award presentation coming up and your services will be required !

[Fulmir]

If you're gonna do something silly with a big piece of equipment why don't you try building your own dif probes instead? That at least seems less likely to fry you or the scope...

[nez]

I see the Micsig DP10007 for £134.52 on Banggood, with apparently free shipping to UK..  I'm sure the noise and CMRR isn't quite as good, but it has a much higher bandwidth and much better price (as well as being better than nothing!)

[nali]

Why don't you isolate the DUT, not the scope?

https://www.screwfix.com/p/carroll-meynell-3000va-intermittent-isolation-transformer-230v-230v/320hv

120 quid.

[Faringdon]

Thanks, the Micsig DP10007 looks good , but shipping time is over a month unfortunately.

The isol txfmr looks good, but i wonder say you dont need it any more in 3 months time, is there anywhere you can drop it off if you cant sell it?..eg back at screwfix, whatever?

[H.O]

I'm not really into "high voltage" stuff, like in the kV range (and I do have a Pintek DP65 Pro that's specified 1300V) so I was going to get a DP10007 due to it having a 10x range but then I just couldn't help myself and got the new DP750-100 instead. No perfomance testing done but it works and it sure has a nice look and feel to it :-)

EDIT: Eleshop in the Netherlands have both the DP10007 and the DP750-100 in stock (that's where I bought my DP750-100).
Price might not be as low as Banggood but you'll have it withing a few days.

[Faringdon]

Isol transformer
https://www.screwfix.com/p/carroll-meynell-3000va-intermittent-isolation-transformer-230v-230v/320hv

...i realsie that the secondary "neutral" of these isol txfmrs is always  connected to earth....so in fact i will have to ensure that i take it apart and unscrew that earth connection first.

[Faringdon]

EDIT: Eleshop in the Netherlands have both the DP10007 and the DP750-100 in stock (that's where I bought my DP750-100).

Thanks, i noticed DP750-100 and DP10007 both go to 700V  but DP750-100 does it from 500:1 and 50:1, but DP10007 does it from 100:1 and 10:1.

So i would say the DP10007 is more accurate?
(they both cost the same on Eleshop)

https://eleshop.eu/test-measure/oscilloscopes/probes/differential-probes.html

[H.O]

Nah, I wouldn't say more (or less) accurate based on attenuation factor alone. Gain accuracy is specified as 2% for both probes.
But measuring something like a 3V differential bus signal with a 10:1 probe will produce 300mV input into the scope while with the 50:1 probe it's only 60mV.

If both probes generates the same amount of noise then clearly having a 300mV signal is "better" than having a 60mV signal and this why I originally wanted the DP10007.

I'll see if I can find some time tonight to run some experiments.

[nali]

Isol transformer
https://www.screwfix.com/p/carroll-meynell-3000va-intermittent-isolation-transformer-230v-230v/320hv

...i realsie that the secondary "neutral" of these isol txfmrs is always  connected to earth....so in fact i will have to ensure that i take it apart and unscrew that earth connection first.

  Where did you get that notion?

We have a couple at work, visually identical though I don't know if they're actually the same. The Earth connection is linked through from the Primary to the Secondary sockets. No connection at all between Earth & the secondary windings.

[Faringdon]

The Earth connection is linked through from the Primary to the Secondary sockets.

......so if powering the scope from the isol txfmr, you can just not connect the earth to the scope.
..in fact, that sounds the cheapest way...just need an ~200W isol txfmr...then scope away and just careful not to touch scope ground whilst powered up.

[wasedadoc]

I power the PSU being investigated from a £40 UPS. UPS unplugged from AC mains and all other output loads. Also use a cheap 100:1 probe. Scope remains earthed as normal. Still not a recipe for poking about with gay abandon.

[BrokenYugo]

The Earth connection is linked through from the Primary to the Secondary sockets.

......so if powering the scope from the isol txfmr, you can just not connect the earth to the scope.
..in fact, that sounds the cheapest way...just need an ~200W isol txfmr...then scope away and just careful not to touch scope ground whilst powered up.

That solves nothing. Do it right before you get yourself killed.

[H.O]

Here's the DP750-100, set to 50:1, measuring a 3V 125kbit/s CAN-bus signal (in a fairly "clean" environment).

[Faringdon]

That solves nothing. Do it right before you get yourself killed.

Thanks, but it solves  the fact that i am skint after spending over £1k getting this project going......i will just  be careful not to touch the scope ground whilst the scope is powered.
A 200W isol txfmr to power the scope with is the cheapest way.
And the local burglars are less likely to steel a 200W isol txfmr.

...no earth to the scope. Being very careful not to touch as discussed.

(I will be using my soldering iron, to inscribe  "This is stolen if youre  not xxxxx" into the plastic on my Tenma 72-8225 40MHz scope)

[bc888]

Dave's Eevblog HVP70 has better bandwidth and is cheaper is it not? I didn't check all of the specs but your linked pico appears to be 25Mhz vs 70Mhz for the HVP70. The other question, I suppose, is what does $450 in AUS vs your currency once you have it in your hands. Appears to be in stock currently. What am I missing here...?

https://www.eevblog.com/product/hvp70/

[Geoff-AU]

I bought a Micsig diff probe.  I'm too young to die.

