Differential signal measure with standard probes disconnected from GND

[tompouet52]

Hi all !

First post here, about oscilloscope probes setup.

I have to measure an RS-485 signal.
I use two "standard" probes (1x-10x, with the GND lead/alligator clip, and the sprung hook, Lecroy PP020-2 for the reference) for that.

On the oscilloscope (Lecroy Wavesurfer 3054), I use the math "substraction" function to perform:

C1 channel (differential signal high) - C2 channel (differential signal low)

I also removed the gnd lead on EACH probe.

Can my measure be considered as "correct" in a way that it should not induced any noise/damadge comming from the gnd,
which is linked to the earth ground through the scope ?

Indeed, my signal is not really clean, although I guess that it could be partly improved by adjusting probes compensation.

If not, what could possibly go wrong with this setup ?

Thank you !

[RoGeorge]

- Yes.
- A lot can go wrong.

I would rather use an isolating transformer for the mains voltage that power the scope, or a scope that has isolated channels by design. For the first, a 1:1 galvanic isolated transformer on the 220V/50Hz mains will do the job, it's a nice to have in the lab anyway. For the second, some handheld unit (battery powered or powered from a UPS) can do the job.

[tompouet52]

Well, thank you for the quick answer !

By "going wrong" I meant destroy / damage any part of the RS-485 "network" (master and/or slave).

I assume that using true differential probes would be safer and better ?

Thank you!

[ace1903]

"I also removed the gnd lead on EACH probe."

May I ask why?
485 common voltage is low and any scope can cope with it.
Without them scope will pick up lot of noise due to incorrect grounding.
I used similar set up yo measure can signal(with scope gnd connected to common wire) up to 1mbit/s
with nice results.

[rstofer]

Look at Figure 26 in this datasheet:
http://www.ti.com/lit/ds/symlink/sn65hvd33-ep.pdf

The output drivers float the output signals at Vcc/2.  The bottom mosfet pulls the signal down to local logic ground and the top mosfet pulls the signal up to Vcc.

In my view, both outputs are ground referenced and you can connect your probe grounds to local logic ground.  Then all you have is two outputs going to two scope channels.  You can choose to do math, or not.

Even inputs from the remote end are referenced to local logic ground at the driver.  Look at the A Input schematic.  To have the bus be truly isolated would require magnetics such as is used on Ethernet or some form of fiber optics.

I'm not much in favor of floating the scope and I certainly don't plan to buy a battery operated scope.  I think I prefer to come up with a measurement plan where neither is necessary.  I've been told, but I haven't proven for myself, that PSU troubleshooting requires floating the scope.  I remain skeptical.  But I will concede that I don't work on PSUs.

[MarkL]

The short answer is "yes", a differential probe is the right way to do this measurement.


The subtract method on the scope can be used with some caveats:

- The signals being subtracted must each be within the range of the ADC on the scope.  To guarantee this, make sure that the extremes of both C1 and C2 are completely on the screen before you subtract them.  (Most scopes have some amount of ADC range beyond the screen, but it's best not to depend on it.)

- Because the signals need to be on the screen, try to find a ground somewhere on the device you're working on, if at all possible.  Otherwise you'll be picking up a lot of noise which may be much larger than your actual signal and you'll be forced to reduce the vertical sensitivity to keep it all on the screen.

- Make sure you stay within the voltage safety limits on each of the probes and scope input.  I don't know what equipment you're testing.

- There will be some mismatch between the channels, so you won't get a perfect result when you subtract them.  To get an idea of the mismatch, look at the subtracted result with C1 and C2 connected to the same signal.  Ideally it should be a flat line.


The probe is 10x and 10Mohms, so it will not affect the RS485 signal or damage it (but possibly if you go beyond the probe voltage limits of 500Vrms; see the probe owners manual for details).

And on the compensation, you usually set each probe separately to its optimal value and leave it.  In this case after you've done that, you can optionally tweak one of them by looking at the subtracted value with both probes connected to the same source, or both connected to the scope's probe compensation output.  Again, a flat line is ideal.

