One of the features I added to the instrument panel early on was a low coolant warning system. This was comprised of a DIY coolant probe that went into the expansion tank, a small circuit with a coolant level chip, and an idiot light on the panel.
Initially, the coolant I used was NPG, which is made by a company called Evans and is basically propylene glycol instead of the antifreeze you get in the local auto store which is ethylene glycol. The rationale here is that the boiling point for NPG was way higher than you could actually achieve, and ran at 0 psi. A good deal if you are worried about blowing out hoses under pressure.
The DIY coolant probe worked “OK”. I had to build two of them over the years because one started leaking. Basically it was just a brass end cap with two small holes drilled and shielded wires passed through to act as the probe. The whole thing was sealed with JB Weld and it worked… well, “OK”.
Delorean forward a few years. After battling cooling issues for a long, long time, I finally swapped in larger radiators and went back to using generic antifreeze for its better heat transfer properties. But that of course meant running a pressure cap of 15-20lbs. I did not have confidence in the DIY probe under pressure, so I removed it and “INOP”ed the idiot light.
Eventually I got annoyed by not having a coolant level check without having to pull the top cowl off, so I started looking at off the shelf probes. I actually found the ideal probe, but after numerous calls to distributors and the manufacturer, I could not track down a single part. I’m still baffled by that one – you would think companies who advertise their products would make it easy for babbling idiots saying “Shut up and take my money!”, but not in this case.
So, back to the DIY drawing board.
The original fitting is basically just a regular old 1/4″ NPT threaded fitting. That makes part sourcing easy. Mostly. I ordered a couple of stainless 1/4″ NPT compression fittings for rigid 1/8″ tubing. I also ordered some 1/8″ steel rod to use as a probe.
The trick here is obviously electrical isolation between the probe material and the fitting. Since the compression ferrule was also stainless, and the rod was a close tolerance fit in the fitting itself, I had to make a few changes.
I ordered some teflon ferrules of the appropriate size to replace the stainless ones. I also had some heat shrink tubing good for 135C. Since the engine would be “in trouble” if it was running hotter than 230F or 110C, I felt that it was OK, but I would have liked a little more margin.
The idea was to drill out the fitting to accept the rod covered with the heat shrink tubing as an insulator. The ferrules would rest on bare metal to hold and seal the rod in the fitting, then another piece of shrink would isolate the exiting probe end from the screw on cap.
After breaking a drill bit inside the first fitting, I very carefully drilled out the second fitting by using lots of cutting oil, and stepping up the drill sizes slowly until I had the correct sized hole through the whole fitting.
Once I was able to push the probe through the fitting with the shrink wrap on it with some modest resistance, I called it done. I then flattened the end of the probe to be able to attach a wire to it and dropped a blob of solder on the whole thing to hold it in place. It’s hard to solder to steel, and I would have preferred a more secure method, but adding a shrink wrap cover to hold everything in place seemed to work fine. I put the connector from the old probe on to the new and added a ring terminal for the ground side.
I finally installed and test ran the engine and everything looked 100%. I was very pleased with the way this turned out. Definitely going to need a little bit of test flight time near a runway just to be sure, but I sure am glad I have my coolant level probe back.