[TMM]

That solves nothing. Do it right before you get yourself killed.

Thanks, but it solves  the fact that i am skint after spending over £1k getting this project going......i will just  be careful not to touch the scope ground whilst the scope is powered.
A 200W isol txfmr to power the scope with is the cheapest way.
And the local burglars are less likely to steel a 200W isol txfmr.

...no earth to the scope. Being very careful not to touch as discussed.

(I will be using my soldering iron, to inscribe  "This is stolen if youre  not xxxxx" into the plastic on my Tenma 72-8225 40MHz scope)

If you're going to isolate the scope, instead of isolating the DUT or buying a differential probe, you might as well just cut the earth wire out of the scope lead and call it a day since it is even cheaper and the chance of dying is more or less the same.

I would buy the DP10007. It's a cheap and idiot proof solution to the problem.

[n1ist]

Someone tried to float a scope at work.  I then had to help them remove the scope from the grounded box they had it sitting on with a crowbar and saw after the rubber feet failed and the scope spot-welded itself in place.

Remember that not only will metal parts of the scope be hot, but the insulation on the rest of the scope (and mouse and keyboard, etc.) aren't designed for high voltage isolation.  Also, don't try connecting the ground clips of different probes to different places or the probe cables will light up.

If you must float a scope, use one of the battery powered ones that are designed for floating; they are aimed more towards electricians than electronic design, but they would be safe.  Otherwise, float the DUT and use diff probes.
/mike

[jc101]

Telonic have several differential probes in stock...

https://telonic.co.uk/product-category/test-measurement/test-measurement-manufacturer-accessories/micsig-accessories/

[Faringdon]

but the insulation on the rest of the scope (and mouse and keyboard, etc.) aren't designed for high voltage isolation.

Thanks,
The one thing that coudl get popped by floating only the scope, woudl be the scopes Y caps....but i reckon a 1MEG 1kv resistor from scopes (isolated) "neutral", connected to real neutral, woudl prevent any overvoltage to the said Y cap.

Excellent points about not touching parts of the scope...the thing is trying to remember to turn off the  DUT every time you move the scope ground clip lead thing.......also, turning off the scope would be best in that situation...but would be a pain as v/div etc woudl all need resetting every time you move the probe.

[jc101]

but the insulation on the rest of the scope (and mouse and keyboard, etc.) aren't designed for high voltage isolation.

Thanks,
The one thing that coudl get popped by floating only the scope, woudl be the scopes Y caps....but i reckon a 1MEG 1kv resistor from scopes (isolated) "neutral", connected to real neutral, woudl prevent any overvoltage to the said Y cap.

Excellent points about not touching parts of the scope...the thing is trying to remember to turn off the  DUT every time you move the scope ground clip lead thing.......also, turning off the scope would be best in that situation...but would be a pain as v/div etc woudl all need resetting every time you move the probe.

You're working alone, in a closed room, what happens to you if you do manage to fry yourself or the kit?  Anyone else about know what to do to revive you?  What is the protection of the power in your room, could you trip an RCB or RCBO (assuming it actually has protection if course...) with an isolated scope?

You life I suppose, but right tools for the job and all that seems to apply here.

[nez]

If the reason to not order from Bangood is the shipping delay, why not order one from the shop that jc101 linked to?  The prices are good there too.

[Faringdon]

Thanks yes i must admit those Telonic prices are  a lot better than some others.

[Faringdon]

Thanks for these great answers, and thanks jc101  i now have the micsig diff probe and its very good.
Of course, when i probe the current sense resistor (burden) of my PFC it is just full of pickup as these diff probes have each probe on separate lead......so i think i will still need to buy an isolation transformer so that i can use a  home-brew coaxial probe to look at it without so much common mode junk in the way.

Alternatively i woudl like to take the back off the scope, measure the DC output of the offline SMPS... disconnect it, then feed in a isolated DC supply and then i have a floating scope.
Its a Tenma 72-8225

[nctnico]

Thanks for these great answers, and thanks jc101  i now have the micsig diff probe and its very good.
Of course, when i probe the current sense resistor (burden) of my PFC it is just full of pickup as these diff probes have each probe on separate lead......so i think i will still need to buy an isolation transformer so that i can use a  home-brew coaxial probe to look at it without so much common mode junk in the way.

Alternatively i woudl like to take the back off the scope, measure the DC output of the offline SMPS... disconnect it, then feed in a isolated DC supply and then i have a floating scope.
Its a Tenma 72-8225

Don't do that (unless you have a death wish). Buy a current probe instead!

[Faringdon]

Thanks, the output current of the secondary  is just 100mA peak, so i dont think i can buy current probe to see that small?

[jc101]

The current probes from Teleonic will see that, I was using them down to 15ma just fine. I can look the model number out tomorrow.

[nctnico]

Thanks, the output current of the secondary  is just 100mA peak, so i dont think i can buy current probe to see that small?

No problem at all. Alternatively you can use a current transformer in series with the current sense resistor. Wurth has several nice current transformers that work up to several MHz.