Saabscene Saab Forum - Saab Technical Information Resource banner

Temperature gauge - odd behaviour

15K views 21 replies 7 participants last post by  rjsdavis 
#1 ·
2005 Saab 9-3 1.9 TiD Vector Sport (6s manual)


I have a similar problem to this. Prior to yesterday, for the previous day or two, I had noticed that my petrol gauge was not as it should be, as the gauge was sitting roughly 1/4 of a tank higher than it should be for the amount of fuel that was in the tank. Usually, the yellow warning light comes on when the gauge needle just about touches the first of the three reserve tank lines. However, now, my petrol gauge yellow warning light comes on when the needle is pointing at slightly below 1/4 full! I decided to trust the remaining mileage numbers on the SID display instead of trusting the petrol gauge itself. Then....


Yesterday, I found myself in London in reasonably heavy traffic. I noticed that the temp gauge was moving up past the 3/4 mark, so was keeping a close and concerned eye on it. A few minutes later, it was level with the first of the three red bars! I started to panic, as I assumed that the gauge was accurate and my car was on the edge of overheating! I pulled over to have a look - the engine was certainly hot, but it didn't seem as though it was ready to pop. I should add, that at no point has the yellow "Check engine" light come on, nor have I had any other warnings from the car/dash at all, that the car was about to overheat. I am getting an occasional "check oil" warning on SID, but when the oil level is checked, it is also pretty much exactly between the min and max lines on the dipstick. The oil/filter is due for service/replacement, but is this relevant? (I had assumed not for this particular temp gauge issue).


When I restarted the engine, the gauge was even higher (top of the three red bars), so the engine had theoretically got hotter whilst it was off and I was checking it. I couldn't work out what was going on, so whenever I got caught in traffic and was going to be stationary for a while, I stopped the engine to help her not to overheat. This seemed to make the gauge worse, as I noticed that when the engine was off, the temp gauge would drop to the half-way point on the dash gauge. By the end of the day, and indeed now, the day after, the temp gauge now rests at the first of the three red bars as it's "reset" or "resting" point position. Prior to yesterday, my temp gauge would always sit, rock-solidly just above the 1/4 mark (on a cold day), and just under the half-way mark (on a hot sunny day) when the car was running at working temperature.


I've read a few posts on the net, which talk about various fixes for this. As a result of following the tips - I've:


* Removed fuse F4 from the passenger glove box fuse box. This is supposed to cut the feed to the dash instrument cluster. It's been removed for a couple of hours now, and it's made no difference, as the needle still sits, at rest, at the first of the three red bars. The fuel gauge is also still wrong too.


* I've also removed F1 and F4 fuses from the engine bay fuse box, as these should cut the feed to the ECU. This also made no difference.


* I've also completely disconnected the battery to effectively "reset" the car. The battery has also been disconnected for a couple of hours, and this has also made no difference either! (I've left it disconnected for the time being in case there is any residual charge anywhere in the system that is propping up the dash gauge falsely.




I was using the car earlier this morning, and checked whether pressing the nightpanel light made the needles drop. It did. The temp gauge dropped, but only back down to its new, stupidly-high resting place of the first of the three red bars, and the fuel gauge also dropped, but only to its incorrect new resting position of the third of the red reserve tank bars.


Therefore, from, what I've done, I feel confident that the fault doesn't lie with either the engine temp sensor or the coolant temp sensor? Given that the engine was bone cold from resting overnight, and has now also had all power removed from it, if the sensors were sending a duff signal - the removal of power should still have reset the temp and fuel needles to their normal resting points? That's my theory anyway.


1 - I've read on another forum that I'll need a Tech2 diagnosis to read the codes. I'm also a Land Rover owner, and there's a large swathe of owners that have Nanocom computers that can be plugged in and do pretty much all of the jobs that a main stealer computer can do. Is there a Saab equivalent?


2 - I've also read on another thread, that apparently gauge needles can "slip" on their spindles. Is this correct? Can the instrument cluster be removed and these needles to "moved" or "reset" back to where they should be so that point in the correct range in the future?


