Low Coolant Message - Resolved
#21
Sounds like the right plan to me.
Just to flesh out the thermostat idea a bit, what I'm thinking is ...
There is a small opening or valve in the thermostat body which is there to allow pressure to equalize even when thermostat is closed. If that opening were clogged or stuck then ... as engine cools down ... after thermostat closes ... we get more cooling and so could end up with a vacuum on the radiator side of the thermostat. Then on the next cold start, when the thermostat opens, coolant might want to flow from the tank toward the radiator, pulling the level down until further heating / expansion brings the level back up.
Or something like that anyway ...
Good luck.
#22
The newer cap didn't fix it. Last night I put the old radiator back on, temp was about 100 F. I then let the car idle til it got to about 180 F. NO MESSAGE!
This morning I ran my errands, took about 1 hour city driving with 1 stop when I had to shut down. At 163 F I got the message again. This was before my stop & shut down. As I continued my trip, the message went out for about 3 minutes only to return again.
BTW I put in a new thermostat in June along with an aluminum housing.
My guess is that something happens to the magnetic float after time that causes it not to operate properly.
This morning I ran my errands, took about 1 hour city driving with 1 stop when I had to shut down. At 163 F I got the message again. This was before my stop & shut down. As I continued my trip, the message went out for about 3 minutes only to return again.
BTW I put in a new thermostat in June along with an aluminum housing.
My guess is that something happens to the magnetic float after time that causes it not to operate properly.
#23
#24
#25
I'm not a pro on these magnetic sensing devices but....
The jet skis I work on use a similar floating-magnet-on-a-rod set up to measure fuel level in the tank. After many years they up and just quit working. Don't know why. Also, the speedometer on lots of watercraft use a magnetic spinning paddlewheel. These, too often just quit working after some years. So maybe our magnet somehow loses it's strength and it's weakened effectiveness is influenced by the temperature of the coolant?
Do I remember from high school science class that you can remagnitize a magnet by rubbing it against another?
By the way, the nearly universal picture of the expansion tank shows a sensor already in it. I wonder if it really comes with one...
#26
#28
I may have missed something but it seems we're speculating about different things that might cause a false low coolant indication without having checked to see if the coolant level maybe actually is lower when the indicator shows up.
How about that cold-start, cap-off test, watching for a change in coolant level when "low coolant" shows up?
If the magnet were shot, we should have a constant indication, but I don't know if it would be low coolant or not.
How about that cold-start, cap-off test, watching for a change in coolant level when "low coolant" shows up?
If the magnet were shot, we should have a constant indication, but I don't know if it would be low coolant or not.
Last edited by Dennis07; 09-06-2012 at 06:27 PM.
#29
I may have missed something but it seems we're speculating about different things that might cause a false low coolant indication without having checked to see if the coolant level maybe actually is lower when the indicator shows up.
How about that cold-start, cap-off test, watching for a change in coolant level when "low coolant" shows up?
If the magnet were shot, we should have a constant indication, but I don't know if it would be low coolant or not.
How about that cold-start, cap-off test, watching for a change in coolant level when "low coolant" shows up?
If the magnet were shot, we should have a constant indication, but I don't know if it would be low coolant or not.
I pulled the tank off again and cleaned the place where the sensor is fitted.I couldn't tell if it was dirty or not. Right now I don't have the message with the key on. I have some errands tomorrow morning so I'll see if it comes back.
Last edited by MRomanik; 09-06-2012 at 07:18 PM.
#30
Did the cold start cap off. Coolant never went low,it actually overflowed out the top.The low coolant light was on.
I pulled the tank off again and cleaned the place where the sensor is fitted.I couldn't tell if it was dirty or not. Right now I don't have the message with the key on. I have some errands tomorrow morning so I'll see if it comes back.
I pulled the tank off again and cleaned the place where the sensor is fitted.I couldn't tell if it was dirty or not. Right now I don't have the message with the key on. I have some errands tomorrow morning so I'll see if it comes back.
