Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Stopgap Fix for Rear Camera Cable Recall?

This site may earn commission on affiliate links.
This article:


demonstrates the problem related to the recall of cars for the rear camera cable. If you go to your car, open the trunk, look at the right side, and close the trunk partway, you'll immediately understand the problem. The cable really kinks as the trunk is closed.

I got the recall email below. The cable hasn't failed, but I really need it to back into our garage.

I wonder if there's a way I could do something about this (duct tape) to keep the cable intact until "parts are available." Any ideas?

email.jpg
 
This article:


demonstrates the problem related to the recall of cars for the rear camera cable. If you go to your car, open the trunk, look at the right side, and close the trunk partway, you'll immediately understand the problem. The cable really kinks as the trunk is closed.

I got the recall email below. The cable hasn't failed, but I really need it to back into our garage.

I wonder if there's a way I could do something about this (duct tape) to keep the cable intact until "parts are available." Any ideas?
so they used non-flexible cable in a known flexi area?

every time I hear how 'good' tesla design team is, I think about huge basic things like this that they just MISS.

I mean, come on, single core wire in a moving application? even interns know better than that, don't they?

wow. I mean, WOW. extreme rookie mistake.

wonder what else they farked up when they hired newbies to design the various subsystems. damn!

this is what I would expect from a dollar item from ebay china.
 
so they used non-flexible cable in a known flexi area?

every time I hear how 'good' tesla design team is, I think about huge basic things like this that they just MISS.

I mean, come on, single core wire in a moving application? even interns know better than that, don't they?

wow. I mean, WOW. extreme rookie mistake.

wonder what else they farked up when they hired newbies to design the various subsystems. damn!

this is what I would expect from a dollar item from ebay china.

Well, the harness is made by Delphi who are the parts group spun out by GM. So either Tesla specified it this way or Delphi said this is the way we've made this stuff for GM for years don't worry about it. Either one seems equally likely...
 
Well, the harness is made by Delphi who are the parts group spun out by GM. So either Tesla specified it this way or Delphi said this is the way we've made this stuff for GM for years don't worry about it. Either one seems equally likely...
its not the supplier's fault.

no supplier swaps solid core for stranded.

that's 100% on tesla. 100 percent. they let this design go as -is and NOT A SINGLE PERSON THERE said 'hey, this is a flexi application, you guys think flexible cable might be appropriate?'

super rookie mistake, but no one in the whole company caught it!

my god. what else are they missing? seriously!
 
Two weeks ago all the cameras failed and I couldn't use AP. The problem fixed itself when the car slept.

Do you think that's related to the wire harness recall issue?
no; unless a connector is making/breaking connection. that's possible. tesla does not 'design' many systems - they slap them together, see it working once or twice in a lab and ship it. they are hackers, not engineers. that's how I now see them.

they could have screwed up thermal mating issues or condensation issues. when things warm up, connectors sometimes 'work better'. that's a bad design but it seems par for the course (sigh).

likely its software. putting things to sleep and then waking up is what causes the most bugs in car software (I know, seen them firsthand and debugged them). since battery life is so important, a lot of software is all about sleep states and how to aggressively put systems into lower compute or hiber states and wake them (in a tree fashion, via a trigger) when something changes. its cool and fun but its not easy and its why you see so many 'sleep/wake' bugs.

anyway, yours is likely software like most of the others. real cable issues would be hard faults that dont self-fix (other than intermittent connections, but those are at the connector level and rarely inside wire).
 
  • Like
Reactions: trisk and PianoAl
its not the supplier's fault.

no supplier swaps solid core for stranded.

that's 100% on tesla. 100 percent. they let this design go as -is and NOT A SINGLE PERSON THERE said 'hey, this is a flexi application, you guys think flexible cable might be appropriate?'

super rookie mistake, but no one in the whole company caught it!

my god. what else are they missing? seriously!

I'm not suggesting that Delphi swapped it. I'm suggesting that normally when you partner with a supplier somebody there would have told Tesla that such wiring isn't a good idea.

Tesla generally thinks it knows better than established industry players. That's partly how they've succeeded and partly why they run into these kinds of issues on basic problems. So it is quite possible Tesla ignored them if they did suggest it was going to be an issue.
 
so they used non-flexible cable in a known flexi area?

every time I hear how 'good' tesla design team is, I think about huge basic things like this that they just MISS.

I mean, come on, single core wire in a moving application? even interns know better than that, don't they?

wow. I mean, WOW. extreme rookie mistake.

wonder what else they farked up when they hired newbies to design the various subsystems. damn!

this is what I would expect from a dollar item from ebay china.
Blanket statements like this are dumb. At my job we use coax in moving applications all the time. They key is to maintain the correct minimum bend radius to prevent this problem, which Tesla did not.
 
