1964 Riviera ST400 Speedometer driven gear

[RiVin]

I have a 1964 Riviera that I've been working on for 8 months.  It's all original matching numbers, 70k miles.  I'm down to minutia now.  This forum has helped me numerous times on varied topics.  But not this time....so far.  I'm trying to find out what the tooth count is on the original factory ST400 speedometer driven gear.  I have the 3.07 rear axle ratio and run original size tires being 7.10 x 15.  I have the 1964 chassis service manual, but it does not cover this, at least not that I could find.  Please help.  If possible I would appreciate seeing a chart out of a parts manual that shows what driven gears go with certain rear axle ratios and tire sizes.  But just knowing what came originally in this Riv would get me moving in the right direction. Thanks so much in advance.

[1965rivgs]

  The information you are seeking appears in the factory service bulletins. I don't have access to them at this time but will in early October. If you havnt found the info by then send me a PM.

Tom

Edited September 10, 2023 by 1965rivgs (see edit history)

[BulldogDriver]

Use this to calculate what should be in the car. You should be able to pull the bullet out after removing the cable and count the number of teeth on the shaft. It should match the color in the chart.

https://www.tciauto.com/speedometer-gear-calculator#2004R

 

EBay lists these parts all the time. FYI make sure the shaft is the same length or a little longer as the one you pull out. Also make sure the seal is seated all the way in the bullet if you replace the seal/bullet. 
 

Ray

Edited September 10, 2023 by BulldogDriver (see edit history)

[NTX5467]

3 hours ago, 1965rivgs said:

  The information you are seeking appears in the factory service bulletins. I don't have access to them at this time but will in early October. If you havnt found the info by then send me a PM.

Tom

All the Buick parts manual has is a listing of the drive and driven gear part numbers and colors.  No listing of axle ratios and tire sizes, in a chart as other divisions have done.  I'll look in another resource . . .

 

NTX5467

[BulldogDriver]

The OP has the axle ratio and tire size. The calculator will give them the drive and driven gears for the transmission. Pulling the bullet will confirm that the correct driven gear was installed and they can replace or adjust as needed. Can’t see that anything’s needed past that. I don’t think Buick was running their own set of gears unique to their application. 
 

From my observation the parts manual is lacking in the coverage of Gen 1 Rivieras. 
 

Ray

[NTX5467]

14 minutes ago, BulldogDriver said:

The OP has the axle ratio and tire size. The calculator will give them the drive and driven gears for the transmission. Pulling the bullet will confirm that the correct driven gear was installed and they can replace or adjust as needed. Can’t see that anything’s needed past that. I don’t think Buick was running their own set of gears unique to their application. 
 

From my observation the parts manual is lacking in the coverage of Gen 1 Rivieras. 
 

Ray

Yes, it seems Buick had a "bad habit" of putting specific model information into "Service Bulletins", even in the later years.  As the first year Regal GN turbo cars and the "end of the series", last gen Riviera convertible top weatherstrips.  Something in the corporate DNA?

 

I suspected the speedometer gear specs might be in the AMA Specifications, but they were not.  Yet, that tire size and axle ratio combination could be found on other GM cars with the THM400 automatic, no doubt, if the OP was only wanting to verify what was in his car might be accurate.  Otherwise, match the color codes (which should also relate to the tooth-counts) on each gear.

 

FWIW, the Buick parts book only lists two drive gears, a 17 tooth and 19 tooth, with no mention of axle ratio or tire size.  Driven gears, there are many, with only the color and tooth counts listed, with no mention of axle ratio or tire size, in the transmission parts section related for the THM400 automatic transmission.

 

Take care,

NTX5467

[EmTee]

Is there even a problem?  I would wait until the car can be driven and then check the odometer accuracy.  There may be nothing to fix.  Worst case is to make a speedometer drive gear change, which takes about 10 minutes to accomplish once under the car...

[RivNut]

Use your GPS for a very accurate speedometer.  On one of my cars the speedometer reads fast, the other one is slow.  The GPS is always accurate.  The newer aftermarket speedometers are all GPS based, no need for a cable other than your odometer.

