Odometer question?

I tried the search function to answer this question and came up empty.  What drives the trip and main odometer in the 1996 Roadmaster wagon?  Is it gear driven off the drivetrain? Controlled electronically from the PCM?

I finally figured out why the gas mileage seems so poor on my new (to me) wagon.  The trip odometer is consistently reading 28% less than the driven mileage.

The spedometer is also reading 68-69 mph, but GPS consistently indicates 75 mph. So, the spedometer off by about 10%, considerably less of a disparity than for the odometer.

how are you determining driven mileage ?  trip odo different from regular odo ?
what size tires do you have and what is the PCM programmed for ?
what gears ?

I haven't had a RM cluster out in a LONG time, but IIRC the trip odo just has a small gear attached to a bigger main gear.   There shouldn't be a disparity between your odometer and trip odometer. PCM reprogramming should be all you require to ensure accurate odometer reading.

Maybe Fred has one lying around and can take a pic for you.

The trip and regular odometer are in sync, so the error is the same for both.  Miles traveled is from known distances (183 miles, my house to Albany exit), and from GPS miles traveled.  Same degree of error by both methods, so that's not a factor.

Tire size is 235 70 15 vs OEM 225 75 15.  Revs per mile differential is minimal (1.2%); nothing near 10 to 28%.

PCM was programmed while in custody of previous owner, who made a number of changes that are the subject of other recent threads.  Long story.

Did you mean spedometer rather odometer in above reference to PCM reprogramming?

dmg4 wrote:

Did you mean spedometer rather odometer in above reference to PCM reprogramming?

no. Reprogramming the pcm should have addressed all of them (speedo/odo/trip odo) on the 94-96 RMW. I wasn't sure how you were calculating your error differential. I would have been astounded to find your trip odo was 28% different then your regular odo. Your speedo should not be registering any significant error from the odo either. A small error from the GPS would be ok, but not 10%. Civilian GPS aren't exceptionally accurate.

As a quick sanity check, are there odometer check points on the road in NY ? I've seen some in the prairies and used them to verify my tune (found a mistake in a tuner's email tune with one). Hack the odo at the start. If the stretch is deserted you can set cruise prior then hack the clock at 0. d=v*t
also, verify what your ECC (14 IIRC) to see what the pcm is outputing.

Driven past that same 10 km verification stretch with 3 different roadmasters with tunes and odo/trip odo/speedo have always been defined by the tune parameters. That being said, garbage in, garbage out.

I didn't read your threads and don't have time to look them up.

if you email me a copy of your OBD2 tune I can look at it for you, but you'd be on your own for reloading it.

I was under the impression that in the sender unit there is a gear that is replaceable. They make (made) the gears in several tooth sizes and used different colors to designate them. The differences helped account for different rear end ratios among any GM vehicle that used the 4L60 transmission. It was the primary GM V8 tranny during the 90's for rear wheel drive vehicles.

The sender transmits a pulse to a module via one wire. The module is near the steering wheel base, where the signal is split, one to the ECM and one to the cluster.

Maybe somewhere along the line it had transmission service, or was replaced from a different car, like a Camaro, or truck using a different rear, and the wrong sender gear was either left in, or the wrong one installed.

I would also point out that the only experience I have is with the 93 model year. It is possible that the method was changed when the LT1/4L60E combo came out.

A transmission swap is quite possible on this car.  It has 114,000 on the odometer, but the transmission pan looks like it was installed yesterday: brand spanking new.  One of the many surprises I am finding.  Yesterday I discovered that the previous owner had installed some heavy duty bags inside the rear coils, bypassed the load-level system, and manually inflated them to dumptruck pressure.  You learn something new every day in this car.

That explained the kidney-puncher ride quality on all but the smoothest surfaces.  Deflated to 5 psi now to save the aged fillings in my molars.  I'll get rid of these soon and install some new coilover shocks in the rear.  I've run those on my previous three wagons, and they are perfect for me.

""It is possible that the method was changed when the LT1/4L60E combo came out."

Yes. The LT1 cars use the same speed transducer regardless of axle ratio. Adjustments are made in the PCM.