  • Like
Reactions: smatthew
I'm not suggesting that Delphi swapped it. I'm suggesting that normally when you partner with a supplier somebody there would have told Tesla that such wiring isn't a good idea.

Tesla generally thinks it knows better than established industry players. That's partly how they've succeeded and partly why they run into these kinds of issues on basic problems. So it is quite possible Tesla ignored them if they did suggest it was going to be an issue.
its true that better vendors can 'help' the oem's make better choices, but I'm not sure I'd want to be the supplier to has to school Tesla on how wiring 101 works.

you might tell them about when chips might be EOL or when some new footprint appears and you have to re-layout the board on the next rev. but to tell the designer 'are you sure you really want this kind of wire?' - that's borderline comical.

would be interesting to be a fly on the wall, as they say, during some of those meetings. wonder how bad they really were? eyeballs always rolling, etc?
 
Blanket statements like this are dumb. At my job we use coax in moving applications all the time. They key is to maintain the correct minimum bend radius to prevent this problem, which Tesla did not.
uhm, no. that's not even close to enough.

you ONLY pick solid for stationary apps. period. full stop. never ever ever for flexi applications.

been doing this for decades, myself, mate. not a summer child, here.

just zero excuse for this. yes, you also do care about bend radius but you NEVER EVER pick solid core for flexible apps. that's just 101 level stuff.
 
Is RG6 coax in a CATV environment considered a flexible or fixed application?
you tell me - how often IS that going to be flexed? in walls and around the house (inside conduit), it sustains bends during install but not regular use, of course.

the real solution to flexible needs is flex pcb material. that orange film looking stuff with connectors on the ends for mating. I would have suggested something along that line for a high flex application.

I am also a fan of fine multistrand wire. when I build quads, I use silicone coated wires and super fine strands. expensive, but really love the stuff. downside is that the insulation is too easily pierced (you can strip it with fingernails and that's actually a recommended way to strip that kind of wire, seriously) but you then surround that with another flex layer of tubing. keeping all things multi strand inside or flex pcb material.

never ever single core. no matter what anyone says. that's just basics, right? this isn't about SI and trombones, man, its just coax ;)
 
uhm, no. that's not even close to enough.

you ONLY pick solid for stationary apps. period. full stop. never ever ever for flexi applications.

been doing this for decades, myself, mate. not a summer child, here.

just zero excuse for this. yes, you also do care about bend radius but you NEVER EVER pick solid core for flexible apps. that's just 101 level stuff.
Then you have been doing it wrong for decades. Plenty of cable manufacturers rate their coax for dynamic applications. For example, almost every single modern camera on a robotic arm. Or any body worn military comms antennas.

Here is an example of a datasheet for coax where the manufacturer has marketed the coax for these types of applications and has specified the minimum safe bend radius for repeated bending:

 
Last edited:
  • Like
Reactions: smatthew
you tell me - how often IS that going to be flexed? in walls and around the house (inside conduit), it sustains bends during install but not regular use, of course.

the real solution to flexible needs is flex pcb material. that orange film looking stuff with connectors on the ends for mating. I would have suggested something along that line for a high flex application.

I am also a fan of fine multistrand wire. when I build quads, I use silicone coated wires and super fine strands. expensive, but really love the stuff. downside is that the insulation is too easily pierced (you can strip it with fingernails and that's actually a recommended way to strip that kind of wire, seriously) but you then surround that with another flex layer of tubing. keeping all things multi strand inside or flex pcb material.

never ever single core. no matter what anyone says. that's just basics, right? this isn't about SI and trombones, man, its just coax ;)
Wow... I just went back and read your comment a little closer.

Silicone wire is great for quadcopters and other applications where soldered connections are ok and you don't need to worry about the service life. It is horrible for crimped connections which need to hold up over time. The insulation jacket is too soft and pulls out of the strain relief crimp on connectors.

That *sugar* is used for RC hobbies, not within the automotive industry. Furthermore, the cameras Tesla use are power and data over a single coax. They operate at ~70 Ghz. It is an application specifically tailored for coax /automotive. Look up GMSL2 cameras.

So your decades of experience is based only on RC toys and not on real engineering. Please don't inject yourself as an "expert".
 
oh, and for the one just added to my ignore list (so I wont see your reply, but others can see this note) - how does bend-radius HELP at all when you keep BENDING the wire?

bend radius does not address REPEATED bends.

not sure why this is hard for some of you to understand. most understand it, but those that insist on white knighting tesla for any flaw they make, I can totally live without seeing anymore of their nonsense posts.

keep bending solid core wire and it breaks. HENCE THE RECALL. uhm, DUH?