Smiths Mechanical Temp Gauge Capillary Tube Repair

[anduril3019]

I came across this article on lotuselans.net about repairing capillary tube gauges and it seemed like fun to give it a shot. I just happened to have a water temp gauge I bought cheap on ebay thinking I might use the face and needle on another gauge, so it seemed like a good candidate for repair with nothing to lose.

 

The article on lotuselans.net seems to be based on some other articles with good details:

 

https://forums.lotuselan.net/t/smiths-dual-oil-water-temp-gauge-repair/24987

 

https://mgaguru.com/mgtech/dash/dt102b.htm#:~:text=Measured diameter of tubing on,is used for electrical wiring).

 

https://www.ply33.com/Repair/tempgauge#:~:text=Do not use a torch,fit for the old capillary.

 

I won't rehash the links above, for more "how-to" detail, read those first, but here are a few things I ran into.

 

The first thing of note is the article suggests finding a cheap mechanical gauge at Advance Auto Parts for $15. That was in 2018. I didn't look too hard, but the best I could do was about $30 for a Dorman gauge on Amazon. One thing to watch for, some of the gauges advertised as analog, or mechanical, use an electronic sending unit and wire that is covered with a metal coil that looks like a capillary tube, don't get one of those.

 

If there is any tube left on the gauge, I think the best way to attach the new tube is to sleeve and solder as mentioned in the article. Mine had a very short stub left so I first tried to de-solder and remove it, but felt like it was getting too hot without releasing the solder. So, I filed the stub flat and drilled it out. The new tube measured .050" diameter, so I drilled the gauge back out with a #55 (.052") drill. That proved to be too tight, and the next best thing I had on hand was a 1/16" drill, so, a little loose, but in the end it worked. I drilled it about 3/16" deep and made sure the ID was clear using a sewing needle.

 

The next step was getting cold water to keep the bulb contents (ether?) in a liquid state. Out of curiosity, I wanted to see what temperatue difference I could get between plain ice water, and salted ice water. Plain ice water got down to about 33° by my kitchen thermometer. I took that chilled water, strained out the ice,  and was able to dissolve about 1 1/2 tablespoons of salt in it. After adding fresh ice, the temp dropped down to about 25°.

 

The next step was chilling the bulb and cutting the tube. I was able to work the coil shield back a couple of inches and expose the capillary tube. I cut it with a cutoff wheel so as not to crush it, then dressed the end and cleaned it out with the sewing needle. I tinned a section of the tube before cutting thinking it might ensure I didn't clog the tube with solder. After cutting the tube, I was trying to push the coil back a little further and got a little spurt of liquid out the end of the tube, I must have compressed something. So care is warranted, and safety glasses!

 

Now is when I found my drilled hole to be too small. The tinned end fit fine in the 1/16" hole, not too loose. Since it was hard to de-solder in the beginning, I was worried about getting enough heat into it with a soldering iron and no flame, but it actually soldered pretty quick and seemed like a well flowed bond. Now is when I wished I'd thought of some kind of strain relief. I'll add something after now. 

 

After what seemed like a successful connection, on to testing! As my water temp rose to around 100° there was no needle movement, the gauge starting at 90°. So, I figured, oh well, nice try, and was deciding whether I would try again or not. But as the water temp rose, the gauge started to move! At boiling temp, it's within a few degrees of my thermometer. Probably not as accurate or responsive as a modern gauge, but good enough that I can call it a working gauge! I compared it to a working dual gauage I have, and temp read about the same.

 

All-in-all, I'm calling it a success!