TCAN1043HG-Q1: Calculation cable length and stub limitation for CAN FD at 2Mbps

Hi Alex,

Thank you for your question. I am unfamiliar with the source of the above snapshot and formula. Could you let me know where it comes from?
We also recommend a 5ns/m cable length propagation delay as a good rule of thumb. Just looking at the formula, it looks like you can produce a plausible estimate if you scale the formula to CAN FD data rates at 2Mbps. Looking at the above example for 500Kbps, it looks like the source has taken a time quanta approach to calculate the propagation delay segment for each CAN data bit. (I'm assuming each time quanta for 500Kbps is 125ns, and the formula assumes 11 time quanta of propagation delay @ 500 Kpbs.) Since the total bit time is 2us for each bit, then the formula assumes 1375ns/2000ns = 68.75% of each bit is propagation delay.

If you take that ~70% ratio and scale it to 2Mbps data rate, then you come out with a propagation delay, t_propseg:

t_propseg = 70% * 500ns/bit = 350ns.

Plugging that into the formula gives:

L_D = 1.4m
L_DI = 7m


This estimate seems plausible given what we have measured for our TCAN1043 technology at CAN FD speeds. We have been able to run at 2Mbps and 5Mbps without bit errors on our own multi-node network with a single stub length exceeding 1.5m and a total stub length exceeding 5m and a trunk length of 40m, just to give you an example.
I think this formula should give you a decent estimate on where to start. Your unique bus network topology, including cable type, bus capacitance, and other factors will contribute to the discussion as well. Do you have a particular design goal you are targeting for your application? If you want, I can provide more detail about the example network I have mentioned above to highlight the CAN FD performance of the TCAN1043. Please email me me separately if you are interested in that info.

Let me know if I can clarify any of the above calculations.

Best Regards,
Max Megee
TI Transceiver Applications
tmmegee@ti.com