0
\$\begingroup\$

When I connect my oscilloscope probes to my SPI lines, the line voltages get affected a lot and so my SPI comms fail.

I am using a breadboard as an intermediary between the two devices and adding cables to the breadboard to scope from.

Is there a better way to probe SPI? I'm currently looking to snip my wires to expose the metal and probe that way to reduce the length of the SPI connections.

Any better ideas would be appreciated.

Running SPI at 10 MHz

\$\endgroup\$
5
  • 3
    \$\begingroup\$ If putting a scope to SPI bus causes such serious issues, it is an indication that there is something horribly wrong with your system. Perhaps you need to improve the system instead of improving the measuring? Post all the details to figure out what is wrong. Usually, putting a scope to SPI bus has no adverse effects at all. \$\endgroup\$
    – Justme
    Mar 29, 2021 at 9:56
  • 1
    \$\begingroup\$ What speed? I do connect scope probes directly without issue for 10MHz. \$\endgroup\$ Mar 29, 2021 at 9:56
  • \$\begingroup\$ What cable length overall (including the breadboard)? \$\endgroup\$
    – Andy aka
    Mar 29, 2021 at 10:09
  • \$\begingroup\$ the line voltages get affected a lot - how do you know that and, what does "a lot" actually mean? Are you using bidirectional SPI all the time i.e. you judge an SPI fail based on not receiving a proper slave response? \$\endgroup\$
    – Andy aka
    Mar 29, 2021 at 11:01
  • \$\begingroup\$ @Andyaka length of wires is roughly 14 cm including the breadboard. I mean on the oscilloscope I can see the clock line doesn't get pulled to ground when it is switching. \$\endgroup\$
    – JC123
    Mar 29, 2021 at 11:14

3 Answers 3

3
\$\begingroup\$

Your problem isn't so much the scope probes as the breadboard.

The breadboard adds series inductance (as though you had installed coils in series with the SPI signals) and parallel capacitance (as though you had put small capacitors from the SPI signals to ground.) This will slow the rising and falling edges, making it more difficult to detect level changes. The inductance can also cause ringing - there can be false level transitions from the ringing.

Attach your probes directly to the SPI signals, right at the point where the wires connect to one of your boards.

Use 10X probes if you have them - they are less of a load on the signal lines.


SPI uses relatively fast signals, and isn't intended to be used between circuit boards.

It is intended for use on a single board with short connections.

The wires between the boards are probably making your SPI signals "unhappy," never mind the breadboard and the additional load of the oscilloscope.

  1. Use short wires.
  2. Keep the circuit neat.
  3. Use straight wires rather than letting things twist around and get tangled.

Put the breadboard down and back away slowly. Nobody has to be injured here.

\$\endgroup\$
1
  • \$\begingroup\$ LOL: I admit, I also had some non-terminated probe wires soldered to my SPI-signals and measured significant noise. I unsoldered them and backed away slowly. Nobody was injured. \$\endgroup\$
    – boink
    Aug 31, 2022 at 15:08
1
\$\begingroup\$

The thing to look out for is the parasitic capacitance you create on the communication lines.

  • I would suggest to keep the lines short and maybe try to bypass the breadboard.
  • Does it work with the breadboard without the scope connected ?
  • Do you see some signal on the scope ?

Also how is your oscilloscope grounded ? Make sure you clip the ground probe somewhere. If your device under test is floating, you can't rely on the earth reference.

\$\endgroup\$
1
  • \$\begingroup\$ It works with shorter wires. My question is basically that I need to attach the scope probes and I'm going to just snip the short SPI wires to expose the conductor then probe it but doesn't seem like the best method to me as cables get exposed. \$\endgroup\$
    – JC123
    Mar 29, 2021 at 11:16
0
\$\begingroup\$

While other answers are fairly correct, in my experience the big problem with breadboards, especially the cheap ones, is that as soon as you use wires of different sizes, it spreads apart the breadboard spring plates and causes bad contacting.

My guess is when you insert your scope wire into the breadboard, this is exactly what happens. Try to probe a different way, use a higher quality breadboard, or solder your circuit on a test PCB (they are cheap these days).

Personally, I mostly stopped working with breadboard altogether because of these issues, and as the cost of PCBs has become so low that you can just make a PCB for 5$, and there are fewer and fewer DIP components.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.