VOXL2 min rise/fall time for 5MHz SPI interface
-
Hello. I'm designing a rigid flex cable and the ground plane hatch spacing seems to be dependent on the min rise/fall time of the signals. Does anyone know what that might be on a 5MHz VOXL2 SPI bus?
-
Hello. I'm designing a rigid flex cable and the ground plane hatch spacing seems to be dependent on the min rise/fall time of the signals. Does anyone know what that might be on a 5MHz VOXL2 SPI bus?
Hi @michaelherrera
Short answer, I do not have a solid answer for you.Long Answer:
The rise and fall time will be dependent on the total capacitance of your cables (length, trace width, etc), your load, and the QRB5165 has many different drive strength options too.That being said, SPI is not rigidly required to have 50-ohms if you are going for that, you may be over-constraining yourself. Especially at 5MHz, impedance matching in my experience has never been an issue at that slow of a frequency. The important thing is more to not have stubs greater than 1/4 wavelength of the rise-time.
For any reflection/matching concerns, you have over 8nS of HIGH and LOW time at the fastest frequency of 50MHz for SPI, so at 1/10th that, you have over 80ns of settling time. if you are worried about emissions, then a simple AC tuning circuit can be placed at the load side to help match and squelch any ripple/undershoot/overshoot due to reflections.If you can better explain what port/connector you are referring to, what SW you are running, and what target device you are reading, I can try to capture an image on the scope to measure some rise times. But, that being said, you are probably on the order of 1-3 nanoseconds which is common on SPI. Neither of which may impact your hatching too much. If you enter anywhere from 1ns to 2ns on your calculator, I would bet you'll get good enough and not see much impact/change between the two values. Then, you can test on your own with your HW and see if you are close. I bet you would be. 1-2ns also translates to ~1GHz max, which is a 1/4 lambda of ~12inches. So, avoid stubs or "T's" longer than 6" (I target 1/2 the 1/4 wavelength rule).
If you are working on GHz links, then yes, rise times are on the order or 10's of pico-seconds and you'd see impacts there, where every 10ps will move the needle.
Hope this helps.
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register Login