Instrument Panel Problem (with video!!!)

Does everything come back on when sitting still or while driving? I suspect a loose connection somewhere. These can be a bugger to diagnose. Probably the easiest way for you or your mechanic is, when the problem returns, start wiggling every connector while checking the display. You'll know when you find it.

 

Intermittent problems can be really tough to fix. Unfortunately your car has become an example of how a high-mileage Prius may be difficult to keep running, due to an electrical problem.

When you say you checked every possible connector, does this include removing the instrument panel plastic trim (which includes the passenger-side airbag) so that you can access the combination meter (speedometer, odometer, fuel gauge, warning light assembly)? I suggest that if this has not been done yet, your mechanic will need to gain access to it. Then remove the meter and inspect the wiring connector and the mating socket. Clean if necessary with the correct spray solvent for electronic switch parts, then reassemble. Maybe that will help, maybe not.

The combination meter receives 12V power from three fuses: 10A GAUGE, 15A DOME, and 7.5A IGN. Check those fuses again to see if they are secure in their sockets and if there's any problem with the wiring. When you experience the problem, use a digital voltmeter to measure the voltage on both sides of those three fuses: if you don't measure voltage on one side of a fuse, that is evidence that the fuse is bad.

The backup lights are controlled by the combination meter, more evidence that part is the source of your problem.

Has the vehicle suffered any accident damage that would be relevant to this problem?

It might also help to disconnect the taxi meter while working on the problem, to make sure that device is not a contributor to the problem.

Good luck.

 

Luscious Garage | Hybrid Specialists | LG TV - Prius Dead Dashboard

 

Hi Carolyn,

Thanks for providing the link to your informative article. One question and one comment:

1. Any sign of coffee or other liquid stains on the failed combination meters that were removed?

2. Thanks for pointing out that the failure of the car to power down is due to confusion about whether it is at speed or not, and that holding down the POWER button is a way to force the power down. Otherwise, I was thinking it would be necessary to disconnect the negative terminal of the 12V battery, then remove the traction battery's orange service interlock.

 

Hi Patrick,

1. The first cab had some powdery substance on the back of the control board and we thought "a ha!" The next one was visually perfect. Go figure.

2. Again at our first brush with this issue, it took a while to get the car to act up consistently. When it did come in BROKE the power button also didn't work. First step was hooking up the scan tool and looking for CAN confirmation of power button inputs (or lack thereof). Datastream showed button inputs just fine (it also showed open requests at the hatch button). So why denying it? I was already back probing powers and grounds at the cluster, and when I finally disconnected it entirely the button worked normally. Then the question was, "what does the cluster offer to override the power button inputs?" The assumption was speed.

This is not uncommon with contemporary (i.e. computer controlled) cars, actually. A common case is when the speedo doesn't work and you hook up the scan tool and look to see if you can see vehicle speed at the engine ECU. If speed shows then you suspect an internal fault with the cluster.

When the car acted normally with the combo meter disconnected, obviously the cluster needed to go. It was pulling down the communication bus, affecting the power inputs and the hatch, etc. I spent some time studying the wiring diagram, which ultimately illuminated the BEAN gateway aspect of the speed information to the Power ECU and Body ECU. The combo meter gets speed info from the Skid ECU (wheel speed sensors) over the CAN bus, which means that everything else on the CAN bus gets that info too. The Power ECU is on the CAN bus too (Body is probably also, but I don't remember and the diagram isn't in front of me), but it clearly listens to speed from the BEAN network, not CAN. Interesting car to be sure.

Very best,
c

 

Thanks to Carolyn for providing the theoretical explanation for why a combination meter failure can cause the problem; and thanks to toi for confirming this was the problem behind the OP.

 

Wow! Great article! This ought to go in a FAQ or KB article somewhere.

Too bad LG is so far away from me. :/ (if I were to need some out of warranty work)

 

It is fantastic that you all have come up with a fix for this problem but, has anyone worked out what is causing the failure that it might be prevented?

I see that all known failures have been in cabs and some of those don't have extraordinarily high mileage. Could it be that the taxi meter is causing the failure rather than the stereotypical behaviours of taxi drivers?

If it is the taxi meter is there anything that can be done to protect the instrument cluster from harm caused by the taxi meter?

I am in awe of the knowledge of the people on this forum.

 

Hi Pat,

This is not true. Two recent threads contain info from several owners who do not appear to be cab drivers, have logged relatively low miles on recent model year Prius (2007, 2008), and have experienced similar symptoms:
http://priuschat.com/forums/prius-main-forum/56597-my-prius-will-not-turn-off.html

http://priuschat.com/forums/prius-main-forum/55865-cant-turn-prius-off.html

San Jose to San Francisco is ~50 miles, not too far to travel to find real Prius expertise. It is likely that few Toyota dealerships have similar caliber techs working on Prius issues.

 

I missed the start of all this, but patsparks has been poking me
to look into this further. [Appreciated, don't get me wrong!]
Okay, I caught up, and as a sidenote it's awesome that
lusciousgarage has joined all the fun over here!
.
I would venture a guess that this is a very similar problem to the
MFD. The reported problems and timing of non-functionality really
feels like a loose connection. It's a big block of electronics. It
gets bounced all over the place when you go over bumps in the car,
too, and after watching it whang up and down over some of the sand
roads last summer I'm amazed it hangs together at all. Diagnosing
a failed vs. good unit would probably be a big more, uh, interesting
than doing the MFD since you'd have to have the whole dash-top off
and possibly even drive around that way.
.
However, if a failed unit is still in someone's hands out there it
might be worth in-depth study if this is going to become a common
pattern failure. [Hmm, I wonder if I happened to get an "improved"
model of some sort in my '04, like with the MFD?] Except that
they're dying in much later model years, possibly indicating some
sort of design *downgrade* may have happened along the way. [Maybe
they took that bad connector out of the MFD and moved it to the
instrument binnacle, heh]
.
Are the clusters substantially different other than layout between
'06+ and earlier years? I haven't looked into the electrical side
of that. It *is* weird how many things get vectored through the
cluster; you'd never think that *reverse lights* control comes from
there but on the upside it made that control wire easy to get to.
.
_H*

 

Or alternatively, some sort of manufacturing change (e.g. new process, new operators, new source of parts). Speaking as a former Process Engineer, the possibilities are endless.

 

Sorry Patrick, I missed those threads. I was wondering if the issue was to do with drivers who are less careful because it isn't their car.

 

I have a 2008 Prius that has 6000 miles on it. At start up this am in my garage, I had a blank instrument cluster - nothing. It took me awhile to figure it was on. I drove it and the instrument cluster came back on after 20 mins and 15 miles. Although the cluster is back there is no light on the start button (green) and the display seems to still be acting strange. Taking it to the dealer Tuesday.

 

I now have a solution for this problem. See [this] post.