ROG100

Only available from Porsche - they restocked. NLA according to the Wholesalers.
I have some in stock.

NC928S4

Before I ordered a unit from Roger I decided to test the existing coolant float assembly to see what is actually wrong with the device. The problem has been that the “check coolant level” warning goes on after the car reaches operating temperature. After it cools down it functions normally.

Firstly, it is a PITA to remove. The float buoy moves without friction and has 0 ohms when low and approximately 1000 ohms when fluid is at correct level. I’m making these measurements in my kitchen so any temperature effect of that may effect the buoyancy is not taken into account. Also, I need to test if the float gets more porous as it heats up, which would make it float less.

Does anyone know if these resistance levels are normal. I have not found any documentation.

dcrasta

The float will react differently in water than coolant (density and all). (2cents. for testing buoyancy use coolant).

NC928S4

I looked up the density of water and glycol densities versus temperature and both go down but not dramatically as temperature increases. Thus the float will become less buoyant(because it displaces less mass of coolant) and sink as temperature heats up and cause the problem. I suspect that the float has absorbed fluid and other particles over time and is now heavier which makes it very sensitive to density changes of the coolant as temperature rises.

The float is a piece of plastic wrapped around a magnetic collar that causes the resistance to change when the correct level position is reached. It’s pretty heavy so it’s easy to see how the displaced coolant by volume may not be enough to float the float.

Rather than hack a new floatation gizmo to increase buoyancy I’ll be placing an order with Roger. At least I can be reasonably confident this is the root cause unless this basic float mechanism has some bizarre relay dependency on the hatch release mechanism.

SanDiego928

What is this you are saying here? Does this affect parts for my 81?

Mrmerlin

FWIW if you have a suspect sender its best to replace it,
fitting a used unit you will usually find its FUBAR

shorting the wires together will not make the light go out,
removing the original from the tank is a bit of a task and should only be done when you have a new unit to install.
Once you remove one of these senders you will appreciate the value of slipping in a new part VS swapping out a few used units hoping that it will work. BTDT.

The electronic system on the sender is usually what fails once it cant sense the proper resistance then the light turns on, sometimes the float also fails.
Again removing the failed part without a new part to slip in will be very frustrating.
You can also use a O ring for the cap seal if the original is somehow damaged.

I hope Roger can find more of these parts

jcorenman

Stan,

Roger said he had some in stock, above. I just ordered one from Sonnen Porsche for $55.

Cheers,

NC928S4

Thanks for the feedback guys. Roger is shipping a new unit today. For grins I’m going rig up a 1.5Mohm resistor across the leads to mimic a full coolant tank to keep the light off until the new unit arrives.

heinrich

Randy I'll give you 5 bux for it

heinrich

agreed

heinrich

Rog, what prevents us from tossing out all that old weird stuff and installing a newer, cooler design? Complete tank I mean. Existing seems like overdesign.