Hi all. Probably a non runner but here goes.

I am building a Haynes roadster with Sierra rear diff (3.62 : 1 I think) probably with 195/50 x 15 tyres and a Mazda 1.6 engine and gearbox. The Mazda would normally have a 4.1 diff fitted with 185/60 x 14 tyres.

I want to use all the Mazda electrics including the speedo. As I see it it will read 88% of the speed it recorded in the MX5 due to the diff ratio change. Does anyone know how I can get it re-calibrated? (I have thought of printing off a new speedo dial altered to indicate correctly and glue it over the original ).
Ian

Ok I've done some checking and find that I have a 3.92 diff not the 3.6 that I had been led to believe. That means that my speedo should read 95.6% of what it did. I've got some 195/45 x 15 wheels and tyres so they should read 3.65% faster. That means I will end up with the speedo reading 99.25% of what it used to. I am depending on the original being a little optimistic so hopefully it will not under read. If it was acurate then at an actual 30mph it would read 29.775mph, not enough to tell the difference given the thickness of the needle (I hope).

I have a problem checking before IVA as I have no local roads to check against a GPS to get actual speed. So, I have a question to those who have done IVA. How do they actually check speedo accuracy on the test?

You should be OK at that. It'd be better to be able to check before but not easy legally.

At IVA they'll put the car on a set of rollers that have a calibrated speed readout. Then they'll run it to set speeds on the rollers (30/50/70 I think) and check the speedo against the rollers.

So my biggest error with them would be at &0 where it should be reading 69.475 if the original set up was 100% accurate - fingers crossed then - thanks

Doh! 70 - fingers wont work

Yes, if it was originally spot on. They very rarely are though; Usually over reading by 5-10% in which case you'd be OK.

Your speedo has to read equal or higher than the rollers actual speed (max +10% from memory).