Instrument calibration

When I bought my Giulietta Sprint, the previous owner warned me that the speedo was offset, and that an indicated 80 kph was in fact only 50 kph. It was probably because of the low gearing at the differential, he said. I set out on the journey home, and saw that the odometer was showing true distance, so decided that I would crack the mystery.

I would take the instruments apart and solve the problem.

Peeking behind the dashboard for the first time

Peeking behind the dashboard for the first time

Right after removal; the instruments are not pretty

I had seen on an italian Giulietta Spider blog how the owner had gotten his instruments dismantled, cleaned and reassembled at some specialist in Italy, and I was very much inspired by the pictures and videos that he had posted on the subject. I had no idea if the instruments upon opening would explode with springs and gear wheels all over the place … or indeed how such an instrument would work. But, as with anything mechanical, it has been designed carefully for dismantling and repair; it is not that difficult at all. Back in 1954, Alfa Romeo chose Veglia as supplier for the three main instruments: A triple-scale clock for petrol level, oil temperature and water temperature, a rev counter with an oil pressure gauge (with a direct oil tube connection from the engine) and the speedometer with a trip-counter and odometer.

While the instruments were disassembled I painted the binnacle

While the instruments were disassembled I painted the binnacle

These instruments, as well as the car itself, were not designed for a 60-year life. This became evident when I saw that the inner brackets of the mechanicals were riveted together. I would have to drill (or otherwise open) the rivets in order to properly disassemble and clean the parts. But I was also surprised at the simplicity of the design. It turns out that speedometers have been made this way since a german inventor, Otto Schultze, took out the patent for his eddy-current speedometer in 1902, and the method was only replaced by the electronic speedometer around 1980.

But last year, when I took the instruments out, I had no idea about such facts. I just went at it and examined the inner workings. I could see that the pointer was spring-loaded, and that it was driven by a disc that somehow was pulled by a magnet-disc of some kind, mounted on the drive-cable. Check out Canada casino online and win real big money! After cleaning and lubricating the mechanism, best as i could, I tried different solutions for calibration. The speedometer was running out of the scale at 180 kph, when the car was actually doing 130 kph. And dyno runs at Skallebøle Autoværksted had shown that the rev counter had a 10% offset as well; it was showing 6500 rpm at 5800 rpm.

I tried to extend the distance between the two discs by inserting shims under the screws that held the bridge together inside the clock. But it was impossible to arrive at decent readings; it was all too sensitive, and it was very cumbersome to take the instruments out, disassemble them, install shims, rebuild the thing and take it for a test drive … etc etc. So in the end I just lost patience and reinstalled all the instruments (now very clean) and settled with the fact that I would have to find that a guy in Italy who had refurbished Alejandro’s clocks.


The internals of an eddy current rev counter

I went all summer 2013 with these mis-readings, and grew accustomed to the offsets. But over the winter I decided to buy new scales and chrome rings for an aesthetic refurbishment, and picked Veloce-type items with 2000-8000 revs and 20-220 kph indications in stead of the Normale-type that came with the car (1000-7000 rpm and 20-180 kph). I had regained hope (sic) and my idea was to really take the instruments apart; open all the rivets and go for it; somehow tighten the spiral springs and use the opportunity to get a clean slate.

Much easier to do most of the work without test drives

Much easier to do most of the work without test drives

In the video you can see how I actually did it; it turned out that the needle can be picked off the spindle, whereupon the spiral spring can be tightened a bit, and, with and the needle in place again, it will now rest at the stop with higher pre-load. This means that activation requires more torque (from the rotating magnet) and that the reading has been changed to a lower value. It also meant that I could do the calibration at my workbench, without having to go for numerous drives to confirm readings. I would just go for a couple of confirmation drives, and then maybe do the last little fine-adjustment afterwards.

With the instruments removed, I could take time to undo the electrical spaghetti behind the dashboard

I is quite evident that the electrical wiring in my car has evolved over a couple of owners. The car was built in march 1962, imported to Denmark in april 1962 and sold to a Mrs. Jette Weiercke as BJ.62200 in july 1962. She kept it until 1975, when it was de-listed and eventually rebuild as a FIA-GT race car by Frank Erwing in 1996. And it shows! But the car works fine, and all the important appliances work satisfactorily to have earned the car another 8-year lease of life as AB.60346.

I will build it as a 1959 Giulietta Sprint Veloce – and I am almost there!

Leave a Reply