Speedo 2.0

[Gis]

Purists look away, this is a restomod. I’ve been struggling with a bouncy speedo needle for some years now. Managed to improve it a bit over time but never managed to fully get rid of it. During that time I’ve changed the speedo cable, cleaned up the speedo and re-sleeved the angle drive and used a bracket to hold it down into the gearbox, greased the whole system etc etc. Nothing really worked that well and the bounce still resurfaced. The source of the bounce is either the angle drive (NLA) the pin gear that connects to the angle drive or the drive gear that sits on the end of the main shaft in the back of the box. Possibly a combination of all three. So the options were to pull the box find and change the gears (expensive option) or look elsewhere to bypass the box. I just couldn’t stand the bounce any longer, the one instrument we always look at. It was time to deal with it. 

So in comes the Dakota digital electronic to mechanical cable drive (ECD-200BT). For a signal source I created 6 simple small brackets that attach to the gearbox side of the driveshaft using the drive shaft screws. Each of these brackets has a small magnet glued atop. The sensor is attached atop via a bracket to the frame. The inboard brakes obviously don’t make things easy here. Using a powered 3 wire hall type sensor allows for a relatively large gap (6-10mm) to the magnets. This is needed since the engine/gearbox combo sits on rubber mounts and might move relative to the frame.

So it’s 6 pulses per wheel rotation, the ECD needs a ppm (pulses per mile) baseline and that was approx 4720 ppm with my current tire size. Setup is done via Bluetooth and an app. Dead easy. Installation of the unit was the usual combo of lotus yoga and scratches since it involved digging into the binnacle and laying new cables and relays etc but proved well worth it. A test drive was needed to verify and slight adjust the ppm (using a GPS speed app and the DD app again). My wife was doing the calibration while I was driving 🙂. So now I have accurate speed and a rock steady needle. Love it, Real happy with that. This solution also allows for flexibility and recalibration to different tire sizes.

A technically better solution would be to use a tone ring as a reluctor and a passive 2 wire sensor but that would mean that the gap between sensor and reluctor is really small (<1mm) so it will not be possible to hang the sensor off the frame. I’m also thinking of using the brake disk itself as a reluctor and machine shallow grooves into the circumference and then mount the passive sensor off some of the screws that hold the gearbox to the engine. It would require a custom bracket but seems possible. Might go for that at some stage. That would be speedo 3.0

the ECD is a really well build unit, cost is 330 USD, the Ford clip-on cable works well with a few simple modifications to fit it to the smiths speedo. Hall sensor, brackets and magnets are a few quid only. should I ever want to go back to the original setup it’ll only take the effort to reconnect the old cable on both ends, so fully reversible. 

[ChrisJ]

Very neat solution.

The Smiths instruments in my Esprit don't bounce around at all, but the VDO ones in my Excel, well that's another matter. I fitted a HUD to get over that issue.

[Christian43]

thanks for the post. I have had exactly the same problem as you. The speedo has now been rebuilt twice, as has the angle drive at the gearbox end. I even took the back of the gearbox off to check the internal drive gear. Have replaced the cable as well. The only thing I have not changed is the sleeve but I just can't believe its that. As you say its pretty irritating as otherwise my car is well sorted. I may try your solution.

cheers

C43

[Gis]

I’ll post some pictures of the sensor setup in the back as soon as I find some time to crawl under it again. As outlined in the post; best would be a tone ring + passive sensor setup. The inboard brakes don’t make that easy. I bought a spare rear disk to experiment with to find a more sturdy solution. The current setup works fine for now. 

[snowrx]

Simple VSS for an older G car- Hall sensor on an angle bracket reading 6 magnets epoxied to outer CV joint. 

[Gis]

8 hours ago, snowrx said:

Simple VSS for an older G car- Hall sensor on an angle bracket reading 6 magnets epoxied to outer CV joint. 

That’s exactly the same hall sensor I’m using. I weatherproofed mine a bit more with shrink tape. How long have you been running this setup Thomas? Any feedback on reliability?