3 - Is there anything I should be checking to ensure that I've diagnosed this correctly?


Many thanks for any help or diagnosis to help sort out this super-niggly problem!
 
See less See more
#2 ·
Hello rjsdavis , from what you are experiencing it sound to me as though there is a fault in the dash panel . I've had the same fault with my tempreature gauge , it normally sits at the half way mark , but when it starts to go wobbly it will clime above the half way mark and nearly reach the heady heights of boiling point , but I lift the bonnet and there's no signs of distress .
I turn the car off and the needle sits a quater way up the gauge , so it's always reading the atchual tempreature + a quater more ( if you catch my drift ) .
I just disconnect both terminals on the battery leave for 10 mins and reconnect and all is well .
But you have already done that ,
There is some other guy on here where his Rev counter goes off the scale , and he dose the same as me ,
Seems like your dash panel has gone completely wobbly .
It's just funny that yours is a vector sport , so is his and so is mine and mines a 2005 , so is yours and so is his , ........"......
I think that all tech 2 will be able to do is confirm the diagnosis , which would be a replacement dash panel .
There is OP-com which is a gm based diagnostic system ( most saabs are GM based , look on that well known auction site ,.
Sorry I can't be of much usefulness
But you are not alone with this kind of problem .
 
#4 ·
2005 Saab 9-3 1.9 TiD Vector Sport (6s manual)


I have a similar problem to this. Prior to yesterday, for the previous day or two, I had noticed that my petrol gauge was not as it should be, as the gauge was sitting roughly 1/4 of a tank higher than it should be for the amount of fuel that was in the tank. Usually, the yellow warning light comes on when the gauge needle just about touches the first of the three reserve tank lines. However, now, my petrol gauge yellow warning light comes on when the needle is pointing at slightly below 1/4 full! I decided to trust the remaining mileage numbers on the SID display instead of trusting the petrol gauge itself. Then....


Yesterday, I found myself in London in reasonably heavy traffic. I noticed that the temp gauge was moving up past the 3/4 mark, so was keeping a close and concerned eye on it. A few minutes later, it was level with the first of the three red bars! I started to panic, as I assumed that the gauge was accurate and my car was on the edge of overheating! I pulled over to have a look - the engine was certainly hot, but it didn't seem as though it was ready to pop. I should add, that at no point has the yellow "Check engine" light come on, nor have I had any other warnings from the car/dash at all, that the car was about to overheat. I am getting an occasional "check oil" warning on SID, but when the oil level is checked, it is also pretty much exactly between the min and max lines on the dipstick. The oil/filter is due for service/replacement, but is this relevant? (I had assumed not for this particular temp gauge issue).


When I restarted the engine, the gauge was even higher (top of the three red bars), so the engine had theoretically got hotter whilst it was off and I was checking it. I couldn't work out what was going on, so whenever I got caught in traffic and was going to be stationary for a while, I stopped the engine to help her not to overheat. This seemed to make the gauge worse, as I noticed that when the engine was off, the temp gauge would drop to the half-way point on the dash gauge. By the end of the day, and indeed now, the day after, the temp gauge now rests at the first of the three red bars as it's "reset" or "resting" point position. Prior to yesterday, my temp gauge would always sit, rock-solidly just above the 1/4 mark (on a cold day), and just under the half-way mark (on a hot sunny day) when the car was running at working temperature.


I've read a few posts on the net, which talk about various fixes for this. As a result of following the tips - I've:


* Removed fuse F4 from the passenger glove box fuse box. This is supposed to cut the feed to the dash instrument cluster. It's been removed for a couple of hours now, and it's made no difference, as the needle still sits, at rest, at the first of the three red bars. The fuel gauge is also still wrong too.


* I've also removed F1 and F4 fuses from the engine bay fuse box, as these should cut the feed to the ECU. This also made no difference.


