Can Bus Arithmetic

Lost my Can Bus on a coupled RNAV(gps) approach, thankfully not IMC.

Found a loose/broken wire at the AOA 260 molex. And reading the other forum posts, I realize I don't have the alternate path wired from the GSU25 to the GDU460 that is RS-232. I will do this ASAP before venturing out in IMC.

Here's my arithmetic question: my meter showed 120 ohms from Can Bus Hi to Can Bus Lo. Just like it's supposed to. BUT I also found Can Bus Hi to ground pin at 30 ohms. And Can Bus Lo to ground pin at 30 ohms. And the last two should be open.

So of course I'll start pulling units off the can bus until I can hopefully find the culprit. But I am puzzled, since I've had 400 hours on this panel....and the measurement of 30 ohms on both Can Bus Hi and Can Bus Lo seems to be more than coincidental.

Any observations that might make my troubleshooting quick and/or unecessary?

Molex

I would not be using molex connectors for can bus, or even rs232. I would rather see cannon D style or duetsch connectors. Basically any connector that uses a proper pin and socket contacts.

I am afraid you are going to be fighting the molex connectors every annual.

With units connected to the can bus, i am not sure how reliable any measurement is from hi or lo to ground or power. Some units may have protection diodes and you could be measuring those.

flipmike said:

Lost my Can Bus on a coupled RNAV(gps) approach, thankfully not IMC.

Found a loose/broken wire at the AOA 260 molex. And reading the other forum posts, I realize I don't have the alternate path wired from the GSU25 to the GDU460 that is RS-232. I will do this ASAP before venturing out in IMC.

Here's my arithmetic question: my meter showed 120 ohms from Can Bus Hi to Can Bus Lo. Just like it's supposed to. BUT I also found Can Bus Hi to ground pin at 30 ohms. And Can Bus Lo to ground pin at 30 ohms. And the last two should be open.

So of course I'll start pulling units off the can bus until I can hopefully find the culprit. But I am puzzled, since I've had 400 hours on this panel....and the measurement of 30 ohms on both Can Bus Hi and Can Bus Lo seems to be more than coincidental.

Any observations that might make my troubleshooting quick and/or unecessary?

Click to expand...

From section
2.3.1.3.6 CAN Bus Troubleshooting

a) Using an ohm meter, verify that the resistance between the CAN HI and CAN LO pins on the connector is 60 ohms. This will verify that the CAN backbone is properly terminated at each end (two 120 ohm terminating resistors in parallel).

b) A resistance of 120 ohms indicates that one of the two required CAN terminations is missing

Not sure about the 30 ohms but that does not sound correct.
Molex for CAN bus, definitely not good....

Thanks

PilotJohn, you are reading my subconcious. Looking at the molex and the 22 gauge wire crimped to the pins, I'm less than thrilled at the safety factor for flying IMC.

And thanks Walt. I've read that same installation manual page...which is kind of why I was curious...60 is Hi to Lo, a missing termination at one end = 120

So "30" is too coincidental...1/2 of hi to lo. (?)

Oh well, binary search time...pull off a connector, see if it changes. Wash, rinse, repeat.

Measuring the resistance from CAN Hi to ground and CAN Lo to ground is an important step for troubleshooting the CAN bus wiring to rule out a short circuit, but just be aware that this won't give you a meaningful result unless you first unplug every device.

flipmike said:

Lost my Can Bus on a coupled RNAV(gps) approach, thankfully not IMC.

Found a loose/broken wire at the AOA 260 molex. And reading the other forum posts, I realize I don't have the alternate path wired from the GSU25 to the GDU460 that is RS-232. I will do this ASAP before venturing out in IMC.

Here's my arithmetic question: my meter showed 120 ohms from Can Bus Hi to Can Bus Lo. Just like it's supposed to. BUT I also found Can Bus Hi to ground pin at 30 ohms. And Can Bus Lo to ground pin at 30 ohms. And the last two should be open.

So of course I'll start pulling units off the can bus until I can hopefully find the culprit. But I am puzzled, since I've had 400 hours on this panel....and the measurement of 30 ohms on both Can Bus Hi and Can Bus Lo seems to be more than coincidental.

Any observations that might make my troubleshooting quick and/or unecessary?

Click to expand...

well it has not been mentioned yet in this thread, but regarding the test numbers by flipmick, the OP, the values described are actually impossible in theory.

If each of 2 points are 30 ohms to a common point, the MAXIMUM resistance between the 2 points (H&L)will be 60 ohm. It can't be 120 ohm. Sketch it out. So, dunno. Likely 30kohm, not 30ohm. Auto-ranging meter?

I just measured a powered off CAN bus with 4 units plugged in with a HF Digital VOM. The high to low rails were 60.1 ohms. Perfect. For the rails to ground, readings were off scale high on the highest ohm setting, so more than 200Kohms - open/floating. Also perfect.

However, when using a Simpson 260 meter which uses more current to make measurements, I got 5K and 36Kohms from rails to ground depending on polarity. Evidently, the higher measurement voltage/current of the Simpson meter will tickle the silicon connected to the bus into conduction causing a small error in the resistance readings of the rails to ground.

The good news for me is that one can do a meaningful check of the CAN bus without ripping into the system and disconnecting everything - though that may come later if trouble is found.

Again, 60 ohms or else. The 2 120 ohm resistors are absolutely essential to the operation of the CAN bus. They function not just as terminating elements but are essential to pull the signal back to the passive state. They are half of the signal drive.

Ron

Ahhh Cumulo...you are certainly correct! GIGO...garbage in and you get garbage out. My most sincere apologies.

Yes, Can Bus Hi to Lo for my system is 60 ohms. (well, 59.7). And of course basic addition for two 120 ohms in parallel = 60 ohms.

Repaired the broken wire, and in an abundance of caution or was it curiosity, I proceeded to then unplug everything on my Can bus. Nothing showed as shorted to ground, either Hi or Lo, and it was open between Hi and Lo.

Most importantly while I had the patient opened up, I put in the optional RS-232 connection between the PFD (GDU 460) and AHRS 1, and also put in the optional RS-232 connection between the MFD (GDU 465) and AHRS 2. Redundancy!

Oh...and vom values after it's all back together? 60 ohms across Hi and Lo. BUT for some reason, and I kept track as I plugged each node back into the bus, when I plugged in my com 2 (GTR 200), the last item to plug back in, the resistance went from open to apx 34 ohms for Can Bus Hi to ground, and from open to apx 34 ohms for Can Bus Lo to ground. I will further investigate that, but with the new RS-232 connections it will not be a safety of flight situation.

Thanks to all responding.

Yup

Glad to hear it is all better then new now.
I think the 34 ohms is reasonable. If you are curious, I would measure it again and change ohm reading scales. If the 34 ohms changes, than you are probably reading thru protection diodes or some similar circuitry. I would not worry about the 34 ohms; as long as it is not a dead short.

Lastly, GO FLY.