mx5cartalk.com

Hi all,

I'm new to mx5s. Bought a 1998 nb about 3 weeks ago now.

The car felt great on the test drive so we bought it. Problems however did start to prop up that we are slowly dealing with one by one...

The biggest issue we noticed is the speedo is out by a big margin. 100kmh on the dial and you're actually 110kmh when checking against a gps. Similar, doing lower speeds it's roughly 10kmh faster than what the speedo says.

Anyone know where I should start my investigations? I understand that the sensor can fail but in this case it doesn't appear to be incorrectly operating.. Just off by 10kmh...

Appreciate any help

My NB8A speedo is about 3kph out, which seems to be the norm.


First thing to check would be tyre size, as a larger wheel/tyre will cause the car to travel faster than what the speedo ways.
Standard size is 195/50/15, which works out to be about 576mm diameter. Any larger and the speedo will be off.

The other possibility is if a previous owner has fitted the diff from an NB8B.
the NB8B had a 3.6 ratio diff, which is roundabout 10% taller than the NB8A diff, if they didnt also use the late model speedo sender, the speedo will be out 10%



Of course, both those possibilities should only affect the speedo by a percentage, rather than a flat 10kph across all speeds. (I assume at rest the speedo needle is on the stopper?)


edit:
worth noting that the NB speedo senders are electronic, unlike the mechanical ones of the NA series.
You may also want to check the wiring is all ok, but i doubt that's the issue, if that were the case you'd expect the problem to be intermittent at best.

hks_kansei wrote:My NB8A speedo is about 3kph out, which seems to be the norm.


First thing to check would be tyre size, as a larger wheel/tyre will cause the car to travel faster than what the speedo ways.
Standard size is 195/50/15, which works out to be about 576mm diameter. Any larger and the speedo will be off.

The other possibility is if a previous owner has fitted the diff from an NB8B.
the NB8B had a 3.6 ratio diff, which is roundabout 10% taller than the NB8A diff, if they didnt also use the late model speedo sender, the speedo will be out 10%



Of course, both those possibilities should only affect the speedo by a percentage, rather than a flat 10kph across all speeds. (I assume at rest the speedo needle is on the stopper?)

Thanks for the fast response.

Running stock wheels and tyres.
Needle could be out 10%, ill have a more careful check tonight. The needle is at rest on the stopper.

Where is the sensor located on the gearbox? I have ramps, so is it better to lift the front or rear to access it?

"IF" is it out 10kph at all speeds. Then it sounds like the needle has been played with in the dash cluster at some stage. Most like removed and not replaced properly at correct angle.

"IF" is was out 10% at all speeds. Then it wound be electrical or mechanical with optioned issue already noted.

Theenyx5 wrote:Where is the sensor located on the gearbox? I have ramps, so is it better to lift the front or rear to access it?

Random internet pic, but it's the one on the passenger side with the two red wires. (pretty sure on the car though it's all inside some loom tube, so you wont see wire colour)
Blue wire one on top would be reverse light switch



Access wise, it's dead centre so it's not that much difference front or rear.
I'd lift the back since that way you have the more empty end of the tunnel for access room (where from the front it's take up by bellhousing)

Never heard of the sensor misreading. Speedos normally read faster than actual not the reverse. Run yr Gps to see what its doing at 60 before yu get pinged. Like HKS said diff rato or wheels / tyres also chk.

Mr Morlock wrote:Never heard of the sensor misreading. Speedos normally read faster than actual not the reverse. Run yr Gps to see what its doing at 60 before yu get pinged. Like HKS said diff rato or wheels / tyres also chk.

It's definitely out. The gps is accurate on my Forester.
I usually drive the mx keeping in mine the speedo being out. Generally driving 10kmh less

hks_kansei wrote:

Theenyx5 wrote:Where is the sensor located on the gearbox? I have ramps, so is it better to lift the front or rear to access it?

Random internet pic, but it's the one on the passenger side with the two red wires. (pretty sure on the car though it's all inside some loom tube, so you wont see wire colour)
Blue wire one on top would be reverse light switch



Access wise, it's dead centre so it's not that much difference front or rear.
I'd lift the back since that way you have the more empty end of the tunnel for access room (where from the front it's take up by bellhousing)

Thanks. This helps.

First obvious question is what your rolling wheel and tyre diameter is.

jack car up from rear with enough room to work under diff comfortably(on stands!).after checking sensor get some chalk,put some thin clear datum marks on diff yoke and inside of left wheel,turn tailshaft so wheel rotates forward one turn.If careful you should be able to tell if diff has turned 3.6,3.9, 4.1, etc turns.
just as a matter of interest I drove a couple of nb's before buying mine and the were all reading low by 4-8kph(mine is about 7kph low)
Mick

noobee wrote:jack car up from rear with enough room to work under diff comfortably(on stands!).after checking sensor get some chalk,put some thin clear datum marks on diff yoke and inside of left wheel,turn tailshaft so wheel rotates forward one turn.If careful you should be able to tell if diff has turned 3.6,3.9, 4.1, etc turns.
just as a matter of interest I drove a couple of nb's before buying mine and the were all reading low by 4-8kph(mine is about 7kph low)
Mick

Will give that a try. Cheers. Will be good to work out if the diff was changed

Vat wrote:First obvious question is what your rolling wheel and tyre diameter is.

Pretty certain it's 195/50/15. Was the first thing I looked into.

noobee wrote:jack car up from rear with enough room to work under diff comfortably(on stands!).after checking sensor get some chalk,put some thin clear datum marks on diff yoke and inside of left wheel,turn tailshaft so wheel rotates forward one turn.If careful you should be able to tell if diff has turned 3.6,3.9, 4.1, etc turns.
just as a matter of interest I drove a couple of nb's before buying mine and the were all reading low by 4-8kph(mine is about 7kph low)
Mick

I'd say chalk mark both wheels.

being an open diff if one moves and the other doesnt (or moves slower) it can stuff up the measurement (since if one wheel is stopped the other wheel will move faster)
shouldnt be an issue, but if one of the calipers drags it can stuff up the measure.

but yeah, count how many turns the tailshaft makes for the wheels to turn once and you should get a rough approximation of the diff ratio. (about as accurate as you can get without opening the diff to count the teeth on the crown and pinion)

Also, if the diff is a 3.6 you have three options to fix the speedo.



1: replace the diff with a 4.1 ratio (or at least the crown and pinion)

2: purchase and install an aftermarket speedo corrector

3: purchase the speedo sender from a late NB to suit the 3.6 diff



Of them all, i'd skip 2 (unless you really want the speedo to be dead on accurate), and go for 1 or 3, depending on the budget.

noobee wrote:jack car up from rear with enough room to work under diff comfortably(on stands!).after checking sensor get some chalk,put some thin clear datum marks on diff yoke and inside of left wheel,turn tailshaft so wheel rotates forward one turn.If careful you should be able to tell if diff has turned 3.6,3.9, 4.1, etc turns.

Unlikely to be anywhere close with an open or torsen diff. The 100% accurate method is:
Have one wheel on the ground to stop rotation
Turn free wheel 10 rotations, while counting the rotations of the tailshaft
Double the number of tailshaft rotations & divide by 10 to give Ratio:1

eg Wheel turned 10 times rotates the tailshaft 20.5 times. Thus ratio is 20.5x2/10 = 4.1 (Diff is 4.1:1)
Cheers