Thanks for the update, Rev.

Is it correct that reprogramming of the PCM will correct both the odometer and the spedometer, or do I need to change a gear that is driving the odometer, and then reprogram the PCM to correct the spedometer?

Reprogramming the PCM should correct your problems. You need to have entered the correct rear and ratio and tire size. I would not rely on what the previous owner has told you. You need to check for yourself.

If I were you, I would swap the PCM's between your two wagon's (providing they are both 96's). If it is the PCM, as I think, the problems should carry over to your Caprice and the RMW should smooth out. May not be perfect, as I'm not sure what rear end you have in the RMW, but the issues will be a lot closer than what it is now.

Tom

Tom

It's going to take someone more savvy than me to read the values that have been entered in the PCM. Tire size is 235/70/15. Service code sticker on the car indicates it came with 2.56 rear gears. Could I estimate the approximate ratio from the speed at 60 MPH as indicated by GPS and engine RPM on a flat, straight, deserted road?

You could use your method as a sanity check, but because GPS speed is imprecise it would make it hard to discriminate between certain close ratios  (but you should drive it to verify speed output on the ECC and compare to speedo and gps).  Those radar panels you see on some roads are more accurate than the gps.  
Easiest method: Spin the wheel count the driveshaft rotations.  Don't forget open diff requires 2 turns of the wheel.

do you have reason to think the previous owner changed gears ?  
do you have a digital copy of your tune ?

there's at least one underlying issue here.  You still shouldn't have a discrepancy between speedo and odo.  1st thing you need to rule out is observer error.

Also, I had no idea you had 2 wagons, so putting that in a sig could be beneficial for troubleshooting (swapping pcms is often a good test).

Dan, you need to stop trying to make this harder than it is. Take members suggestions! Providing they are both 96 PCM's, swap them. If the problem gets better, well the PCM has to be reprogrammed. You don't have to worry about reading the values.

Tom

Quote :

You still shouldn't have a discrepancy between speedo and odo. 1st thing you need to rule out is observer error.

Quote :

Dan, you need to stop trying to make this harder than it is. Take members suggestions!

Please, be patient with me. I ask a lot of questions, and when information conflicts, I ask follow up questions. It's how I learn, and it's perhaps unfortunate for my mentors that this may take a bit of time. However, it's not for lack of effort on my part.

For example, the odometer (trip and main) and spedometer conflict in degree of error by more than a factor of 2. The speedometer is off by 10%, the odometer by 26% in repeated independent tests. Trust me here. I know how to measure stuff with GPS, measured miles, and known waypoints. The above values are accurate. I've driven the car on several trips over 100 miles to confirm the values. I just drove from Geneva NY to Millinocket Maine. Same 10% and 26% error all the way by GPS, measured miles, and elapsed time (lots of elapsed time!).

I'd be happy to swap the PCM in my 1996 Caprice towpack car with 3.23 rear gears with the 1996 Roadmaster with (presumptively) 2.56 rear gears. It woudl only take a few minutes. But my original question (still not definitively answered) is what drives each meter? More to the point: why is the odometer error 26% and the speedometer reads 10% off?

David

What tuning software are you using? TunerCat has a speed correction module built in that re-sets the parameters based on axle ratio and tire circumference. Will also re-set shift points to match. Gets you within 1% of where it's supposed to be.

Bill

I'm not saying you don't know how to measure distance, but often the simplest answer is often the most likely.  People measure stuff wrong all the time.

again, this is going off 10+ year memory of having the RM cluster out once.  The trip odo has a small gear driven from a bigger main gear.  I don't have time to go look up the FSM to see if there's a diagram in it.  On 94-96 roadmaster the calibration is done by PCM.  All programs I have used have used 2 parameters (wheel size,diameter or revs/mile; and rear end ratio) to modify ODO/SPEEDO/TRIP ODO. Caprice is the same, but the digital odo data is stored in the cluster itself. Unlike the previous cars, there's no gear to change.