[RiVin]

Ah, yes!  Of course you are correct.  But I have an opportunity here.  My speedometer is reading 7% slower than actual.  I also have a fairly decent leak at the ST400 speedometer gear housing.  So, I figured that if I must pull out the housing to replace the seals, why not replace the driven gear at the same time.  I'd like to purchase the new gear before tearing into it. It looks like this may not be possible.  Thanks for the advice.

[RivNut]

58 minutes ago, RiVin said:

Ah, yes!  Of course you are correct.  But I have an opportunity here.  My speedometer is reading 7% slower than actual.  I also have a fairly decent leak at the ST400 speedometer gear housing.  So, I figured that if I must pull out the housing to replace the seals, why not replace the driven gear at the same time.  I'd like to purchase the new gear before tearing into it. It looks like this may not be possible.  Thanks for the advice.

How do your present tires compare to the original equipment tires?  Is the 7% consistent from 10 mph to 80 mph, or only at a given speed?

[RiVin]

The original tires were 7.10 x 15 as I'm sure you know. It was 27.8" diameter.  I am now running 215/75R15.  Diameter is 27.7".  First thing to notice is that the speed inaccuracy is in the wrong direction.  The 7% is consistent from 30 to 60mph.  I haven't run it past there yet.  Thanks

[NTX5467]

Should be an o-ring around the basic speedometer gear "plug" outside.  Then a smaller "lip seal" which seals the speedometer gear shaft.  As soft and flimsy as that gear shaft seal is, it's a wonder it seals at all for years.

 

According to the 1955 Buick AMA Specs, a 7.10-15 tire is rated at 750revs/mile at 30mph.  Currently, a P225/70R-15 Radial TA is rated at about 754 revs/mile @45mph.  Which puts a P215/75R-15 in the same ballpark.  GPS can measure distance, but considering how different some of the Google maps locations are for a physical address, distance might be a bit suspect.  I'd feel better about using the Interstate mile markers for a distance check, personally.

 

Enjoy!

NTX5467

[RivNut]

7.60 x 15 was an optional size for the Ribiera, white sidewalls or black sidewalls.  The replacement size in radial tires is 225/75R15.  Would that bigger tire go towards correcting the speedometer error or make it worse?

[RivNut]

Road speed should have no bearing on revolutions per mile.  See what this tire calculator shows in difference between the 215 and 225.

 

https://tiresize.com/calculator/

[NTX5467]

I use the revs/mile spec for comparison to modern sizes of tires.  The 7.60x15 size shows 738 revs/mile, which equates to a modern P225/75R-15 size.  This is important as it relates to the number of revolutions/mile the speedometer cable makes, which relates to odometer calibration and accuracy.  The reason I'm concerned about odometer accuracy is that the odometer is geared to the speedometer's rear speed cup.  The speed reading (which results) is done magnetically by the bar magnet in the rear speed cup.  That means that such magnetism can be varied with what I term a "zapper" so the speed reads as accurately as it can . . . a purely cut and try situation. 

 

Police-certified speedometers (in genuine law enforcement vehicles from the factory) are accurate to + and - 1mph across the complete speed range.  Due to a "close-tolerance" clockspring and calibration of the magnetism, as I understand it.  The reason I know this is that I ordered two police speedometer heads for my '80 Chrysler Newport.  Each came with a paper detailing the accuracy of the unit.  I didn't want to look at an 85mph speedometer all of the time, plus they were inexpensive.

 

One other reason I'm concerned about the accuracy of the odometer is that that reading is what fuel economy is run off of.  Taking some runs on the Interstate at a constant 60mph, checking the milemarkers after 5 miles (although 10 miles is easier to do the calculations from) checks the speed accuracy.  60 secs/mile at 60mph velocity.

 

From what I've seen, all of the tire size comparators operate on "exact" measurements.  In an article on the then-new Appha-numeric tire sizes of 1968, where it gave gov't specs for such, it noted that a 7% variance from the minimum specs listed in the article.  In looking at the tire specs charts in TireRack.com, no two tire brands in the same size have an exactly the same number of revs/mile specs.  Many off by a few revs, which is good for a flexible product, but none exactly the same, by observation.

 

Y'all enjoy!

