Final drive ratio

  • For registered users, a Google custom search option has been added that will search across several dedicated GMC Motorhome sites including this forum, GMCMI.com, and bdub.net. To use it, either click the "G" icon at the bottom of the search dialog or select the "Google search" tab in the search results. This new feature is also useful when your search includes words that are shorter than the 4 letter minimum supported by the forum's built-in search engine.
The standard gear that came in our coaches was 31 teeth (403417) To correct it to a 3.42 ratio you would need a 34.5 tooth gear. Seeing it is difficult to make gears with 1/2 a tooth you would have to compromise to 34 or 35 teeth. that is why they use correctors. In the parts book it looks like the pinion drive gear is part of the rear hub assy. and not easily replaceable. In any case, even if it was, it would be a fairly big job to change that gear. Even if you could get a 34.5 tooth gear I doubt it would fit into the housing.
The speedometer gear is driven off the governor gear. That is a metal spiral gear that is part of the governor shaft. Its not hard to get to if you have OEM exhaust manifolds, but there is only one choice for that part.
I have not taken out the speedo gear, but it appears it is held with a bullet shaped bearing assembly which is held in place by a simple bracket. It should just pull out once that bracket is removed.

One thing I don't understand is if you get a gear with more teeth, isn't it going to be larger in diameter? And therefore the shaft of that gear will need to be held further away from the governor gear it is meshing with. But I don't see any reference to this...

You can see the spiral gear just under the governor house cap in this photo:

Here is a good photo of the speedo gear and the associated parts. Just one bolt to remove the bracket and it should slide right out.

My transmission leaks at the speedo cable so I likely need a new O-ring and shaft seal. And this photo has all the info I need in the comments!!
 

Attachments

  • TH425 speedometer drive.jpg
    TH425 speedometer drive.jpg
    250.7 KB · Views: 9
The speedometer gear is driven off the governor gear. That is a metal spiral gear that is part of the governor shaft. Its not hard to get to if you have OEM exhaust manifolds, but there is only one choice for that part.
I have not taken out the speedo gear, but it appears it is held with a bullet shaped bearing assembly which is held in place by a simple bracket. It should just pull out once that bracket is removed.

One thing I don't understand is if you get a gear with more teeth, isn't it going to be larger in diameter? And therefore the shaft of that gear will need to be held further away from the governor gear it is meshing with. But I don't see any reference to this...

You can see the spiral gear just under the governor house cap in this photo:

Here is a good photo of the speedo gear and the associated parts. Just one bolt to remove the bracket and it should slide right out.

My transmission leaks at the speedo cable so I likely need a new O-ring and shaft seal. And this photo has all the info I need in the comments!!
Often the speedometer gear housing is on a eccentric so it can be turned to adjust the gear engagement. Ours don't look to have that feature! I guess the Th425 was only really designed for the Tornado.
 
Often the speedometer gear housing is on a eccentric so it can be turned to adjust the gear engagement. Ours don't look to have that feature! I guess the Th425 was only really designed for the Tornado.
However the Oldsmobile Toronado also had a 2.78 ratio and the 3.07 was an option I believe. And the Eldorado had a 3.21. So they must of had different gears for those ratios. Unless the gears were designed to mesh at a different point along the side of the governor gear.

I use the GPS speed and compare it to the speedometer reading, but I'm kinda anal to have it read correctly.
 
  • Like
Reactions: eserv
However the Oldsmobile Toronado also had a 2.78 ratio and the 3.07 was an option I believe. And the Eldorado had a 3.21. So they must of had different gears for those ratios. Unless the gears were designed to mesh at a different point along the side of the governor gear.

I use the GPS speed and compare it to the speedometer reading, but I'm kinda anal to have it read correctly.
I would be bothered if my speedometer wasn't at lest close too! My wife followed me when we brought the motorhome home and we compared speeds with the van, they were pretty close. I'm fairly convinced my machine has a 3.42 FD but I'll do some more tests. Sure wish I could talk to the guy who owned it and loved it most before me but he died in 2013!
 
Holy Over-Literal Interpretation, Batman!

Of course you're not going to get your odometer to read over 100,000 miles LOL

What should absolutely happen though, is after 1000 turns of the cable, the odometer should read 1.0 miles more than it did. There is a _direct_ mechanical connection between the transmission and the odometer.

There is _not_ a direct mechanical connection between the cable and the speedo needle. There's some voodoo magic there they may need to be fixed/calibrated at a speedometer shop.

Adding a correction gearbox into the cable path to get the speedometer to read correctly at the expense of the odometer accuracy not fixing the speedometer calibration problem, just masking it.


Actually after 50 years of inaccurate mileage readings and the Odometer having only 5 significant digits (actual mileage can be off by multiples of 100,000 miles), the Odometer doesn't even make it on my list of concerns. If the Odometer is off by +/- 10% who cares? Mine is reading around 53,000 miles. I know it has rolled once because I have Invoices from the previous owners at around 80,000 miles. So to me it makes no difference if the real actual mileage is 153,000 or 145,000 or 160,000 or even 250,000. It is what it is and no one knows for sure.

