How do I measure a LVDS signal with single-ended oscilloscope probes?

[mariam.f]

Hello
I would like to measure a LVDS signal, on the board that is sending this signal. There are two signal pins on this board for this signal, signal-p and signal-n. How can I measure this signal? Can I use single ended probes and an oscilliscope? I tried to do that but I can't find a signal. Am I making a mistake in measuring this way (what if so?) I don't have differential probes and they are expensive, is there another way?

 

[DEvens]

Hello mariam.f. Welcome to the forum.

It might help to have some context. What is LVDS? What sort of voltage and current are you anticipating? What sort of frequency or rate of change in signal are you anticipating?

Probably what you want to do is Google up the maker of the sensor and find out what they suggest for getting the signal collected.

If you are talking about a "few" volts with frequencies up to a few mega-hertz you may get good results with an oscilloscope. You may need to put in some resistors to match impedance to get a good signal. The usual situation for a scope is a very high resistance across the leads. This may not be suitable for the kind of sensor you are using. Maybe what you need to do is put a resistor across the leads of the scope to balance things. If you have a high-end scope you may be able to coax it to collect and store the data for you.

If you are talking about milli-volts the scope may, or may not, let you go that low. You might need an amplifier. You might need a power source going through the sensor. Before you start guessing, you almost certainly need that info from the maker of the sensor.

 

[Nidum]

https://www.fairchildsemi.com/application-notes/AN/AN-5017.pdf

 

[analogdesign]

Use two single ended probes with a 100 ohm resistor between them (one side of the resistor connected to each to probes). Attach one probe to signal-p and the other to signal-n. Then, in your oscilloscope, set the readout for the difference between the probes (may be called A-B or DIFF, it depends on different scopes). Then you should see your signal.

 

[meBigGuy]

Don't use the 100 ohm resistor if you are measuring with a terminated LVDS receiver already attached. But, otherwise analogdesign is correct.

Basically measure each signal with respect to ground and subtract them, either visually (which I do most often) or with "scope math".

 

[leright]

One possibility is to build a simple op-amp circuit that will take the difference between the two signals (p and n) but if your scope will take the difference for you then analogdesign's suggestion is the best.

 

[meBigGuy]

Again, no reason to subtract them in the scope or an opamp. I find it is actually more meaningful to look at the two signals single ended with the scope trace grounds overlayed. That way you see what each line is doing, and it is easy to subtract them visually.