Differential probe

[LA2BPA]

I'm planning to buy a differential probe, maily for troubleshooting SMPS's. In Norway our mains voltage is 230V RMS (+/- 10%, but usually is quite stabile around 230V).

I want the probe to be as "universal" as possible, so that I can use it for some low voltage stuff as well.
 
I have been looking at the different Micsigs and Dave's HVP70 probe.
The higer attenuaton ranges on the Micsig DP20003 and DP10013 aren't suited for low voltage stuff.
The DP10007 has a maximum differential test voltage at 70V (10X) and 700V (100X). As our P-P voltage is about 650V, this probe should be sufficient? (in 100X mode).

Does anyone have experiences, good or bad with these probes? Other recommendations?

[Noy]

Take care.. 700V could be enough but with an active PFC this could also be more.. And maybe a higher voltage spike? It's only CAT I 1000V / CAT II 600V against CATII 1000V..

I got the DP10013. Of coure 100x would be nicer but a blown scope isn't it worth..

[rvalente]

For higher voltage, check the fluke DP120, has been discontinued but you may find used ones.

https://www.fluke.com/en-us/product/accessories/probes/fluke-dp120

[goaty]

I like my Hameg HZ115 which I got cheap. Only 30MHz though.

[Feynman]

The HVP70 is as universal as it gets, I guess. I can recommend it, although I have the 25 MHz version with a different badge.
Just be prepared to have a wall adapter at hand, since the batteries don't last very long (probe has a DC jack).

[techneut]

You are not going to get the top-top voltage on your probes, only the top voltage.  There can be higher peak voltages in an SMPS but 700V should be enough.

[swgertsch]

I too am looking into purchasing a "high voltage" differential probe to "safely" (for me and my Rigol DS1054Z scope) investigate some noisy LED driver circuits that output ~100VDC that drive an array of LEDs. Also, I want to be able to checkout/troubleshoot my 12VDC to 120VAC "Pure Sinewave" invertor.

I have been really uneasy about doing anything above 24V, but I am willing to branch out.

It is my understanding that a differential probe will give me the freedom to safely probe between any 2 points in a circuit without having to be concerned about ground referencing.   

Micsig sells 3 very reasonably priced differential probes:

DP10007 Differential High Voltage Probe, 700V 100MHz 3.5ns Rise Time 10X/100X Attenuation Rate (~$180)
DP10013 Differential High Voltage Probe, 1300V 100MHz 3.5ns Rise Time 50X/500X Attenuation Rate (~$180)
DP20003 Differential High Voltage Probe, 5600V 100MHz 3.5ns Rise Time 200X/2000X Attenuation Rate (~$220)

First question I have is the DS1054z probe inputs are rated CAT I, 300V RMS. I understand CAT I is for relatively low energy, well protected devices under test environments. There is probably enough energy to still kill you (stop your heart) if you really are careless and too break/fry discrete components, but you should be safe from major explosive events if you stay within a CAT I test environment. Since the scope's inputs are rated to 300V RMS I get the impression there should be no problem probing a 100VDC signal, however, I am a chicken to attempt it even with a 10X probe. Am I a chicken?

So if I need/get a differential probe what model should I get? Either the DP10007, or DP10013. I do not see myself dealing with really high voltages at this point so the DP20003 model is not being considered. I am correct in assuming with larger dividers there will be a corresponding loss in signal fidelity. 

All 3 Micsig probes quote a 3.5ns rise time. I am not exactly sure what this refers to. If I had to guess, I would think it is the amount of time (delay) between what the probe sees and what is subsequently shown on the scope. I think this would only be a concern when multiple channels are being displayed together. The Micsig probed signal would be phase shifted 3.5 ns relative to the other signal(s). I am also curious if this 3.5 ns rise time is constant, or varies with signal frequency. Is this a correct understanding? 
 
 

[techneut]

Measuring the 100V is not that big a problem, just make sure you dom't touch it. The main point is were you put the ground clip because you must realise that the scoop is grounded. This is even more crucial when using 2 probes. This is where differential probes come in. They have a resistance of several Mohm to ground and each other so there is no shorting to ground.
I would suggest the DP1007 because of the lower attenuation so you can use it with smaller signals the other 2.
Rise time is how fast it can follow rising and falling signals like with a square wave. A square wave of 1Ghz would look more like a sine because it can' t follow the speed of the rises anymore.