Now if the speedometer is off by 10%, that could get you a nasty speeding ticket. Now that concerns me.
 
Holy Over-Literal Interpretation, Batman!

Of course you're not going to get your odometer to read over 100,000 miles LOL

What should absolutely happen though, is after 1000 turns of the cable, the odometer should read 1.0 miles more than it did. There is a _direct_ mechanical connection between the transmission and the odometer.

There is _not_ a direct mechanical connection between the cable and the speedo needle. There's some voodoo magic there they may need to be fixed/calibrated at a speedometer shop.

Adding a correction gearbox into the cable path to get the speedometer to read correctly at the expense of the odometer accuracy not fixing the speedometer calibration problem, just masking it.
All I'm saying is I don't really care if the odometer is off. I changed from a 3.07 to a 3.42 which means the output of the transmission is turning 1.11 times faster than OEM. When my GPS is reading 90km/hr, my speedo is reading approx 100Km/hr which is the same ratio. Fixing the speedo will bring the odometer back very close to what ever error there was from the factory (direct drive).

My point is whatever the odometer says is what it is. It's not going to affect the price I get when I sell it. The person will pay whatever he feels is fair base on the condition of the coach.
 
...

Adding a correction gearbox into the cable path to get the speedometer to read correctly at the expense of the odometer accuracy not fixing the speedometer calibration problem, just masking it.
I'm confused. If we change the final drive, then we are changing the distance traveled for each rotation of the transmission output shaft, and therefore the cable going to the odometer. I would think a correction box in the cable is the only way to correct that, short of revising the gearing arrangement in that drives the speedo cable inside the transmission. The final drive, like the tires, is downstream of the odometer gear.

But the correction should fix the odo, and then adjustments in the instrument should fix the speedo, if the two are not aligned in the instrument.

Back in the deeps of time, I developed a data-collection system to measure speed-time trajectories for traffic performance measurements using instrumented vehicles. We used a Masstech speed sensor that turned a clear disk with opaque markings on it through an optical sensor. The data-collection device sent it a steady signal and the optical sensor interrupted it, turning it into a pulse stream. It mounted inline with the speedometer cable, and directly converted the mechanical motion to an electrical signal. Even though speedometer cables were supposedly standardized to make odometers accurate, I still had to program a calibration routine to allow someone to dial it in. 2% of scale would have been plenty good but all it took was for someone to install non-standard tires and their car would be out of calibration.

Rick "like Bruce more interested in an accurate speedo than odo" Denney
 
I don't have a tach.
A tach is actually a handy thing to have and they don't cost much.

It can help you to determine when to shift down manually when climbing hills. Also tell you when the engine is over-revving and to shift up to Drive again.

There used to be a gauge called at "Tach-Vac" that showed both engine vacuum and RPM's. I believe it was designed for aircraft and was expensive, but it was handy to have in one gauge.
 
  • Like
Reactions: Seankidd
Adding a correction gearbox into the cable path to get the speedometer to read correctly at the expense of the odometer accuracy not fixing the speedometer calibration problem, just masking it.
I'm confused. If we change the final drive, then we are changing the distance traveled for each rotation of the transmission output shaft, and therefore the cable going to the odometer.

You're right, Rick, if the final drive is changed, than the odometer *should* need correction if all else is kept the same. Tire choice sometimes makes the final drive swap less straightforward though. The main point in Todd's argument is that we shouldn't aim to fix the speedometer by screwing up the odometer in the process.

I think what Todd is saying, is the odometer *should* be corrected first. I tend to agree with him. The odometer and the speedometer should theoretically agree with each other. If they don't agree, it's the speedometer that's wrong--not the odometer. Once the odometer is accurate, there are other measures that can be taken to correct the speedometer. The odometer should be the primary "source of truth" if you will.

But, if a guy ONLY cares about the speedometer and has no concern for the odometer, he can certainly correct the speedometer as needed. Dealer's choice I suppose.

I personally would like my odometer to be correct for purposes of long-term fuel tracking, knowing how long my trips were, maintenance logs, etc (or just for my anal-retentive peace of mind). Speed is more of a secondary concern. I'll generally go with the flow of truck traffic, or by the sound of the engine, or I'll just look at the GPS. I'm less concerned with speed accuracy in general. I can't imagine ever getting a speeding ticket with the motorhome.
 
BTW, I always wondered...
Is the speedometer just a D'arsonval movement type gauge with a built-in "generator"--so to speak? It seems like that'd be an easy way to make one. If that were the case, any disagreement between the odometer and the speedometer should point to a clockspring issue, or bearing/debris problems.

I reckon the rest of the gauges are air core motors, but I'd think those would be harder to drive linearly like the speedometer is.
 
BTW, I always wondered...
Is the speedometer just a D'arsonval movement type gauge with a built-in "generator"--so to speak? It seems like that'd be an easy way to make one. If that were the case, any disagreement between the odometer and the speedometer should point to a clockspring issue, or bearing/debris problems.