* I've also completely disconnected the battery to effectively "reset" the car. The battery has also been disconnected for a couple of hours, and this has also made no difference either! (I've left it disconnected for the time being in case there is any residual charge anywhere in the system that is propping up the dash gauge falsely.




I was using the car earlier this morning, and checked whether pressing the nightpanel light made the needles drop. It did. The temp gauge dropped, but only back down to its new, stupidly-high resting place of the first of the three red bars, and the fuel gauge also dropped, but only to its incorrect new resting position of the third of the red reserve tank bars.


Therefore, from, what I've done, I feel confident that the fault doesn't lie with either the engine temp sensor or the coolant temp sensor? Given that the engine was bone cold from resting overnight, and has now also had all power removed from it, if the sensors were sending a duff signal - the removal of power should still have reset the temp and fuel needles to their normal resting points? That's my theory anyway.


1 - I've read on another forum that I'll need a Tech2 diagnosis to read the codes. I'm also a Land Rover owner, and there's a large swathe of owners that have Nanocom computers that can be plugged in and do pretty much all of the jobs that a main stealer computer can do. Is there a Saab equivalent?


2 - I've also read on another thread, that apparently gauge needles can "slip" on their spindles. Is this correct? Can the instrument cluster be removed and these needles to "moved" or "reset" back to where they should be so that point in the correct range in the future?


3 - Is there anything I should be checking to ensure that I've diagnosed this correctly?


Many thanks for any help or diagnosis to help sort out this super-niggly problem!
UPDATE - RESOLVED:


This was fixed the same day, as you can see from the above that I had had removed fuses and disconnected the battery at the time of posting originally.


* 4 hours later, I reconnected the battery - no change.
* I then reinserted both of the engine bay ECU fuses and checked the dash - no change.
* Finally, I reinserted the F4 fuse behind the glovebox. There was a series of fast "clicking" that went on somewhere behind the dash. After the clicking stopped, and as I watched the dashboard, both engine temp and fuel gauge needles dropped from their false/incorrect positions, back down to where they should have been all along.


It's been about a week now, and so far, touch wood, the needles have remained in their correct positions, and my engine temp now reads normally at hovering around the 1/4 mark, and my petrol gauge now correctly matches the SID (miles to empty) data. Result.


I had thought that a full car reset had also cured my intermittent alarm problem too... until it went off today 15 seconds after arming the car, and having reinstalled the boot fuse for the alarm system! (Arse).
 
#5 ·
Hello rjsdavis thankyou for the update , that was a bit of mission re-setting it , but it worked .
As I said with mine it's just battery terminals off , 10 mins , back on , reset everthing and away I go .
Once again thankyou for the update .
 
#8 ·
Bloody hell, what a palava. You have to love (and sometimes hate) the quirks that are Saab! Try reading my sai delete and remap thread sometime and how one thing can affect another in a way that makes no sense and applying standard car mechanic faultfinding to a saab can be a pointless waste of time and frustrating as hell!

Having said that i really enjoy my car, its getting on now but i cannot think of anything (other marques) i really want to swap it for. Will keep it going for as long as possible.
 
#9 ·
I was thinking E class ...... I don't know whats available for sensible money tho' ......... Would love a E ...... Ideally perverible .......... Wonder how much i'd get if I traded in my left testicle ? ...... Mind you I am getting sick of spanners, in my old age ......... So I'm probably going to have to settle for a VW Up Or a Yaris ...... Poor me :(
 
#11 ·
Update:

Am finding that this remains an intermittent problem. Simply removing the F4 fuse from the behind the glovebox for an hour or so does the trick. I get the fast "clicking" of something resetting somewhere in the dash when the fuse goes back in, and the two needles drop back down to their starting positions.

My temp guage has always read just over the 1/4 mark when running normally (and getting nice cooling airflow from motorway speeds for example). It only gets to 1/2 way when it's properly warm and has been pissing around in town and in traffic queues getting hot and receiving very little natural cooling airflow.
 
#17 ·
Thanks Aerojon - but how can this be? I don't understand.

