Using scope terminations

[FreshmanNewbie]

I am using a scope that has two termination inputs. 1Mohm and 50ohms internal termination on the probe inputs.

My question is :

1. Suppose I connect the probe to a USB differentia signal. I connect only 1 probe to D+. The ground of the probe is connected to the ground. Which termination scope mode should I use? 1Mohm or 50ohms?

I am not aware on what would be the single ended impedance of D+ line with respect to ground. So, I am not sure which termination setting I need to keep. So, I assume I need to keep 50ohms mode.

2. Suppose, I use a differential probe, I need to connect the differential probe terminals to both ends of the USB lines. I need to make sure that the impedance of the differential probe should be 90ohms matching the USB differential impedance. Correct?

 

[KlausST]

Hi,

doing HF measurements is not a simple task.

Thus there is no simple answer.

****
USB should be terminated at both sides. (I expect it to be)

So when you add your scope in high impedance mode: You add a new (unterminated) branch. Causing an echo (at the scope inside end).
When you add your scope in 50 Ohm mode, then you also add a branch, causing an impedance jump and thus also causing an echo (at the branch).

So neither one is perfect.

Klaus

 

[D.A.(Tony)Stewart]

The USB 2.0 specs requires a differential impedance of 90 Ohms and a common mode impedance of 30 Ohms +/-%.
This, by definition, gives a target of an odd mode impedance of 45 Ohms and an even mode of 60 Ohms.

The USB should be terminated locally and the signal will be >20 MHz so HF test & measurement techniques are critical.

You ought to use a high speed Differential FET Probe ( 1Gohm || 1 pF) but these Rhode & Schwarz or TEK probes are expensive ($$$).

The next best budget method is to use two 10:1 probes rated for 200MHz+ calibrated with spring probe adapters to pin sockets near signal source.
You can even twist the two probe cables for better HF CMRR. These common springs eliminate ground clip wire and expected coax resonance and pulse ringing from ground clip ESL+ coax C. It uses the exposed probe pin (with clip removed) with a ground spring coil tip around the barrel of the probe.

If designing your PCB, it is useful for DFT reasons to add PTH holes for this test method for differential probes. If not, use the nearest pads to the signals.

Calibrate each probe with a perfect square wave, then calibrate both probes on the same signal using DSO Ch. A-B or 1-2 to get a "flat line" up to max frequency. Now you have a decent Diff. Probe setup. using 2 probes. This is Ok for the slower speeds of USB2.

A common problem with USB is when one end uses a floating SMPS portable to another DC powered device with PE grounding. Then the SMPS noise becomes a large interference to USB signal integrity and challenges the CMRR of signal from passive unbalanced wires and traces. (Often connecting the mobile SMPS to PE gnd. shunts this interference and prevents connection errors.) The floating SMPS supplies have feedthru C in the transformers which inject a large CM noise signal. Sometimes the CM chokes used are ineffective depending on the spectrum.

 

[FvM]

1. The correct oscilloscope input impedance setting depends on the used probe. You don't tell which probe you want to use.
2. You'll only see meaningful signals on USB in operation. The bus is terminated by the connected devices and must not be loaded by the probe. Respectively you need a high impedance probe. Standard 10:1 passive probe may be used with fullspeed (12 MBPS) but not highspeed ( 480 MBPS) USB.

 

[danadakk]

Probing :

Probing Tips for High Performance Design and Measurement

FIGURE 1. Probe loading is dependent upon the input impedance of the probe and the sensitivity of the circuit to this impedance. When

www.tek.com

ABCs of Probes

Learn about the fundamentals of probes and discover how to choose the right probe for your oscilloscope and get reliable measurements every time.

www.tek.com

https://www.eeweb.com/wp-content/uploads/articles-app-notes-files-oscilloscope-probing-for-high-speed-signals-1348764017.pdf

High speed diff probes pretty pricey. Google "DIY Oscilloscope Probe"


Regards, Dana.

 

[D.A.(Tony)Stewart]

What you need to verify depends on the method of signal capture chosen to have reasonable accuracy. Can you define it?

 

[dick_freebird]

Some 'scopes offer ability to bond two channels and
get a differential -display-. Quality subject to the two
signal chains' identicalness of course.

A balun could be an answer if at a frequency range that
makes this suitable (maybe cap-block it to keep from
shorting the lines at DC, and throw a balanced pattern)?.
Maybe separate DC / LF measurement tasks from "eye"
type stuff, if too many compromises for one-bench-setup-
fits-all.