Topics

Rail Probes?

dnmeeks
 

Hey team -
I changed roles at work and I ran into a new animal called a rail probe. I have never seen one of these things before, even though I have been doing analog and power for a few decades now.
Does anyone know what's inside one of these? I see the input impedance is relatively low, and it gets lower with frequency... and the attenuation is something like 1.2.
It would help me understand their usefulness if I could see the circuitry inside - compared to a good passive probe with proper shielding.
Thanks
dan

Bruce Atwood
 

Data sheet:
https://www.tek.com/datasheet/active-power-rail-probes

Interesting, but U$4300.00 for the cheap one. Accessories are extra.

dnmeeks
 

Thanks, yeah I see the data sheet. I was hoping to know exactly how it's made. What's inside. The real scooby.

Ed Breya
 

I had never heard of these, but from the data I'd say their usefulness is mostly in the form of convenience. The exact same kind of measurements have been done before, by many, for decades - I've done it. You measure the DC and low frequency content with a regular scope input, and the high frequency content with an AC-coupled 50 ohm wide BW input. These "power rail" probes just combine the measurements into one channel, and add DC offsetting - look at the 7A13 for an earlier example.

They are intended for low impedance sources, like power supply outputs, to look at interference, ripple, and noise. There is a generalized term for all of this, called PARD. If you want to see really high frequency content, you use 50 ohm wide BW gear - a scope for time domain, or an SA for frequency. The important thing is to isolate the DCV of the supply by AC-coupling, and most importantly, to protect the equipment from surges that occur during connect/disconnect, power up/down, and PS or load faults, etc.

In the block diagram, it may be shown simply as AC-coupling, but in the details you would find that the HF signal path is all RF, 50 ohm environment, including the RF coupling cap, with sufficient voltage rating.(not just any old cap), and various protection circuitry on the 50 ohm output side going to the equipment.

So, if you can live with the inconvenience of using a regular scope input or a 7A13 for DC and LF, and having to rig up a fairly simple AC-coupling and protection circuit (the fanciness and complexity depend mostly on the required BW) for HF, you can save the 4 grand or so.

Also note that the key to HF performance and CMRR is to carry the signals directly in coax, with minimal impedance connections. You can get some ideas looking at the optional accessories and how they attach to the DUT - that's the part that really makes it work.

Ed

Jim Ford
 

PARD being an acronym for Periodic And Random Disturbance, for those who haven't seen the term before. Jim Ford Sent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------From: "Ed Breya via groups.io" <edbreya=yahoo.com@groups.io> Date: 4/23/20 2:43 PM (GMT-08:00) To: TekScopes@groups.io Subject: Re: [TekScopes] Rail Probes? I had never heard of these, but from the data I'd say their usefulness is mostly in the form of convenience. The exact same kind of measurements have been done before, by many, for decades - I've done it. You measure the DC and low frequency content with a regular scope input, and the high frequency content with an AC-coupled 50 ohm wide BW input. These "power rail" probes just combine the measurements into one channel, and add DC offsetting - look at the 7A13 for an earlier example.They are intended for low impedance sources, like power supply outputs, to look at interference, ripple, and noise. There is a generalized term for all of this, called PARD. If you want to see really high frequency content, you use 50 ohm wide BW gear - a scope for time domain, or an SA for frequency. The important thing is to isolate the DCV of the supply by AC-coupling, and most importantly, to protect the equipment from surges that occur during connect/disconnect, power up/down, and PS or load faults, etc. In the block diagram, it may be shown simply as AC-coupling, but in the details you would find that the HF signal path is all RF, 50 ohm environment, including the RF coupling cap, with sufficient voltage rating.(not just any old cap), and various protection circuitry on the 50 ohm output side going to the equipment.So, if you can live with the inconvenience of using a regular scope input or a 7A13 for DC and LF, and having to rig up a fairly simple AC-coupling and protection circuit (the fanciness and complexity depend mostly on the required BW) for HF, you can save the 4 grand or so.Also note that the key to HF performance and CMRR is to carry the signals directly in coax, with minimal impedance connections. You can get some ideas looking at the optional accessories and how they attach to the DUT - that's the part that really makes it work.Ed

John Gord
 

In addition to good connection methods (very short grounds, etc.), a valuable technique for these measurements is to loop the probe cable a few times through a large ferrite core (split cores are easiest). This minimizes the added high frequency noise from ground currents between the DUT and measuring scope. I learned this from a designer of multi-kilowatt power converters.
--John Gord

On Thu, Apr 23, 2020 at 02:43 PM, Ed Breya wrote:


I had never heard of these, but from the data I'd say their usefulness is
mostly in the form of convenience. The exact same kind of measurements have
been done before, by many, for decades - I've done it. You measure the DC and
low frequency content with a regular scope input, and the high frequency
content with an AC-coupled 50 ohm wide BW input. These "power rail" probes
just combine the measurements into one channel, and add DC offsetting - look
at the 7A13 for an earlier example.

They are intended for low impedance sources, like power supply outputs, to
look at interference, ripple, and noise. There is a generalized term for all
of this, called PARD. If you want to see really high frequency content, you
use 50 ohm wide BW gear - a scope for time domain, or an SA for frequency. The
important thing is to isolate the DCV of the supply by AC-coupling, and most
importantly, to protect the equipment from surges that occur during
connect/disconnect, power up/down, and PS or load faults, etc.

In the block diagram, it may be shown simply as AC-coupling, but in the
details you would find that the HF signal path is all RF, 50 ohm environment,
including the RF coupling cap, with sufficient voltage rating.(not just any
old cap), and various protection circuitry on the 50 ohm output side going to
the equipment.

So, if you can live with the inconvenience of using a regular scope input or a
7A13 for DC and LF, and having to rig up a fairly simple AC-coupling and
protection circuit (the fanciness and complexity depend mostly on the required
BW) for HF, you can save the 4 grand or so.

Also note that the key to HF performance and CMRR is to carry the signals
directly in coax, with minimal impedance connections. You can get some ideas
looking at the optional accessories and how they attach to the DUT - that's
the part that really makes it work.

Ed



K9FFK
 

Dan:

"TSP #76 - Keysight N7020A 2.0GHz Power Rail Probe Review, Teardown and Experiments"

<< https://www.youtube.com/watch?v=d_Ybe6xnMIg&feature=youtu.be >>

Part of this presentation gets into the probe itself. Might be informative.

ymmv

Dick K9FFK

On 4/23/2020 2:03 PM, dnmeeks wrote:
Thanks, yeah I see the data sheet. I was hoping to know exactly how it's made. What's inside. The real scooby.