I see. Seems my idea is on life support. Hope your case is resolved.
It occurs to me that we should be able to test the sensor by wiggling the float up and down by hand. We should be able to turn on/off the low coolant warning at will this way if the magnet and sensor are good.
The following users liked this post:
MRomanik (09-07-2012)
#31
had the same problem with mine. The first problem that I found my radiator hoses were clasping when the engine cooled down and coolant was being discharged out of the overflow. As soon as the engine warmed up or thermostat opened i would get a low coolant message.
stopped the message by removing the cap (COLD ENGINE) make sure that the res. is full, squeeze on the radiator hoses and see if air is comes goes into the tank. If it does you have an air lock. If you can not get the hose to collapse when squeezed your thermostat (the small hole in the top) is not working. I replaced the hoses and thero and housing.
Next problem was the coolant sensor light started coming on about a month later again. Went to the Jag parts dept and they said that the coolant sensor does not fail on a regular basics. He recommended to remove the sensor clean it and the tube that it slides into. Also run your finger in the tank and make sure your magnet is moving freely. This solved my problem and have not had a coolant message for over 12k.
hope it helps Jim
stopped the message by removing the cap (COLD ENGINE) make sure that the res. is full, squeeze on the radiator hoses and see if air is comes goes into the tank. If it does you have an air lock. If you can not get the hose to collapse when squeezed your thermostat (the small hole in the top) is not working. I replaced the hoses and thero and housing.
Next problem was the coolant sensor light started coming on about a month later again. Went to the Jag parts dept and they said that the coolant sensor does not fail on a regular basics. He recommended to remove the sensor clean it and the tube that it slides into. Also run your finger in the tank and make sure your magnet is moving freely. This solved my problem and have not had a coolant message for over 12k.
hope it helps Jim
#32
had the same problem with mine. The first problem that I found my radiator hoses were clasping when the engine cooled down and coolant was being discharged out of the overflow. As soon as the engine warmed up or thermostat opened i would get a low coolant message.
stopped the message by removing the cap (COLD ENGINE) make sure that the res. is full, squeeze on the radiator hoses and see if air is comes goes into the tank. If it does you have an air lock. If you can not get the hose to collapse when squeezed your thermostat (the small hole in the top) is not working. I replaced the hoses and thero and housing.
Next problem was the coolant sensor light started coming on about a month later again. Went to the Jag parts dept and they said that the coolant sensor does not fail on a regular basics. He recommended to remove the sensor clean it and the tube that it slides into. Also run your finger in the tank and make sure your magnet is moving freely. This solved my problem and have not had a coolant message for over 12k.
hope it helps Jim
stopped the message by removing the cap (COLD ENGINE) make sure that the res. is full, squeeze on the radiator hoses and see if air is comes goes into the tank. If it does you have an air lock. If you can not get the hose to collapse when squeezed your thermostat (the small hole in the top) is not working. I replaced the hoses and thero and housing.
Next problem was the coolant sensor light started coming on about a month later again. Went to the Jag parts dept and they said that the coolant sensor does not fail on a regular basics. He recommended to remove the sensor clean it and the tube that it slides into. Also run your finger in the tank and make sure your magnet is moving freely. This solved my problem and have not had a coolant message for over 12k.
hope it helps Jim
I cleaned the inside of the tank but not the tube the sensor slides into. I can't imagine it loads up with dirt & oil enough to effect the readings, but you never know...
Last edited by Spurlee; 09-07-2012 at 08:53 AM. Reason: Clarity
#33
Scott,
You said right at the start you changed your sensor ?
Have you done Dennis's top off overflow test?
If you get the same result it has to be electrical.
It might be worth checking the sensor feed to the ECM.
As usual the ECM puts out 5V on pin 23 (for a 2004) which is grounded by the reed switch in the level sensor.
It looks as if the reed is closed when the level is high so with no warning light you should get 0v there.
If you push the float down you should see 5V and get a warning.
If you know from the tank overflow test that the coolant level is OK then if you're getting 5V when you get your warning the fault can only be in the sensor (new), plug connection, ground connection or harness.