[mikerj]

If you are using the simple A-B math function then you are basically taking two single ended measurements, so you do need a ground reference otherwise the signal will be noisier.

You can only dispense with the grounds if the scope has a proper differential input mode, which is uncommon (particularly on cheaper scopes).

[vk6zgo]

- Yes.
- A lot can go wrong.

I would rather use an isolating transformer for the mains voltage that power the scope, or a scope that has isolated channels by design. For the first, a 1:1 galvanic isolated transformer on the 220V/50Hz mains will do the job, it's a nice to have in the lab anyway. For the second, some handheld unit (battery powered or powered from a UPS) can do the job.

He isn't looking at the AC Mains--it's only RS485,so there are no high voltages involved.
This is how it was done for decades when differential probes were very expensive.

[T3sl4co1l]

RS485 is not fully differential.

- The receivers need to see a signal that's within some volts of their supply voltages (usually +/-14V or so).

- The transmitters electrically connect the transmission cable to +V/GND alternately, to send data.  This fixes the transmitter output's common mode voltage firmly at (VCC + GND) / 2.

The oscilloscope is not fully differential.

- Each probe input channel can only resolve signals between the maximum and minimum range the ADCs can measure.  This is usually not far beyond a "full screen" reading, at whatever V/div setting is in use.  Signals outside of that range are ignored: limited to MAXVAL or MINVAL.

- Each probe will have a different ground return impedance, because they are not in exactly the same physical locations, nor are they linked together along the way (according to your description).  Thus, they will measure different (high frequency) signals as well.  That is to say, the CMRR of the probe configuration itself is poor.

- To fix the latter condition, you can join the probe grounds to each other.  Now, at high frequencies, a step change at one probe tip, will cause a current into that probe, which is carried out of the other probe.  As long as the probe impedances (from tip to tip-ground) are equal, the differential (and common mode) voltages will also be symmetrical.  This preserves CMRR, but does nothing about the former problem (scope input range).

- To fix both problems, ground the probes to the transmitter's ground reference.

One more important factor to note:

- Isolating the oscilloscope will have very little effect, if any at all, on the high frequency CMRR of the measurement.  The isolation transformer itself will have quite a low common-mode impedance to ground.  That is to say, it isn't isolating high frequencies.  And even if the transformer were ideal, the fact that the oscilloscope, and any attached cables (including the probes themselves), all take up physical space, means they also have equivalent capacitance or impedance to free space.  This provides a current-return path for common mode currents.  Thus, you will have little or no change in high frequency errors.  (The only way to truly solve that, is an ideal isolated probe, where the isolation barrier has extremely low capacitance, and the isolated probe section is no larger than the probe tips and handles themselves!)

When is it okay?

- If you don't need to resolve high frequencies (such as the risetime of the signal, or high data/clock rates), you can still use the ungrounded or bridged-ground probe method, and filter the high frequency content by enabling input bandwidth filters (probably for a cutoff frequency of 20MHz or below).  You may still see artifacts due to clipping.  You will have to decide what is real and what isn't, and mentally subtract that from your measurement.

Tim

[danadak]

These might be useful -

http://materias.fi.uba.ar/6644/info/osciloscopios/avanzado/Diferential%20oscilloscope%20measurements.pdf

http://www.newark.com/pdfs/techarticles/tektronix/FFM.pdf


Regards, Dana.

[RoGeorge]

...it's only RS485,so there are no high voltages involved.

Indeed, if everything works properly, it's safe to assume that.
If the measurement is done on an equipment installed in the field, nothing is guaranteed.

[Jeroen3]

CAN bus is electrically similar, and a dual channel scope is perfectly capable of displaying this using it's two single ended probes.
(channels on AC)

[tompouet52]

Well, thank you all,

It seems that I have strongly under-estimated the consequences of what I thought was a simple setup...
Thank you for all the information, I have now to read and understand all of that

Have a nice day !