I reckon the rest of the gauges are air core motors, but I'd think those would be harder to drive linearly like the speedometer is.
How speedometer works
 
When I did my dashboard 7-8yrs ago, I got a GPS Speedo from Speedhut. https://speedhut.com/

I was able to personalize it and it has worked accurately. The only issues were loosing signal in tunnels, no big deal, just go with the traffic for that short distance. And sometimes in a VERY heavy rain storm it will loose signal. Those are rare instances considering the total milage it is in use. There is only a sensor that lays on the dashboard at the windshield and no connection to the transmission or drivetrain. Accurate no matter what FD, transmission chain drive or tire selection you are using. It only cares if it can see the satellites.

Also, while I was rebuilding my dash, we traveled in our MH for about 2yrs without a dashboard. I hooked up a bicycle speedometer with the trip mechanism magnet on the right hand driveshaft and programed that little speedo for the diameter of the MH tire. It was right on accurate when checked against radar speed checks, and the manufacture claimed accuracy to 99mph. Worked as a good transition till I got new gauges mounted on new dash. JWID
 
Is that really the ratio? Neat. I always wondered, but never took them apart and counted the gear teeth.

According to the interweb that's the ratio for GM. It must be true if it's on the interweb right? 1000 turns per mile does sound like something GM would do though.
 
Todd,
You can't credit GM with the odometer/speedometer calibration. That came out as an SAE standard some time in the 20s (iirc). Oh, and it isn't 1000/mile, it is 1001/mile to get away form gear count issues. (Like why final drives are never exact i.e 3.75, 4.10 ).
More than a few of the transmissions I have messed with have the speedo drive on an eccentric so the correct gear can be rotated to engage the tail shaft gear. I don't know why the 425 does not have this feature.
I know my speedo is off, but the GPS in my cell reports speed to a tenth of a unit so I really don't care.
Matt_C
 
How speedometer works
Ah, eddy currents. Even simpler than the coil-wound D'arsonval movements. Although he doesn't get into it, I'd imagine the mechanism is similar, with a clockspring and whatnot. Probably still only susceptible to spring and/or bearing/debris issues. There's not much to go wrong with a magnet and a conductive disc.
 
Last edited:
According to the interweb that's the ratio for GM. It must be true if it's on the interweb right? 1000 turns per mile does sound like something GM would do though.
The Masstech Speed Sensor I used when developing PC-Travel put out a pulse every 0.88 feet, or 6000 pulses per mile, at least as I dimly recall. I also recall that the optical disk in it had six interrupters, so that would correspond to a thousand rotations per mile. At the time, that was the standard for all mechanical speedometers on American cars that used the same threaded cable connections. That was the only speed sensor we needed to supply for American cars. But users still had to calibrate the distances for enough accuracy that the trajectories would line up with each other repeatably.

Rick "digging up really ancient memories there" Denney
 
I have a Vehicle Speed Sensor (VSS) which connects to my EBL engine control for the fuel injection. It outputs 2 pulses per speedometer cable revolution so 2000 pulses per mile (PPM). The EBL instructions says the standard is 2002 pulses per mile, so I'm not sure where the other 2 pulses come from.

I can change the PPM in the EBL programming so mine is set to 2224PPM which lines up the EBL computer display with the GPS and the change in final drive.

There is a speedometer calibration place just over a mile from me. Interestingly they don't deal with mechanical speedometers, only electronic ones and unit units built after 2000. I'm not sure why an electronic speedo would need calibration, its only counting pulses, unless the differential ratio or tires have been changed. Anyway, their main business is used vehicles (mostly pickup trucks) being exported to the USA. Every vehicle I've own in the past 20 years or more can select Miles or Kms from the dash.
 
When I did my dashboard 7-8yrs ago, I got a GPS Speedo from Speedhut. https://speedhut.com/

I was able to personalize it and it has worked accurately. The only issues were loosing signal in tunnels, no big deal, just go with the traffic for that short distance. And sometimes in a VERY heavy rain storm it will loose signal. Those are rare instances considering the total milage it is in use. There is only a sensor that lays on the dashboard at the windshield and no connection to the transmission or drivetrain. Accurate no matter what FD, transmission chain drive or tire selection you are using. It only cares if it can see the satellites.

Also, while I was rebuilding my dash, we traveled in our MH for about 2yrs without a dashboard. I hooked up a bicycle speedometer with the trip mechanism magnet on the right hand driveshaft and programed that little speedo for the diameter of the MH tire. It was right on accurate when checked against radar speed checks, and the manufacture claimed accuracy to 99mph. Worked as a good transition till I got new gauges mounted on new dash. JWID
GPS speedos are the way to go. If you don't car about longterm ODO readings, any old garage sale GPS will give you speed.
I also used a bicycle speedo back when I rallied. Most go stupid at 115 KPH but Sigma brand worked as fast as I ever went.