Obviously a heat related open circuit in any of these would give a warning.
Why it should happen so reproducibly beats me though.
You said right at the start you changed your sensor ?
Have you done Dennis's top off overflow test?
If you get the same result it has to be electrical.
It might be worth checking the sensor feed to the ECM.
As usual the ECM puts out 5V on pin 23 (for a 2004) which is grounded by the reed switch in the level sensor.
It looks as if the reed is closed when the level is high so with no warning light you should get 0v there.
If you push the float down you should see 5V and get a warning.
If you know from the tank overflow test that the coolant level is OK then if you're getting 5V when you get your warning the fault can only be in the sensor (new), plug connection, ground connection or harness.
Obviously a heat related open circuit in any of these would give a warning.
Why it should happen so reproducibly beats me though.
The following users liked this post:
MRomanik (09-07-2012)
#34
Scott,
You said right at the start you changed your sensor ?
Have you done Dennis's top off overflow test?
If you get the same result it has to be electrical.
It might be worth checking the sensor feed to the ECM.
As usual the ECM puts out 5V on pin 23 (for a 2004) which is grounded by the reed switch in the level sensor.
It looks as if the reed is closed when the level is high so with no warning light you should get 0v there.
If you push the float down you should see 5V and get a warning.
If you know from the tank overflow test that the coolant level is OK then if you're getting 5V when you get your warning the fault can only be in the sensor (new), plug connection, ground connection or harness.
Obviously a heat related open circuit in any of these would give a warning.
Why it should happen so reproducibly beats me though.
You said right at the start you changed your sensor ?
Have you done Dennis's top off overflow test?
If you get the same result it has to be electrical.
It might be worth checking the sensor feed to the ECM.
As usual the ECM puts out 5V on pin 23 (for a 2004) which is grounded by the reed switch in the level sensor.
It looks as if the reed is closed when the level is high so with no warning light you should get 0v there.
If you push the float down you should see 5V and get a warning.
If you know from the tank overflow test that the coolant level is OK then if you're getting 5V when you get your warning the fault can only be in the sensor (new), plug connection, ground connection or harness.
Obviously a heat related open circuit in any of these would give a warning.
Why it should happen so reproducibly beats me though.
Yes, I did replace the sensor but when that didn't solve anything and the problem persisted I sent it back. Now I am using the old sensor.
Sounds like you know a lot more than I about reed switches - I will try some testing as you suggest and report back... I'm going to study up on reed switches now that I know what they are called.
#35
Scott,
It would be very rare for a reed to malfuncton in this way.
After many 1000s of cycles they can fall apart, but tend to fail permanently rather than intermittently.
I'm intrigued by this - the only electrical answer I can come up with at the moment would require the switch to work the other way - open when the level's OK and closed when it isn't.and if you've got a slight coolant leak which gets into the connector it could ground the ECM input until things got warm and dried it out but that would mean the sensor was 'fail dangerous' - open circuit was good - and I don't think even Jaguar would go down that route.
Interested in more detail about your S/S top lines - have you posted about these?
It would be very rare for a reed to malfuncton in this way.
After many 1000s of cycles they can fall apart, but tend to fail permanently rather than intermittently.
I'm intrigued by this - the only electrical answer I can come up with at the moment would require the switch to work the other way - open when the level's OK and closed when it isn't.and if you've got a slight coolant leak which gets into the connector it could ground the ECM input until things got warm and dried it out but that would mean the sensor was 'fail dangerous' - open circuit was good - and I don't think even Jaguar would go down that route.
Interested in more detail about your S/S top lines - have you posted about these?
#37
one way to find out if it is electrical or the float and float sensor is to disconnect the plug to the float sensor. Then put a shorting wire across the two terminals of the connector. Leave it disconnected from the sensor in the expansion tank. Now the tank and the sensor are out of the picture. The warning message should be off all the time regardless as to where the coolant level is. Drive the car. If you still get the crazly messages the problem has nothing to do with the coolant level, sensor or float. It is in the electrical wiring somewhere.
