I've got a T56 with 4.11 gears in my car and I'm running a 9396 ECM running a modded A1 codemask. I've calculated that I'm at around 55,000 ppm with the T56's 17 ppr VSS and 275/40r17 tires.

I can get the ECM to see the correct speed but I understand that over 44,000 ppm the ECM can no longer divide the value enough for the 4,000 ppm output that the speedo is looking for (speedo reads high).

So my question is whether there is an easy way to divide the VSS signal in two, that way I end up at close to 28,000 ppm which the ECM should be able to divide to get a fairly accurate speedo.

 

I got this.... I have 3.73 and a T5 in a 91 RS. I could not find the drive gear so i went this way. I am still deployed so it will be a few months till I am able to get it all installed.
http://www.hawksthirdgenparts.com/pr...Interface.html

 

Thanks, but I'm looking for something simple and cheap . I only need to divide the signal in half; if it comes down to spending $100 ill probably just mentally figure out speed based on RPM.

Anyone know if teeth in the VSS reluctor can be removed? I think I remember somewhere reading about 93 T56s having an 11 tooth reluctor. How difficult are they to find/swap out?

 

You could use a DRAC or VSSB to divide the signal, simple, cheap (auto wreckers), and adjustable, by changing jumpers (or installing DIP switches).

 

Hmm is that anything like the VSS buffer box that comes in our cars? I also figured a 7730 could be used to divide the signal but probably WAY overkill and not nearly as cheap .

I found a site that describes how to modify a DRAC/VSSB but I'm unclear on what jumper settings to use to divide the signal in two.

 

Not the cheapest way but easy nonetheless. I use this also

 

the DRAC route would work great.

the "simple" method i was coming up with turned out to be slightly less simple... and then i went full banana and decided to see how difficult it would be to come up with a full microcontroller based solution to do more than simple integer division. turns out, not that much more difficult for a massive increase in capability/accuracy.

 

Is a VSSB described here the same as the VSS buffer box that came stock in my car? Too bad I threw it in the trash...

 

SAAR Could you tell us more about your micro controller based solution

 

i don't think any 3rd gens got a DRAC.... so i'd say the only use your buffer had was to convert a cable speed into a square wave signal to be fed into the optical VSS input.

looking into using an ATMEGA to measure the 4K PPM output signal and knowing the actual PPM that will be generated(since getting it to line up perfectly almost never happens), then applying a 0-2 multiplier to that time period and recreating the signal to actually output 4K PPM. or 4004 PPM, either way.

it's obviously easier said than done, but it would make circuit cost REALLY low.... probably less than $10 for a fully adjustable unit. anyways, should allow the PCM to accept anywhere from 2,000 to 88,000 PPM VSS signals, have the PCM do it's divide, send it to the ATMEGA, have it apply it's correction, then send it on to the cluster with excellent accuracy.

i IMAGINE.... that the optical stuff could benefit from this as well.

 

Hmm weird my car has always had an electric speedo and there was never a speedo cable in the car. Isn't the optical signal generated from a cable speedo?

That ATMEGA sounds like a better solution either way so keep me posted please!

 

Correct, an '89 Firebird would have a magnetic VSS in the trans tailhousing. Gear driven the same as a cable speedometer set up. The 4-out buffer box used is completely different then a DRAC.

RBob.

 

Note that a DRAC may not do the trick. The lowest divide ratio is 0.500, which brings the tooth count to 20. This is still higher then the 17 in the T56. I'm not up for the math this early in the morning, but with the rear ratio and tire diameter it may be just enough to work.

As trucks usually ran 4.10 along with a larger diameter tire (they cancel each other out some).

Note that the Dakota Digital SGI-5 is only around $80. Just be sure to twist the wires from the VSS to the SGI-5 box.

RBob.

 

But if it divides in half, the ECM would be able to divide it further. That was my idea anyway.

 

For the '7730 'type' ECM, the magnetic input can be divided by 1, 6, 7, 8, 9, 10 or 11. This divisor only affects the output data streams. Not the internal VSS calculation.

So yes, if the 17 tooth wheel is in that ballpark, it can be used. Note that the rear ratio and tire diameter will also affect this. Along with the ECMs internal divisor that affects it's MPH calculation.

RBob.