A car that is cruising up the motorway and receiving lots of cooling natural high speed airflow will always be running at a lower overall engine temperature than a car that it is ticking over and stuck stationary in gridlocked traffic receiving no natural high speed airflow at all. Therefore, the temp gauge on any car will never always be fixed at one single position (9 o'clock) "regardless of what driving you are doing" - I don't understand this at all.
 
#13 ·
At first, I would also assume that the thermostat is shot, but given the other details (eg Fuel gauge) it does sound like the Dash is at fault here.

Perhaps, the cheapest way forward would be to get a cheap generic ODB scanner and software. This will read any engine fault codes but also more importantly should be able to show basic senor outputs like Temperature.
If the ODB scanner says that your engine temperature is 90*c but you dash is anywhere other than the half-way (9o'clock) position then you can be fairly certain that it's your Dash. If the ODB & Dash agree however, then it's something else.

Given that Lizzard & Zohan have already stated that it's a common fault & you have fuel gauge issues too, I`m inclined to suspect the Dash is at fault here.


Frank.
 
#14 ·
When Thease dash panels go wobbly , you either get one , or two , or three needles doing all sorts of stuff . As I said someone's Rev counter went so far over it met the back end of itself .
But as rjsdavis has found out just resetting it his way sorts out the problem .
I have a tempreature gauge that sometimes reads a quater when engine is off , so it shows I'm nearly overheating , but the o,d battery disvonnect sorts it .
As Aerojon has said in 2 posts
Your tempreature gauge should be at 9.0 o'clock and should get there in abouts 15-20 mins
If it's quater way up the scale then yet again as aerojon has stated your car is being over cooled .
With Thease cars when it goes like yours then rep,ace the thermostat and housing ( they come as a comp,eye kit ) and are available on that well known auction site .
 
#15 ·
But pulling the battery should only correct the crazy dash, it won't affect the thermostat surely?
Or are you thinking that his car has 2 issues? To be fair the thermostat is also a fairly common fault too, but he has also reported funny things happening with the fuel gauge too.
If the dash is suspect then we can't trust the temp gauge, hence my suggestion for an OBD reader. My thinking being that we need a second opinion about the actual engine temperature.

Frank.
 
#16 ·
Thanks for these posts. Very much appreciated.

CaptainFrank - as indicated earlier, when the gauges go funny (and end up being higher than they should be), it's usually both of them pissing around together - i.e. the temp gauge will run higher than normal, as will the petrol gauge, but not necessarily off by the same proportions - therefore, the temp gauge might be as much as a whole half of the scale wrong, but the petrol gauge might only be a fifth too high for example. They'll both also sit far higher at "rest" when the engine is off and key removed and this resting error state is proportionate to how much higher they are reporting when they are buggering around and running - i.e. if the temp gauge is a quarter of the scale out when running, it will also be a quarter of the scale too high at rest also.

I have always assumed that the levels that they display on the gauges is driven by some sort of relay or computer box that sits somewhere in the dash, and it is this very relay / computer box that is being reset by the F4 fuse removal and the fast "clicking" that occurs when the fuse is re-inserted. N'est pas?

It's interesting that you suggest that the temp gauge might also be playing around, and also that my engine might be being overcooled. I don't know if you have seen that I have a niggling issue with the a/c system (that is right beside this post in this forum section), where the a/c system shuts down after the engine is being put under load, which seems to shut down both engine fans. Curiously, this would seem to imply that if anything, my engine is NOT being overcooled, as far as I can tell, as the fans aren't actually working properly at the present time!!

The interesting thing about having the concurrent a/c issue, is that the fault codes are being read and checked by the guys trying to fix the a/c system, and they are telling me that there are no stored or current fault codes to read! This is why we're struggling so badly at the moment.

What do you make of that? I'm confused as hell...
 
#18 ·
Prolly due to a very efficient thermostat .
Maintaining a correct even engine tempreature , no matter what load .
My saab sits at 9.0 o'clock
My fiat 500 sits at halfway
My transit van sits at halfway
My vectra sits at 3/4 .
It's like a boiler thermostat . Maintaining tempreature .
So an efficient thermostat , will open and close frequently to maintain the desired tempreature ,
So no fluctuation in the gauge .
 