#38
Scott,
It would be very rare for a reed to malfuncton in this way.
After many 1000s of cycles they can fall apart, but tend to fail permanently rather than intermittently.
I'm intrigued by this - the only electrical answer I can come up with at the moment would require the switch to work the other way - open when the level's OK and closed when it isn't.and if you've got a slight coolant leak which gets into the connector it could ground the ECM input until things got warm and dried it out but that would mean the sensor was 'fail dangerous' - open circuit was good - and I don't think even Jaguar would go down that route.
Interested in more detail about your S/S top lines - have you posted about these?
It would be very rare for a reed to malfuncton in this way.
After many 1000s of cycles they can fall apart, but tend to fail permanently rather than intermittently.
I'm intrigued by this - the only electrical answer I can come up with at the moment would require the switch to work the other way - open when the level's OK and closed when it isn't.and if you've got a slight coolant leak which gets into the connector it could ground the ECM input until things got warm and dried it out but that would mean the sensor was 'fail dangerous' - open circuit was good - and I don't think even Jaguar would go down that route.
Interested in more detail about your S/S top lines - have you posted about these?
My S/S Top Lines are from Colliflower - Colliflower, Inc - Your ParkerStore for Hose and Fittings - http://www.colliflower.com/default.html Gus worked with them to assemble a kit for the XK8. They were expensive - over $300 US as I recall. The kit includes a variety of 90 degree fittings to make the pump side connection. Top shelf components; much, much beter than OEM. There is an OEM set of hoses on eBay (US) right now.
I also rebuilt the top latch through a company called Top Hydraulics. Top Hydraulics | Rebuilt Convertible Top Hydraulic Cylinders - Top Hydraulics, Inc $150 US Upgraded seals, excellent workmanship. He may have the latch in stock by now, I was one of the first. My top latch blew right after I replaced the hoses - and I have the pressure reduction valve...
With the hoses, pressure valve, latch and Remote Top feature I have almost $1,000 in hardware invested in the stupid top. On the other hand, I don't worry about using it like I did before...
Last edited by Spurlee; 09-08-2012 at 08:18 AM. Reason: Clarity
#39
one way to find out if it is electrical or the float and float sensor is to disconnect the plug to the float sensor. Then put a shorting wire across the two terminals of the connector. Leave it disconnected from the sensor in the expansion tank. Now the tank and the sensor are out of the picture. The warning message should be off all the time regardless as to where the coolant level is. Drive the car. If you still get the crazly messages the problem has nothing to do with the coolant level, sensor or float. It is in the electrical wiring somewhere.
#40
Debugging procedure
This test is confirmed by experiment to work, on my car anyway. For use in sorting out low coolant warnings ...
By reaching inside the expansion tank, and holding the float up or down, it is possible to turn off or on the low coolant warning. I think it's fair to say that for any car which acts this way, we know that the magnet on the float, the reed switch (sensor), and all downstream electronics are good. At least at a particular point in time.
Conclusion: if a warning appears on such a car, seems like it must be a true low cooloant condition, or a sticking float.
Doing this test is a little tricky though. The low coolant condition is apparantly checked every 15 seconds or so (roughly). No immediate feedback either inducing or removing the low coolant warning. The test can be done with the engine not running but ignition on.
FWIW ...
By reaching inside the expansion tank, and holding the float up or down, it is possible to turn off or on the low coolant warning. I think it's fair to say that for any car which acts this way, we know that the magnet on the float, the reed switch (sensor), and all downstream electronics are good. At least at a particular point in time.
Conclusion: if a warning appears on such a car, seems like it must be a true low cooloant condition, or a sticking float.
Doing this test is a little tricky though. The low coolant condition is apparantly checked every 15 seconds or so (roughly). No immediate feedback either inducing or removing the low coolant warning. The test can be done with the engine not running but ignition on.
FWIW ...
Last edited by Dennis07; 09-08-2012 at 09:19 AM. Reason: clarity