Coolant light on but weird symptoms

igeak691 · Post by **igeak691** » Thu Apr 10, 2014 4:00 am

I've got an odd issue at hand with the coolant light. I had to replace my expansion tank when I bought the car, there was a hairline crack which is very popular with these cars. So after I flushed the coolant, replaced tank, and refilled it with coolant, I bled the system when the heater and engine on. No problems then. After driving around a few times, the coolant light turned on and would go away after about 20 minutes or so of driving. So I thought there was a leak, however, I checked the marker when the engine was cold and it had not moved at all. So I know there is no leak, even to this day. So here's the weird symptom. The coolant light would turn off after about 20 minutes or so of driving, but that was when the coolant expansion tank was overfilled with coolant. When I removed the coolant so that the marker was between max and min, the coolant light would be constantly on, no matter how long I drove. Have any idea what I should look into? I verified the working condition of the sensor itself as well as the wiring. It works as it should, but something inside the coolant expansion tank isn't sitting in the right position, I'm assuming. I've looked on many threads, and even some that cuts open the tank to see what's inside. Here's some info that I know though:

If the circuit is closed, the coolant light is off. If the circuit is open, the coolant light is on. The car checks once when put into the "on" position, but does not constantly check, I verified this by having my dad watch the light on the dash and bridged the contacts over and over again simulating the sensor function. Any more ideas as to what to check? Should I overfill the coolant expansion tank again (I heard this was bad because it would overpressurize the system, but on the other hand I also heard it will bleed any excess coolant)?

bcworkz · Post by **bcworkz** » Thu Apr 10, 2014 5:12 pm

It still sounds like a sensor issue to me. They can work fine when tested cold, then when heated, they stop working correctly. Or it could be reading the level incorrectly even though it is turning on and off correctly. I don't know enough about how it reads level to speculate in more detail. It sounds like a temperature related issue anyway.

igeak691 · Post by **igeak691** » Sat Apr 12, 2014 3:22 am

It measures coolant level by a magnetic reed switch and the magnet is inside the coolant tank, whereas the sensor sits in a compartment that never touches coolant. When the magnet comes close to the switch, it closes the circuit, thus turning OFF the coolant dash light. That's as much as I know about how it works. I just don't get why when the level of the coolant is correct, that the magnet doesn't fall on top of the switch.

bcworkz · Post by **bcworkz** » Sat Apr 12, 2014 8:30 pm

What I've seen happen many times is there's a marginal connection somewhere in the sensor circuit. When cold, everything is tight and the connection is closed and everything works. Once heated, everything expands, loosening up, causing the marginal connection to open, so the sensor fails to work.

Thanks for the sensor explanation. I hate to question information you so kindly offered, but are you sure it's a reed switch? The only reason I ask is there are many magnetic related position sensors on the car and every one I know anything about is a hall type sensor. Seems a bit odd they'd go to a reed switch here. OTOH it could explain things as hall sensors are usually very reliable,all hall sensor failures I've heard of were from improper positioning, not because the sensor failed.

igeak691 · Post by **igeak691** » Sun Apr 13, 2014 1:42 am

You are probably correct that it is a hall magnetic switch and not a reed. That just rolled off the tongue for some reason.

The thing was that before the expansion tank failure it would read correctly, so I don't think it just so happened that after I fixed the expansion tank that it started having issues with a marginal connection. And I have proven (I guess not proven, but seen through trial and error) that the coolant level affects whether the light goes off. If overfilled, the light goes off after the engine warms up and I've driven for about 20 minutes. If filled correctly, the light never turns off. So what should I do.. overfill the expansion tank? It's supposed to bleed off excess anyway right?

bcworkz · Post by **bcworkz** » Sun Apr 13, 2014 10:24 pm

Yeah, excess coolant would get spewed out on the ground, then be replaced by air when the system cools. Still, I can't get myself to recommend overfilling anything, even though I can't imagine what would be the problem. It's just wrong!

I agree it does seem like the thing to do.