#19 ·
Yes, but this doesn't particularly allow for fluctuations in the ambient local temperature either. For example, in the UK, we generally have our engines running in ambient temperatures that range from, let's say -5 degrees, upto +35 degrees - a 40 degree range.

Clearly an engine running at motorway speeds runs at broadly the same operating temp, give or take a degree or two. However, running at the same operating temperature when it's mid-Winter at, let's say 0 degrees, and running at the same broad operating temperature when it's mid-Summer and cooking at 30 degrees, the engine is still running in the same way as it's still doing the same work. In fact, there is an even greater difference, as the engine parts themselves are that much cooler to begin with on a freezing cold day just from the cooling effect of the ambient temperature. The air that's being drawn into the engine intakes massively differs in temp too, as does the cooling effect of the ambient air as it strikes the engine block whilst cruising up a motorway.

The net effect of all of this is differences in running temps. It can never be "always the same". Running up a motorway in mid-winter, the engine will be running cooler, at the same speed and doing the same work, than doing the same journey in mid-Summer on a baking hot day. The engine management doesn't just "allow the engine to run 25 degrees hotter" on a freezing cold day so that the temp gauge stays level at of the time all year round. Remain confused....
 
#21 ·
rjsdavis - thanks for clearing up the wobbly display issue. I`m not as familiar with the 9440 models as I am with the 9600 and I`d never actually heard of this before but lizard-t and The Zohan had so I was happy to go along with them at the time. Does anyone know the actual root cause of the needles behaviour? Is it as simple as the needles spinning on their spindles? I`m fairly sure that when working on my sisters 9-3 that when the power is restored the dash quickly moves all needles up to full then back to 0 again. If the needles are spinning with might explain why this reset them back to where they should be sitting (this is pure conjecture on my part as I am not familiar this this system at all!)

On the subject of the thermostat though, my sisters 9-3 1.9 TiD had a faulty one that caused it to never reach the halfway mark on the temp gauge. As soon as a new one was fitted it would get upto halfway within about 10 minutes.

When I replaced the thermostat on my 2.2 TiD, I noticed that it was designed to open at 92*C - I imagine that the 1.9 TiD is probably the same.

I would also like to point out that the 9-5 (and I assume the 9-3) has a deadspot on the temp display where anything between roughly 85-95*C (I don't recall the exact temps but they can be changed with a Tech2) will show as exactly halfway.


Frank.
 
#22 ·
Does anyone know the actual root cause of the needles behaviour? Is it as simple as the needles spinning on their spindles?
I don't think it's "spinning on their spindles" - ie that is to say, that they become dislodged from their usual position. If this is so, a "gauge reset" or a reset of whatever component still feeding some level of voltage into them (one would imagine) to keep them from returning to their rest positions. One would imagine that the needles are designed to act as a resistor of sorts, with the more voltage supplied to them, the higher the needle travels. I.e. No voltage - needle is at rest (which is where it should be when the ignition is off), and varies right upto the far reach of its travel. I am guessing that when the needles play about, an incorrect (excessive) level of charge is fed into this particular dash clock - hence them only ever being higher than they should be. It would appear that cutting out the F4 fuse, resets whatever actually controls the level of voltage that feeds this particularly clock. It's definitely not the needles "spinning" - i.e. like when you manually dislodge them by touching them by hand inadvertently for example.


On the subject of the thermostat though, my sisters 9-3 1.9 TiD had a faulty one that caused it to never reach the halfway mark on the temp gauge. As soon as a new one was fitted it would get upto halfway within about 10 minutes.
Ok, I accept that there's probably an issue with the thermostat on my car. I'll start sourcing one to replace it and see what happens. As I say, this has always been like this during the lifetime that I've owned the car (6/7 years now). So, sadly, this doesn't appear to be connected to concurrent a/c issue.... Any ideas on this super-irritating problem?
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top