[theHWcave]

@techneut: amen to that. Getting the ground lead connection sorted is the most difficult part.

I have a DS1054Z and hardly every use more than one input (except when checking logic timing in digital boards) because of that. Applies to all voltages not just high voltages.
For what it's worth: I have done investigations on mains powered stuff, SMPS and such but I use an isolation transformer to power the device under test (no common ground with the scope) and my self-built passive 1:1000 probe to protect the scope input. Sometimes I use the 2nd channel with a current transformer to measure current at the same time. This is maybe a bit old-fashioned and inconvenient but so far it yielded good enough results

Example: This is mains voltage and current of a USB charger. The dark blue line (power) is generated by the scope's math function.

[Purity Vendetta]

I'm a bit late to this thread but it's been very helpful as I'm looking at a set of differential probes for diagnosing SMPSs, building motor controls and my valve radio and amp projects.
With regard to protecting our scopes, would measuring voltage between domestic electrical ground and the point where the earth lead from the scope is connected help to assess the differential between ground and the connection? I'm very much a beginner so please be gentle...

[blurpy]

I've got the DP10013, and whether it works for low voltage stuff is a matter of definition and needs.

It's a bit noisy, as can be seen on this image with no signal:



The DP10007 should be less than half of that.

Turning on 20MHz bandwidth limit cleans up the signal quite a bit:



We can see the effect of the noise in this 1Vpp signal:



And that it improves a bit with the bandwidth limit:



A 3.3Vpp signal looks better:



Slightly better again with bandwidth limit:



And lastly a 5Vpp signal:



Also with bandwidth limit:



If I was only working low voltage I would get the DP10007, but DP10013 is an ok compromise I think, where it works for low voltage stuff even if not ideal.

[tggzzz]

I'm a bit late to this thread but it's been very helpful as I'm looking at a set of differential probes for diagnosing SMPSs, building motor controls and my valve radio and amp projects.
With regard to protecting our scopes, would measuring voltage between domestic electrical ground and the point where the earth lead from the scope is connected help to assess the differential between ground and the connection? I'm very much a beginner so please be gentle...

Yes, it would. However, if you connect two points at different potentials, a current will flow. In general that can ruin the probe, the scope, and the UUT. The result can be loud noises, nasty smells (and in extreme cases) followed by deathly silence.

FFI on safety, have a look at the relevant references: https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

[blurpy]

All 3 Micsig probes quote a 3.5ns rise time. I am not exactly sure what this refers to. If I had to guess, I would think it is the amount of time (delay) between what the probe sees and what is subsequently shown on the scope. I think this would only be a concern when multiple channels are being displayed together. The Micsig probed signal would be phase shifted 3.5 ns relative to the other signal(s). I am also curious if this 3.5 ns rise time is constant, or varies with signal frequency. Is this a correct understanding?

It's not about delay in the probe, it's about how fast a rise time the probe can follow in a signal. See this example of 2 square waves:



The green square wave has a rise time of 9ns and the pink has a rise time of 5ns, as in going from 0V to 1V in that amount of time. 100MHz scopes and probes can typically handle a rise time of up to 3.5ns. If the signal is any faster, like 2ns, it will just show as 3.5ns.

[EEVblog]

Micsig have advised that there will be a new version of the DP10007 coming soon, presumably to fix the CMRR issue.

[sijmen]

Micsig have advised that there will be a new version of the DP10007 coming soon, presumably to fix the CMRR issue.

Did they give any indication when the new DP10007 would be available?

Regards,
Sijmen.

[LA2BPA]

Looking at some probe specs, can someone please tell me what "+/- 700V (DC+AC peak)" mean?

I'm a little comfused because +/- (p-p) and AC peak is in the same sentence 

[blurpy]

Looking at some probe specs, can someone please tell me what "+/- 700V (DC+AC peak)" mean?

I'm a little comfused because +/- (p-p) and AC peak is in the same sentence 

There is some recent discussion about that specification here: https://www.eevblog.com/forum/testgear/new-eevblog-hv-probe/msg3628594/#msg3628594