This tip was provided by Mike Moore in Cheshire, UK.

These vapor pressure temperature gauges are not difficult to repair but there is a problem of getting hold of the fluid that goes in the brass bulb. I’ve repaired a couple of these, one in the mid 1960s and the other, an MGA one last year. When doing the first one I used diethyl ether (DEE) to go back into the bulb just because I had some lying around from the days when I mixed my own model diesel engine fuel. Those were the days when, barely out of short trousers, you could go into a chemists shop with a corked empty pint bottle and get them to fill it up with ether albeit leaving a dent in your pocket money. Last year when I called in several local chemist’s shops and asked for ether I got blank expressions with comments back like ‘never heard of it’ or ‘even if we had it we couldn’t sell it to you’! A year ago (2009) you could buy it on Ebay but only in large quantities (gallons) costing the earth.

This got me thinking about whether there was an alternative to DEE. I first had a look on the internet to see if Smiths had actually used DEE in the early post war period but couldn’t find anything. Then I decided to find the pressure required, using an air compressor, to give a full scale deflection of the temperature gauge: this turned out to be 100 psi which is, of course, exactly the same as for the oil pressure gauge. So Smiths had actually used identical Bordon tubes in the construction of these two gauges.

Now I looked for a liquid that would have a vapor pressure of 100 psi at 110 deg C (230 deg F). The closest I could find was n-pentane giving 100 psi at 109 deg C and next closest were DEE at 107 deg C and ethyl mercaptan at 105 deg C. Because of the close link of vapor pressure with boiling point all three of these liquids have a boiling point of around 35 deg. C.

However, even though n-pentane is a constituent in petrol I wasn’t able to find a supplier of the stuff in small quantities at a reasonable enough price. In the end I asked an industrial chemist I knew whether he could get hold of something giving him the choice of the first two and he came up with 100 cc of DEE which is plenty because only 2 cc are required to refill the bulb.

For my two gauges both leaks had occurred near the brass bulb, one at the soldered joint and the other through a split capillary which I removed by shortening it by an inch or so.

I found the best way to insert the fluid into the brass bulb was to drill a small hole, around 1 – 1.5 mm diameter, through the tip of the bulb large enough to insert a hypodermic syringe needle and inject a couple of cc into it. Soldering over the drilled hole to seal the system is a bit tricky because you obviously can’t use a naked flame and it must be brought up to soldering temperature quickly otherwise the fluid at the bottom will boil away at 35 deg. C. I used a large soldering iron of 75 – 100 Watts and held a wet rag around the bottom of the bulb to keep the fluid cool.

You have an additional problem of getting hold of some new 1/16 inch capillary tubing but there seems to be plenty around on internet sites, although when I looked last year I couldn’t find a supplier of the protective coiled steel wire that surrounds the capillary. ………… Mike