I swapped in a T56 into my camaro, and until I can afford to get a digital cluster (I want all my gauges to match so probably going to buy the auto meter phantom set eventually) I'm thinking about just making a little arduino gadget to drive my speedometer cable based on the pulses from the T56. I'm pretty sure I read the T56 sends out 17 pulses per drive shaft revolution. With that I can figure out my speed based on the pulses and my tire circumference easy enough. I just need to know how many revolutions the T5 spins the speedometer cable at per drive shaft revolution so I can wire in a DC motor to match. I could do an LED display and skip the cable easy enough but I wanted to keep my odometer accurate for now if I could.

 

Cable speedometers are supposed to run at 1000 revs per mile. Dunno about the '89-up VSS pulse count.

 

The rate the cable spins is related to the driveshaft by the speedo gear. IOW, it varies.

The # you REALLY need is: 1000 turns of the cable per mile.

 

Thanks!

 

Just wanted to post and confirm something. Shopping for the motor I'm going to use to power this and wanted to make sure:

Both of you said 1000 revs per mile, is it a mile distance wise, or 1000 revs per mile per hour? It makes a pretty big difference, if you are going 60 mph and it's 1000 revs per mile then that is 1000 revs per minute, but if it's 1000 revs per mph then that is 60k revs continuously. I'm assuming it's the former but just wanted to be sure and order a motor capable of the required revs. 1000 revs per minute is pretty easy to achieve, cheap < $1 dollar motors can do that (although I'd probably go with a higher price one for better quality), but 60k revs is a pretty high amount of revs for a small hobby motor, not sure I can even find one to spin that fast (and actually I'd want it to be capable of going to 120 so I'd need 120k revs continuously at that point!). That is partially why I'm assuming it's the former, makes a lot more sense that way.

 

How exactly do you plan on controling the speed of the motor? The factory cable isn't spinning a set rpm. It varies with speed. So you would have to have something constantly controling the speed of the motor to even make this work.

 

Yeah that isn't a problem. I can simply vary the voltage output to a basic motor, or I can go with a stepper motor like this one: https://www.sparkfun.com/products/10551 where I can control the number of steps it takes and how fast it takes them to control the RPM of the motor. That one in particular I linked is capable of 2000 RPM (which would work out to max 120 MPH if it's 1000 revs per mile distance wise) if I run it as fast as I can. Stepper motors are better because they can handle a little bit of torque resistance that might be present in the cable, and they also give me more precise control.

EDIT: Oh and as far as what actually controls all of that, it would be a program that I write on an ATMega chip. Arduino based most likely, I'd probably use the ATMega328P, unless I can get away with using an ATTiny chip for a smaller footprint. Essentially, I'm duplicating what the Dakota Digital box does except I'm going to send the output into a motor to control my speedometer cable instead of converting to a different digital signal. I'm also doing it for much cheaper than a dakota digital box costs =p Where it costs 85 dollars for one of their boxes, I could duplicate the function of their box for less than 15 dollars.

 

It's 1000 revs per mile.

Forget about the speedo and speed and miles per hour for a minute; simplify.

THINK about the odometer.

Pretty simple little gizzy... shows how far the car has moved, eh? well, its input turns .... 1000 times ... to move it one mile.

NOW it's OK to go back to thinking about the speedo.

If those 1000 turns happen to occur in exacty one minute, then the speedo needs to read exactly 60 mph. If he cable spins 500 times in one minute, that means the car only moved half a mile in one minute, so the speedo needs to indicate 30 mph.If it spins 250 times in a minute, 15 mph. 2000 times in a minute, 120 mph. And so on for every other number of turns of that cable per minute.

As always, not rocket science; just a car. Really really simple stuff for the most part.

 

If you're going electric speedo, many people use the Dakota Digital SGI-5.

If you're going cable, the Abbott Cable-X works; another version is under development by some folks; a thread about it is on ls1tech. There's an early one and a late one. My experience with the early one is a delay in moving off zero. After that, it's cool.

There are existing bike speedometer adjustment devices (VSS in, electric out) which cost less than the SGI-5 but kind of have fewer features. I'd have to look but there's at least one on eBay I was considering but my hang-up is I need to build my speedo first.

1001 cable rpm = 60mph on Camaros through 89. The speedo gears thus change gearing of the ratio of driveshaft revolutions to speedo cable rpm. They had a 2 pulse per rpm VSS built into the speedo head (so 2002/mile)

Electric speedos in third gens are all 4004 pulse / mile input. But the signal from a VSS is divided so the CCC, TBI and MAF TPI ECMs get a 2002 pulse / mile input.

It's kind of tough to search and find just the right threads to find all this information across third gen and other places but it's out there.

 

Thanks for the info.

I will be going to an electric speedo eventually, but for now I'm going to continue to use my cable speedo. As stated previously though I can build a device that serves the same function as the SGI-5 for a whole lot cheaper.

And the Abbott Cable X... bahaha that thing is even more ridiculously over priced. I understand companies make things and want to make a profit, but 280 dollars for a device that can be duplicated for less than 15 - 20 dollars. Give me a break.