we can't explain your last question (why is one at 28 and the other at 10), because there are too many uncertain parameters (actual gear ratio, programmed gear ratio). Well, the only thing we can answer is that almost certainly, at the very least, your tune's parameters don't reflect your car's specifications.  By doing nothing, or by tuning, they are supposed to read the same.   If you can't call the PO and ascertain what he did (gears, tune, like did he change any speedo gears, attempting to correct the speedo ?  unlikely...) you have to figure out which, if any, of the parameters is correct.

a) go outside, jack the car, spin the rear wheel and determine actual gear ratio.  You need this info. Period.
b) while driving at a constant speed, use the ECC to read the speed (ECC function is RM specific, not sure if the caprice pcm will support it).  Does it correspond to the speedo?  odo ? neither ?
c) swap pcms.  Assuming the caprice pcm is accurate (aka known tire diameter and gear ratio)  What's the deal now ? are the errors still proportional?

edit: I have a computer with obd2 tunercats on it. I'd be happy to adjust the tune for you, but the cable is in BC. If someone has the obd2 cable at wagonfest, I could upload the tune to your pcm (but then they could realistically also do it, I guess).

The odometer and tripmeter are Geneva gear mechanisms. The motor has a pinion gear, that runs against a ring gear like the differential but no bevel, or hypoid cut, turning the first drum. The first drum has one tooth on it that turns an intermediate gear that turns the second drum, and so on.

The two meters should run at the same speed, because they use the same signal. There could be a ground or input signal fault. One or both of the motors may even be worn out, or the mechanism is binding.

The 91-93 odo and trip meters run on a 4000 pulse per mile signal. They also have a signal modifier box that feeds 2000 pulses per mile to the computer. The 94-96 cars all do this through the computer directly from the VSS. The odo and trip meters are still driven by one signal from the computer.

After a long silence, I was finally able to reach the previous owner. He was in the hospital for open heart surgery, poor guy. He assured me that it was 2.56 rear gear, but that he had PCM For Less reprogram the PCM for a 3.23 in anticipation of a swap of a spare he had on hand. I'm still in the dark as to how that results in a greater odometer error than spedometer error, but I live in the dark on so many things that I'll just add this to my list.

Car is now being fitted with four new General Altimax 235/70/15, new oil pan, valve cover gaskets, and oil temp and pressure sending units. It was leaking oil everywhere. No biggie in itself, as it was drops at a time, but those drops were falling on both cats and exhaust pipes, the interior stank like the Deepwater Horizon every time you stopped moving. Excellent rustproofing of the entire undercarriage however from the resultant oil aerosol at 80 mph. The garage will confirm the rear gear ratio when they have it on the lift.

The car will eventually be for sale for what I have into it. PM me if interested. I bought it for a friend from Norway to use when he visits with his family in late July. I'll bring it to Wagonfest one of the days.

nice.
options: put in stock pcm cal
put in 3.23 gears

Quote :

options: put in stock pcm cal
put in 3.23 gears

The previous owner no longer has the 3.23 posi rear, and frankly this car is quite fast enough for me with the various other mods already. So, correcting the code in the PCM is the best option for me.

Do you have any ABS codes or problems? With the error in the actual speed from the axle sensor, the ABS computer has invalid data to compare with front wheel speed.

Nothing showing up with the ABS yet, but I haven't had to hit the brakes hard enough for it to really be a factor yet either.

The ABS and Speedometer circuits do not have anything in common.

As my Nephew is so fond of saying "Wrong again Uncle Bob".

Oh no. I just sent the PCM out to be reprogrammed for the correct gear ratio (2.56) and tire size (235/70/15). The odometer is now spot on, but the speedometer now reads 76 mph at a constant speed of 60 mph; exactly the degree of error that was corrected in the odometer.

Any suggestions as to how I can get the speedometer to read correctly without messing up the odometer again?

On the positive side, she now does 0 to 60 in less than 6 seconds.

You may have a faulty speedo. The speedo and odo get their signal from the same place.

Quote :

You may have a faulty speedo.

The error is now a consistent 26% high from 20 to 80 mph. Before the PCM reprogramming to change the code from 3.23 rear gears to 2.56 gears to correct the odometer, the speedometer was reading about 10% low. It that a typical mode of failure?