I still can't wrap my head around how this thing measures level. I get the sensor part, but how does the coolant level move the magnet? The sensor appears to be a plastic stick a few inches long inserted into the bottom of the tank. How can that relate to the coolant level several inches higher?

igeak691 · Post by **igeak691** » Mon Apr 14, 2014 2:28 am

From what I've gathered, the floaty thing has a magnet on the bottom of it, so that when it drops, it covers the sensor. But that doesn't make sense with our cars since having the magnet over the sensor would make the circuit close, and turn OFF the warning light, when that's the time that it should turn it on since the level is low. I'm postulating that there is a separate compartment for the z4 coolant tanks, but I unfortunately threw mine out before thinking about dissecting it..

bcworkz · Post by **bcworkz** » Mon Apr 14, 2014 4:25 pm

That can actually work I think, if the hall chip is in the plastic stick, the magnet could fall past it to trigger the warning. I'm now wondering if the tank isn't quite a direct replacement even if externally identical, so that the magnet location is mislocated for your sensor. If so, perhaps you need a matching sensor for that tank? Grasping at straws really, I can't imagine there being parts so similar yet incompatible. It's just that this is sounding like a mislocated sensor, except in this case there is no adjustment. I'd be very tempted to glue a magnet to the sensor to disable the light and just manually check the level from time to time, just like I do for every other car I've ever owned.

I'm sorely tempted to launch into a rant about electronic crutches for idiot drivers, but I'll leave it at that.

igeak691 · Post by **igeak691** » Tue Apr 15, 2014 3:35 am

Yeah I get what you mean. I can easily bridge the connection so the light doesn't turn on, but periodically checking surely isn't as safe as a light that actually works. I definitely plan to do some more research about the height that the stick should be, but you're probably right about small differences on the inside that makes this magnet fall at the wrong height.

If my coolant light turns off after 20 minutes of driving, and this is when it is overfilled, that means my magnet is too low when filled correctly.

Mind doing a test for me? Have the expansion tank cover off, and hold the stick at the highest level, then turn on the car and see if the coolant light is on. It SHOULD be on I think. I think the magnet falls right over the sensor when the top of the stick is right where the cover would push it down to. As the coolant level drops for whatever reason, the magnet will fall BELOW the sensor and cause it to trip.

This is based off the fact that the light turns on when the circuit is open, which means the magnet is NOT over the sensor.

bcworkz · Post by **bcworkz** » Tue Apr 15, 2014 6:35 pm

I'd be willing, but my car is actually still inconveniently located in cold storage. Still lots of snow on the ground here

Roads are clear, but it's still fairly cold. All just rationalizations, I'm actually just busy with a few other things and just don't have time to go get the car right now.

It'd likely be quite a while before I can get back with results. No idea if anyone else is following along, perhaps they could do the test for you.

igeak691 · Post by **igeak691** » Wed Apr 16, 2014 1:11 am

Thanks. Not a pressing issue but would be nice to have it figured out. I never noticed any residue even when the tank was overfilled so my guess is that it can be a temporary fix.

bcworkz · Post by **bcworkz** » Thu May 15, 2014 1:56 am

Sorry for the long delay, I finally found some time to sneak away and go out and play with the Zed for a while. I did your little test, at least I think I did. I was reluctant to open a hot radiator cap, so I let it warm up without the cap on. Without the cap the stick floats to the highest possible position on it's own. (It's always been this way, I never made any attempt to alter the level, this is the level when I bought the car over 4 years ago) The radiator lamp never illuminated. For extra information, I also pushed the stick down with a nut driver so the top is barely above the internal stop several inches down. No radiator light in this position either. I don't know if this is normal or perhaps my level sensor is also malfunctioning?

I don't recall you mentioning what variant you have, maybe it is a M54 vs. N52 thing? I'm personally not particularly concerned because I check the level regularly despite having never had to add any coolant to the system. However I don't think my findings really tell you anything about your car.

igeak691 · Post by **igeak691** » Fri May 16, 2014 4:50 pm

I'm on the M54 engine and thanks I appreciate you doing the test. I noticed that the coolant light signal gets checked every so often and not in real time. I mean, it doesn't make sense to continually check it, and I think once a minute is just fine. As I think is apparent now, the level of the actual bobber isn't what decides whether the light goes on or off. There's a separate compartment with a magnet float somewhere that definitely triggers the light. When I drive for over an hour, I'll notice the light goes off for a few minutes and comes back on. Oh how lovely those few minutes are without any yellow light on the dash

I was told by my dad to just bypass it and trick it into submission, since I open up my hood to check fluids every so often anyway. Since the expansion tank is new, I don't foresee issues coming up out of nowhere, and the car has been real reliable in the few months I've had it. So my next course of action is just to bridge the contacts so the light will be gone.

Lynnbg123 · Post by **Lynnbg123** » Sun Jul 11, 2021 10:04 pm

Any new updates on this problem, I just replaced my expansion tank, and my light is doing the same thing.

Z4-forum.com

Z4-forum.com