Hi,
I'm trying to make a level sensor for my coolant reservoir. I think that I can use a capacitive system (capacitance changes when liquid comes between 2 metal plates)

. I had a system like that on an RV water tank and I've been testing various sensing methods with a milk jug and some water.

The thing that I'm struggling with is a clamping system that will stand up to the temperature range. My thought at this point is to have metal plates that are slightly larger than the reservoir and to screw them into plastic rods made of something like noryl or polyester. I want to use a clamping system and not any glue or tape.

I'd like to tap into the ends of the rods, but I don't know if that's a strong enough method. I hope that it is so that I could easily take everything apart.

I'd appreciate any thoughts on clamping the plates on and whether delryn or polyester would work.

Thanks,

hj

 

Well, I machine Delrin (NOT acetal but REAL Delrin) all the time and it would be plenty strong for your intended use. Small diameter bar stock is cheap as well. Just drill and tap the ends.
I would consider using stainless for the plates and any hardware to prevent any corrosion.
Sounds like an interesting project but it begs the question of why you need to have a coolant level sensor on your reservoir. Barring major problems, it's a more or less a sealed system and shouldn't be losing much coolant. But, like I said, an interesting project.

 

Thanks for the info. I am losing some and I don't know why -- no sign of external leaks and no real sign of a head gasket.

If I do it, it will be more out of interest. Also, I think that I can rig it up with an LED or two to show status. Otherwise I have to remember to pop the hood and take a look -- doesn't seem like a big deal, but it's annoying. Not to mention that the stuff that's in there now is a little hard to see. (Yes it's 50/50).

Thanks,

hj

 

If you're looking for a binary switch (it says "Hey there's fluid here" or "hey, there's no fluid here"), why not just go on Rock Auto and put in N326-15-610? It's the coolant level sensor that goes in the top of a 86-88 Mazda RX7 radiator.

In the stock application, it supplies a ground signal when the probe is immersed in coolant to the warning lights & cluster to tell it "Everything's okay". When not immersed, it puts out no signal which makes the cluster go...(unhappy language ahead, be warned)

HOLY F**KING DOGS**T STOP THE F**KING CAR RIGHT NOW YOU STUPID F**KING R**ARD...

Done by triggering the buzzer inside the cluster.

Humor aside, mounting it is pretty simple. Make a hole for the probe, attach a block of compatible plastic with suitable threads (M12x1.5 I think, doublecheck before drilling). Add a ring terminal before threading it in and wire routed to a convenient ground location and you should be all set. If you need to reverse the sensor's behavior or have it operate something that eats current up, wire in a relay.

 

Thanks Travis, but that's not quite what I want to do. I want to sense changes in the level in the overflow tank -- before it gets to the $%^$%^&$ stage .
Also, I don't think that I could adapt the fit to what I've got.

I was hoping to do 2 or more levels with a capacitive sensor, but it's a lot harder than I thought. Not the eelctronics, I'm pretty sure that I can do that, but when I looked it over, I don't know how I could mount stuff.

Plan B is similar to your idea -- I want to run a few wires down to the liquid area much like sensors used in some sump pumps and the like. Should be fairly easy to do, but I want to get another cap for the tank. They DO make them, but they're a bit hard to find. (Rock Auto will sell me one with another tank).


Thanks again

hj

 

large trucks use the simple switch like Travis cited. as long as the coolant level is up, the circuit is grounded and the light is off. once coolant level drops the ground breaks and the coolant light comes on.

this setup gives you an indication of coolant level too, because for 2-3 days it will light up cold and go out once coolant warms up.
once it gets low enough for the coolant to stay below the sensor the light stays on.

 

I like the concept just fine -- it's all about trying to find something that will physically replace the cap on my overflow tank.

Also, it's a pretty log stretch from their to the actual tank -- and the passage is angled and narrow.( Why -- because they could.) So, the sensor has to work with those restrictions.

I'm sure I can rig something up if I can get a hold of an extra cap. Probably have to go to a yard. The days of a big line of 'Help' products seem to be over.


Thanks,

jh

 

oh, ok. i thought you were looking to add something to the side, not through the cap.

 

If a Present/Not Present switch like I suggested won't be suitable and you're aiming for a more 'analog' solution, there is another option though it's a bit more work. Use a fuel level sensor with the rod shortened to fit inside the overflow bottle and the float moved to the now shorter rod. Gonna be tricky as hell to get sufficient travel in such an odd shape (that's what she said... ) but you'll have an analog resistance-based signal that can be run to a gauge inside or a series of LEDs if you prefer.