[Christian43]

BTW like you I have changed nearly everything in terms of the speedo. The only part I have not changed is the outer speedo cable. I am not sure how this can make the speedo wobble but was wondering if this could have an effect, did you try changing yours? Also did you try grease on the cable? I am using graphite lubricant but its quite dry and therefore no damping effect. I was wondering if grease on the cable would add some damping? I know but long shots but it is strange that we are having this problem. 

[skiing]

Mine has always bounced all over the shop... anyone else managed to sort this without fitting an electric version?

[fjmuurling]

On my 1990 SE (VDO gauges) it was a loose/slipping magnet inside the speedo. Had it repaired a looooooooooooooooooooooong time ago (worked perfectly after) but recently it started to bounce again. 

[Gis]

6 hours ago, Christian43 said:

BTW like you I have changed nearly everything in terms of the speedo. The only part I have not changed is the outer speedo cable. I am not sure how this can make the speedo wobble but was wondering if this could have an effect, did you try changing yours? Also did you try grease on the cable? I am using graphite lubricant but its quite dry and therefore no damping effect. I was wondering if grease on the cable would add some damping? I know but long shots but it is strange that we are having this problem. 

I changed the  complete cable inner and outer Christian. It improved it a bit but didn’t cure it. Greasing also didn’t make a difference. 

[NG5]

Almost all car speedo cable outers since the 1970's have been nylon lined. I've not cut an Esprit one in two to check, but there is no reason to assume it's unlined. (Motorbike ones are often unlined)

Nylon lined cables should never be greased. The grease swells the nylon causing binding. I don't know about the effect of a dry graphite lubricant, but the cables are designed to operate dry. The second effect of grease on a cable is that it can work it's way into the speedo which can damage the speedo.

With heavy use or tight bends the nylon liner can wear through in places which also causes binding. So it's never a good idea to put a new inner in an old outer even if it works ok to start with, it won't have the life of a new inner in a new outer.

The most likely cause of a bouncing needle is too tight a bend in the route of the cable, and the longer the cable the less tolerant of tight bends the inner cable is. It winds itself up and then releases. It's always worth checking your full cable route and making sure all bends are as big a radius as possible.

Bouncing most of the time is caused by the cable. Occasionally you get damage to the drive gear teeth which then slip causing bouncing, but if the teeth look ok they usually are. If the gauge becomes stiff (usually because of grease contamination) that can cause the cable to wind up causing bouncing, normally the cable doesn't last long if that's the case. It's easy to check the speedo with it out. Use a match stick or similar to turn the speedo with your fingers. It should be smooth and almost effortless.

Speedograph Richfield http://www.speedograph-richfield.com/ repair Smiths Speedos and instruments at a very competitive price. I don't know if they do VDO ones though. They also repair speedo angle drives. I've used them a few times with excellent results.

Speedos don't bounce for no reason. It's got to be the drive, cable or instrument, there is nothing else.

 

[NG5]

One other issue I've just remembered. It's possible sometimes not to seat the drive cable properly into to drive or instrument when attaching the cable. (quite common on old mini's where access to either end is difficult)

You would think 1/4 turn of the cable in use and it would seat, but it never seems to, just drives using end friction. The inner compresses into slack space in the outer and binds causing bouncing

[snowrx]

13 hours ago, Gis said:

That’s exactly the same hall sensor I’m using. I weatherproofed mine a bit more with shrink tape. How long have you been running this setup Thomas? Any feedback on reliability?

I think this went on as a quick and dirty test setup to give speed info to my Megasquirt ECU for boost by gear and datalogging, but I can't remember when. Photo meta data says 2014. It's been dead reliable for me, although I used a metal-filled epoxy (JB weld) to attach the magnets and it was quite messy when the iron in the epoxy mounded up over the neodymium magnets. Someday I'll clean it off and redo it.

[Gis]

Here’s the setup on the inner CV joint. Works well. Love the steady needle 😀

[Clive59]

Those magnets look neat, did you make the brackets and heatshrink magnets to them?

[Gis]

Brackets are from a local hardware store and modified. The thin neodymium magnets are glued on with loctite 638. 2 layers of shrink tube provide some protection against the elements since they can corrode.