Is it the faulty part within the instrument cluster itself, and can the speedometer be changed out as a separate part without messing with the odometer or other gauges?

The fact that it was lower initially indicates there may be a problem with the speedo. You had it programmed, and it is now worse. The odo is correct, so there may be an issue with the motor that runs the speedo. There may be a chip in the cluster that could also fail. I do not have a schematic for the 96. You could try another cluster to verify what is the issue.

Does anyone out there have a working 1996 Roadmaster instrument cluster with 117,000 on the odometer that they would like to sell?  There are several listed on eBay for $59 to $99, but they all have about 135K on the odometer.  There are "services" also listed on eBay that will.... ahem, "adjust" the mileage for you (who knew?), but I'd prefer to find one that is close to or just under the present mileage, if possible.  The bezel can be swapped from the present cluster, so the condition of the bezel is not an issue.

Posted above query under parts needed also.

You can swap your odometer into the donor cluster. Easy peasy.

Hmmm....Tell me more.  None of my service manuals cover an odometer swap.  The attached image shows a plastic cover over the odometer.  Is it accessed and removed from the back, and how is it retained? Unplug and plug wiring, or soldered connections necessary?  Is the cover fused to the circuit board?  I swiped the image off eBay.  I have not removed the errant cluster yet.

If you remove the circuit board, the odometer has screws holding it from the other side, or behind the circuit board.

It appears that the put the cover over it to allow them to make the rest of the cluster thinner, and save some plastic.

OK. Found a donor cluster. Will report back on the cluster and odometer swap in a few days to see if that solves the speedometer problem.

The FSM considers the instrument cluster to be a swappable, nonserviceable assembly so it doesn't go any deeper than that.

Or just try swapping the speedometer stepper motor into your cluster.   Now that your odometer is accurate at least the problem is narrowed down to either the speedometer stepper motor, the speedometer circuitry in the cluster, or the needle is not properly positioned on the stepper motor (probably not true for your case).

For what it's worth, I've not seen a GM analog speedometer from the 1990s that was accurate better than +/- 3mph.  In my cars they'd be high at 45mph and then low at 55 and then a little high at 60 and so on.  So at some point you should just call it good enough once you finish playing musical clusters.

OK, this problem is solved. It was a defective speedometer (Nick: now you've hear of one). I removed the dash trim and various screws holding the original instrument cluster in place, removed it, and popped in the donor cluster from eBay. Both the odometer and speedometer worked fine on the donor cluster. Next step was to remove several screws and disassemble the various layers of each cluster, and remove and swap the odometers. In the clusters I had, these were not held in place by screws, but by three very small, delicate and brittle plastic clips on the rear of the white plastic panel between the panel face and the printed circuit board. Navigated that hazard, disconnected the tiny flat ribbon supplying power, swapped the odometers, reassembled and replaced the new cluster (with working speedometer), now with original odometer, replaced, test drove, and both odometer and speedometer are spot on, plus all idiot lights and gauges are working. Time to drink beer and think good thoughts.

Somebody finally listened to me.

Quote :

Somebody finally listened to me.

I feel the same way when I get questions about plants.

Yes first one i've heard of,.
Such a complicated matter,. But once u got "stuck in" so to speak,. It was an easy remedy,.

Too often people have a problem on this forum,. And they like to ask a lot of questions,. Every question often has a choice of options,. Internet diagnosis can take some interesting turns,
But because folks who like to fix their own stuff are usually helpful folks some educated and not so educated guesses are involved ,,
Sometimes the original poster will then actually get their hands dirty and fix the problem,. And in some cases such as this one will like to point a finger at all the wrong diagnosis'

I was quite honest in my answer when i said it'll be the first one i've ever heard of. And judging by some other replies i think some others will have learned from this too.

Now i think richard,.(can i call you dick?)you should insert the Spare parts only cluster into one of your many body orifices for safe keeping in case you need it for parts in the future,

Hi Nick,

Ouch. I can assure you that all offered help was sincerely appreciated, and no finger-pointing was intended.

As for Richard, I think he's likely an innocent bystander. Nonetheless, if you send me his address, I'll mail the donor cluster to him.

Cheers,

David