Gauge calibration

[BDA]

Does anybody have any direct experience with mechanical Smiths temp gauge calibration?

I just sold my first fleabay item (sebring mirror. Woohoo!!) and an item I want to sell soon is a mechanical Smiths oil temp gauge. It's been sitting in my garage for most of forty years so I put the bulb in boiling water. I was going to just check that it came off the peg - meaning that the ether in the capillary tube and gauge is still there (I know there's ether in them because I broke a capillary tube once -Note: if you do that, just get another gauge. It will cost you about the same amount to have it repaired as it would cost to buy a new one). Then I thought I'd check to see how accurate it was. Maybe I should have left well enough alone because it would only register 96 deg. C in boiling water. I took it out and put it in a few times in case it needed to be "loosened up". There is certainly no external adjustment so the only thing I could do is send it off to a Smiths dealer for calibration, but that would probably as much as I would expect to get for the gauge. That got me to thinking - I don't remember ever checking the accuracy of any of my Smiths gauges over the years. I sent the gauges for my Europa to be checked over before I put them in my car, but I don't know what they actually did. For mechanical temp gauges, they may just have made sure it still had ether. I doubt other people who are selling mechanical temp gauges test them for accuracy and there's no real functional difference in the oil if it's really 4 deg. C warmer than the gauge says it is.

So, is that as accurate as I should expect? Would you sell that as a working gauge? I think I might scare off any potential buyer if I said it was off by 4%. Any comments? Thanks!

[jbcollier]

What altitude are you at?  I'm at 1800 ft and water boils at 97°C.

[BDA]

I considered that but dismissed it without checking. I just checked and Wikipedia says I'm about 320 ft. above sea level. I could be higher but not enough to make a difference.

Would you say that 4% is an acceptable error?

[EuropaTC]

I considered that but dismissed it without checking. I just checked and Wikipedia says I'm about 320 ft. above sea level. I could be higher but not enough to make a difference.

Would you say that 4% is an acceptable error?

Hi there,

Personally I'd say 4% on an old instrument is well within reason. We're so used to digital displays now that we expect that if the gauge says 99C or 43psi then it's exactly 99C or 43psi, plus or minus nothing.  But back in the 60s when these things were made ?  If a temperature gauge went into red or steam came out then we'd worry but otherwise I think we'd just be happy and carry on.  Even now, as long as mine stops roughly around the same point on the dial I'm perfectly happy whether it's the indicated 85C or not.

As an aside, over in the UK the law used to allow a 10% margin on speedometers, so if you were doing 32mph in a 30 zone you'd be ok.  34mph and the officer might sharpen his pencil if he was having a really bad day.  In another life I remember being told "+/- 5% experimental accuracy" was reasonable, although of course these days we expect much more with modern techniques.

But....   these are 50yr old designs even if this is a new reproduction made yesterday. 

So I'd leave it alone. If selling I would proudly state in the advert "functioning and tested in boiling water"  I would show a picture of the gauge reading 96C alongside a digital thermometer if you have one. (those £2 things from Ebay for cooking are surprisingly useful at times)  The buyer will see what they're getting. If they wish to recalibrate they can, but I'll bet they won't bother. 

Brian

[Bainford]

JB makes a great point regarding elevation, but I also agree with Brian that 4% is not that bad for a vintage instrument. I don't know about Smith's gauges specifically, but many some industrial gauges are calibrated by plucking the needle from the spindle and replacing it in the correct location.

[BDA]

When you say that a gauge is only off 4%, it sounds a lot better than what it looks like on the gauge (the sweep is 300 deg. I guess. It's not the 80 deg. sweep of the gauges on a Europa which is why I didn't use it)! If I had a good way of taking the bezel off and putting it back on, I might try calibrating the needle, but I'm more concerned about buggering it up so I'll sell it as is and as Brian says, the buyer can do that if he wants to. Hopefully he won't ding me too much on the price. Thanks for all your input, guys!

[BDA]

So why am I reviving this stale thread? I had the occasion to view this graphic that I think I got off the either the lotuseuropa yahoo group, lotustalk, or maybe it was here! Anyway, my concern that my temp gauge read 4 degrees C less than the actual temperature seems insignificant. This graphic is supposedly from the Lotus about Smiths temp gauges. The accuracy is plus or minus 7.5 degrees!

[EuropaTC]

Now that is an interesting post BDA, I've not seen that graphic before so thanks for reviving the thread. I'd guess someone has been irritated by the lack of numbers on the dials and gone to the trouble of cross checking with ye olde thermometer - well done that man !

But now......     it would appear that my car now runs at around 75-80C, just a smidgen over the 75C mark.  I think I shall gloss over the numbers, assume my gauge is one of the 7.5% "low" reading batch and wait until it changes position on the dial   

Brian

[BDA]

Or you could pull the gauge and sender out of the car and test it on the stove! 

[2483R]

You can check the gage on the stove, I have done it several times. Slowly warm the water, as the gage does not react that quickly. Remember to use the voltage stabilizer, or use a 10 volt source,  when wiring everything up. Improper voltage will cause false readings.

On the back of the gauge are 2 small holes revealing a plate inside the gage. Adjustment of the gage is by sliding this plate, use a very small screwdriver. Each hole has a different purpose in the adjustment.

But if your gage reads 96 degrees when water is boiling it sounds close enough to me.

[jbcollier]

Alas, his is a mechanical gauge.  I don't know of any external adjustment points.

[Mike Mileski]

Plus, I would bet if you checked the water temp at different locations within the engine and the radiator, you would find variations of lots more than 4 degrees. It's just not enough to be concerned about. If you start to have overheating issues, your 4 degree inaccurate gauge will surely tell you.

Mike Mileski
Tucson, AZ