Differential probe advice

[jakesvdm27]

Hi, I would like some guidance on which differential probe I should buy. I am sure this is not a straight forward answer, but let me try to explain where I am at right now.

Some background - Electronics is a hobby of mine. One day I decided I need to bite the bullet an get myself an oscilloscope. Back then I was faced with the same problem, and that was which one I should get. I knew nothing about this topic, and there are many choices out there. Luckily this was a simple problem to solve, since Dave did a review on the Rigol 1054 and it just happened to be affordable. So I saved up for it and now own one.

Since I knew nothing about scopes, I started watching videos on the topic and soon realised that I could get hurt if I fall into the trap for young players, as Dave explained in his video on how not to blow up your scope. Guess what, I am a young player here. The bottom line is that I have only used this scope once. That was to read a signal on an arduino board powered by a battery just to see if the scope works. The truth is that I am afraid to connect it to anything not powered by a battery after I watched that video of Dave.

What do I want to do with my scope - I need to probe around on a PC board of my mom's embroidery machine. The issue is that the power supply is ground referenced, as it plugs directly into mains (I also measured it). So this is a good time and excuse to take the next step and get myself a differential probe. This way I can use my 1054 safely.

There are so many things to look out for when it comes to differential probes, and to be honest, I am not sure what to look for.

All I want to do is be able to poke around a pc board with my scope probes and be safe. I.e, don't blow up anything or hurt myself. I expect the signals I would be looking at will all be less than 24v. The current PC board I am looking at has a 24v power supply (dropped down from mains), with a 5v voltage regulator on the board. So 24v into the board, then dropped down to 5v for the IC's on the board.

Considering I would only be looking at 5 - 15v voltage signals (most I have ever measured), but want to do it safely, would the HVP70 be a good choice for me ? Or is there something else I should look at ?  BTW, I see this is out of stock right now on the EEVBlog store.

Your help is much appreciated.

[Phil1977]

First, it´s great you care about avoiding ground loop short circuits - that´s already the most important part!

Then, with a state-of-the-art differential probe like the mentioned HVP70 or MICSIG you can't do anything very wrong. These things have high input impedance and quite some voltage tolerance, so you really can measure with them similar to measuring with the multimeter.

But: For low-voltage DC systems it´s not really necessary. You can also start by probing on your boards with the "x10 setting" on the tip and without any ground clip. That´s as safe as measuring with the differential probe, and as long as the scope and the tested PCB are ground referenced as you describe, you will also get reasonable results - though with more noise as if you would connect the ground clip.

As soon as you feel safe about the voltage potentials in the system you also can use the ground clip - if the scope and the tested board have no potential difference, nothing bad will happen. Just ensure the ground clip is always clipped to 0V potential, never to anything else.

The differential probe is clearly necessary if you want to probe on the primary side of the power supply. There you have lots of circuit that are on high potential against ground. If you connect your ground clip there - like mentioned in Daves video - you will blow fuses and more.


PS: The Rigol 1054 still is a very capable instrument and perfect as well for a beginner as for many advanced tests. It can accompany your electronic projects for a long time.

[jakesvdm27]

Thank you for the advice, it is really appreciated.

So anyone of the following will do then:
DP10013
MDP 700
MDP1500 (I see on Amazon it is nearly the same price as the MDP700)
HVP70 (when it is available again)

[tatel]

A fast google search about "eevblog differential probe " will give you quite a few answers, both videos and forum threads. Look for the CMRR spec. Attenuation could also be relevant. Differential probes are not that cheap so you might want to learn the basics before pulling the trigger.

[jakesvdm27]

Thank you for the videos, will watch them tonight.

Really appreciate all the help !!

[Phil1977]

If you just start doing these measurements you won't notice the differences in CMRR or frequency response too soon.

Of course, the attenuation factor and the voltage levels should be allright from the beginning.

And, as mentioned, you can always start by using the normal probes without their ground clip. You can even simulate a differential probe by measuring with two standard probes and using the MATH-function of your scope to do the subtraction - though a proper "real" differential probe will have much lower noise.

[ledtester]

What do I want to do with my scope - I need to probe around on a PC board of my mom's embroidery machine. The issue is that the power supply is ground referenced, as it plugs directly into mains (I also measured it). So this is a good time and excuse to take the next step and get myself a differential probe. This way I can use my 1054 safely.

Where else on the circuit board are you thinking the scope's ground clip? I would argue that it doesn't make sense to and it's not necessary to connect it anywhere else except for the power supply ground.

[David Hess]

Thank you for the advice, it is really appreciated.

So anyone of the following will do then:
DP10013
MDP 700
MDP1500 (I see on Amazon it is nearly the same price as the MDP700)
HVP70 (when it is available again)

Any of those high voltage differential probes will work for general purpose applications, and it is a good idea to make a differential measurement when there is any risk of a ground loop.  The ones with lower attenuation, while having sufficient voltage range, will have lower noise, so the HVP70 is likely the best option unless measurement above +/-700 volts is desired.  The DP10013 and MDP1500 have much higher noise, consistent with their attenuation, so I would avoid them.

DP10013   x50/x500
MDP700   x20/x200
MDP1500   x50/x500
HVP70   x10/x100

High frequency common mode rejection is often the most limiting factor in measurement accuracy, however this is difficult to evaluate from datasheets, and probably does not matter in your application.

[pdenisowski]

My personal opinion is that the best "accessory" for a hobbyist scope user is a differential probe - there are many, many applications for them and they are relatively cheap insurance to avoid blowing up your scope or hurting yourself. It really is money well-spent.

I've used lots of differential probes: both our own (which are not hobbyist-priced) and those from other manufacturers, and for hobbyist purposes the "hobbyist" level differential probes work just fine.  As some other posters have mentioned, specs like CMRR are probably not going to be that important for what you are working on at that level.

Incidentally, I did videos both about differential probes and CMRR

[pdenisowski]

And, as mentioned, you can always start by using the normal probes without their ground clip. You can even simulate a differential probe by measuring with two standard probes and using the MATH-function of your scope to do the subtraction - though a proper "real" differential probe will have much lower noise.

Using two single-ended probes (the "quasi-differential" approach) actually works reasonably well for many "non-demanding" applications.  Biggest issue in my experience with this approach is running out of hands

[David Hess]

Using two single-ended probes (the "quasi-differential" approach) actually works reasonably well for many "non-demanding" applications.  Biggest issue in my experience with this approach is running out of hands

I have gotten very good results using the add and invert method with 2 channels and 2 probes, but only with analog oscilloscopes (1) where the analog vertical variable control and compensation could be used to adjust the DC and AC common mode rejection.  It is almost unsettling when doing this calibration to see the displayed noise decrease with a sharp null.  DSOs tend to have problems with quantization noise being amplified if this method is used.

I test a modern Tektronix high voltage differential probe and it just worked.  I assume that would be the case with the other major manufacturers.

I've used lots of differential probes: both our own (which are not hobbyist-priced) and those from other manufacturers, and for hobbyist purposes the "hobbyist" level differential probes work just fine.  As some other posters have mentioned, specs like CMRR are probably not going to be that important for what you are working on at that level.

Hobby level differential probes seem to have problems with AC common mode rejection getting worse over a time period of months to years, without any support for recalibration.  I suspect this is caused by humidity interacting with the FR4 substrate used for the high impedance divider, but it could be aging of the FR4 substrate.  Not all FR4 substrates are suitable for high impedance attenuators.

(1) Or old DSOs with analog vertical variable controls where the sum operation occurs in the analog domain before digitization.