CAN Bus Errors, Before & After

[JamesClarkIV]

Here is a look at the CAN bus error stats from a 14 LRU can bus that was originally wired with 96 feed to aircraft grade 22 AWG twisted pair.

It was rewired with 56 feet of Garmin recommended expensive Gigaflight 120 ohm twisted pair. Here is the difference in the data:

Flights 1-5 were old bus / long wire. Flights 6-10 are new bus / shorter quality wire.




Related writeups:

N997CZ — CAN Bus Rewire: Fixing the Basics Before Chasing Vibration

After the ADAHRS vibration analysis across the first five flights, Garmin’s guidance was unambiguous: fix the CAN bus first, then work the vibration problem. The data showed sustained % devia…

jamessclarkiv.com

And the subsequent .pdf below:

 

[msires]

Yep, the right wire may be expensive at $6+ a foot, but you don't need that many feet. You'll realize it was worth it and you should have used it after you spend days under the panel trying to find an intermittent problem on multiple avionics.

 

[Walt]

96' old vs 54' new had a lot to do with the results.

 

[bjdecker]

Walt's right -- 96' is getting really close to the single node limit for a 120ohm twisted pair running at ~1MHz (40 meters, subtract 1M for every node). Since you changed two variables -- hard to know what really "fixed" the errors, but I'd say it was the bus length more than the wire type.

 

[emes]

Related writeups:

N997CZ — CAN Bus Rewire: Fixing the Basics Before Chasing Vibration

After the ADAHRS vibration analysis across the first five flights, Garmin’s guidance was unambiguous: fix the CAN bus first, then work the vibration problem. The data showed sustained % devia…

jamessclarkiv.com

And the subsequent .pdf below:

This inspired me to chase down a lingering problem segment, so thanks!