When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Some time ago I got the wild idea that another sender could be adapted to use in the 928. I bought a bunch of them and then had some different screw caps made. It was a nice project, but I didn't learn until it was ready to go that the new sensors work opposite from the 928. DUH!
Some time ago I got the wild idea that another sender could be adapted to use in the 928. I bought a bunch of them and then had some different screw caps made. It was a nice project, but I didn't learn until it was ready to go that the new sensors work opposite from the 928. DUH!
Unrelated to Kevin's question, but I always thought that you could probably set up a couple of relays to switch the signal on Jerry's parts. It complicates things, for sure, but would at least provide a path forward. Mind you, I'm not the person to design that circuit, but some googling indicated that it's possible with off the shelf parts.
To my thinking the list in that document encapsulates all the failure modes:
The float no longer floats the way it's supposed to.
The reed switch is broken in some way (the reed itself, or the connection between the reed and one of the connections. This might include oxidation or gunk buildup that impedes good electrical connection).
The connections from the reed switch are broken someway (either on the way to the connector, or at the connector itself).
I have never had my level sensor out, so I don't know how easy it would be to diagnose, but I imagine that some simple continuity tests would help to clarify where the fault is in the circuit.
With one, when the float drops it breaks a circuit. With the other when it drops it makes a circuit, which I think is intuitive. I think the 928 sender is the one that breaks a circuit and somehow turns the warning light on.
Was on a road trip in the 84 and got the low coolant alarm. Pulled over and the coolant was fine. Back at home, removed the coolant lever sensor to discover it was full of coolant making it sink. Drilled a hole in it, drained out the coolant, patcged hole with JB Weld and went around the top, bottom, and perimeter of the flost, let cure, and reinstalled.
Worked for 2 weeks and then false low coolant warning again - bought new sensor from Roger and been fine for over a year.
Leads me to believe the failure happens on the inside of the float where it rides up and down
These sort of devices are old hat technology. Nowadays sensors mounted on the side of the tank are the way to go and can be retrofitted.
My expansion tank is typical of most such examples and is probably on borrowed time. To cover for failure I purchased an after market Mercedes Benz tank that is almost identical but does not have level detection capability. I this regard I did a bit of research and found this item linked below that one simply sticks onto the side of the tank. Whether they are any good [or not] remains to be seen.
Seems the Land Rover guys have some nice stuff developed for them.
I have always been fascinated by their 300 TDI rad Overflow Tank with built in sight glasses.
Past discussions here on Sight Glasses brought up issues such as it is 'a weak point' introduced into the system.
Their forums swear by this one as well as other versions of sight glasses
Last edited by Shirah; 07-01-2023 at 09:21 AM.
Reason: add
Unrelated to Kevin's question, but I always thought that you could probably set up a couple of relays to switch the signal on Jerry's parts. It complicates things, for sure, but would at least provide a path forward. Mind you, I'm not the person to design that circuit, but some googling indicated that it's possible with off the shelf parts.
I never thought of that, but it is an interesting suggestion. What is needed is a relay that has a constant connection, but when it is energized it breaks the connection. I suppose there is such a relay.
The other problem with my conversion is that the sensor/sender is about an inch shorter than the one in the 928. I have always had in the back of my mind that it may not be possible to simply fill the overflow tank about an inch above the level mark on the tank and have it stay there versus draining the excess during the normal heat cycle back to the level mark. If the coolant level can be held at about an inch above the level mark then I think the relay suggestion might bring my sensor project back to life.
I don't mean to hijack this thread, but what I have to offer is at least a step-brother to Kevin's question.
Last edited by Jerry Feather; 07-01-2023 at 03:16 PM.
I never thought of that, but it is an interesting suggestion. What is needed is a relay that has a constant connection, but when it is energized it breaks the connection. I suppose there is such a relay.............
Yes Jerry, the center pin (87A Terminal) of a 5-Pin Relay does that. It is normally closed (outputing +12v) until the coil gets closed and then the 87 Terminal gets 12v out
I built a working replacement from the one with a good float and a bad reed switch.
I had another level sensor without a float. Not even sure where it came from or how the float went missing.
I tested the reed switch with the remaining magnet.
Cut off the tip, removed the magnet ring. I slid the float on and glued the tip back on. Gorilla glue rocks.
I then tested the complete assembly in a bowl of water to confirm the float closed the reed switch.
06-28-2023, 10:52 AM
Why do coolant level sensors fail