NTX5467

[RiVin]

Okay!  Tire size calculator says:  215 = 27.7" dia.  225 = 28.3" dia.  Same results as when I did it.  And the 225 would make the problem worse.  I would be traveling more ground with the 225.  I'm not sure what your definition of road speed is, but if it means "the speed of the car itself at a given speedometer reading", then I would have to disagree with you.  If I was to put put the 225 tire on and hold the speedometer at the same speed as the 215, then the actual road speed would have to increase since the car will be traveling %1.02 more distance per each revolution of the wheel and the driven gear. So the revolutions per mile will decrease and therefore show a lower speedometer reading if the road speed stayed the same.

[RiVin]

NTX5467.  I like what you put down there.  Makes sense.  Thanks

[BulldogDriver]

The odometer which is what you would be basing the MPG number should be accurate as long as the tires and differential gear ratios are correct for the what the odometer was setup for. Speedometer readings can be off from wear and inherent error’s without effecting that accuracy. Using mile markers would be efficient for checking that.

 

In my experience with GPS it is the most accurate way to gauge true speed and distance traveled. Even if the signal is off by whatever amount, usually no more than 30-50 feet, it is effecting the accuracy of pin pointing exactly where you are at any given time and not so much as speed/distance. Just use a known accurate unit.

 

Ray
 

 

[RiVin]

I just recalculated the difference between the 215 and the 225 tire.  The difference in diameter is actually 2.2% and in circumference would be the same 2.2%. Thanks

[1965rivgs]

6 hours ago, NTX5467 said:

All the Buick parts manual has is a listing of the drive and driven gear part numbers and colors.  No listing of axle ratios and tire sizes, in a chart as other divisions have done.  I'll look in another resource . . .

 

NTX5467

  NOT parts manuals...SERVICE BULLETINS

Tom

[RivNut]

1 hour ago, NTX5467 said:

I use the revs/mile spec for comparison to modern sizes of tires.  The 7.60x15 size shows 738 revs/mile, which equates to a modern P225/75R-15 size.  This is important as it relates to the number of revolutions/mile the speedometer cable makes, which relates to odometer calibration and accuracy.  The reason I'm concerned about odometer accuracy is that the odometer is geared to the speedometer's rear speed cup.  The speed reading (which results) is done magnetically by the bar magnet in the rear speed cup.  That means that such magnetism can be varied with what I term a "zapper" so the speed reads as accurately as it can . . . a purely cut and try situation. 

 

Police-certified speedometers (in genuine law enforcement vehicles from the factory) are accurate to + and - 1mph across the complete speed range.  Due to a "close-tolerance" clockspring and calibration of the magnetism, as I understand it.  The reason I know this is that I ordered two police speedometer heads for my '80 Chrysler Newport.  Each came with a paper detailing the accuracy of the unit.  I didn't want to look at an 85mph speedometer all of the time, plus they were inexpensive.

 

One other reason I'm concerned about the accuracy of the odometer is that that reading is what fuel economy is run off of.  Taking some runs on the Interstate at a constant 60mph, checking the milemarkers after 5 miles (although 10 miles is easier to do the calculations from) checks the speed accuracy.  60 secs/mile at 60mph velocity.

 

From what I've seen, all of the tire size comparators operate on "exact" measurements.  In an article on the then-new Appha-numeric tire sizes of 1968, where it gave gov't specs for such, it noted that a 7% variance from the minimum specs listed in the article.  In looking at the tire specs charts in TireRack.com, no two tire brands in the same size have an exactly the same number of revs/mile specs.  Many off by a few revs, which is good for a flexible product, but none exactly the same, by observation.

 

Y'all enjoy!

NTX5467

If you feel that you’re being a nuisance running at only 60 mph on an interstate highway, you can divide the number of seconds it takes you to cover the measured mile into 3600. (The number of seconds in an hour.) That will give you a very accurate speed in mph. If it takes you 48 seconds to go the measured mile, then your speed is 75 mph.  Conversely, 48 mph will take you 75 seconds.  40 seconds would be 90 mph and 90 seconds would be 40 mph.  60 for 60 both ways. You’ll need a good stop watch (your cell phone’s clock feature and something like your A-pillar for a marker.) 

[NTX5467]

As I mentioned, Buick has been known to put specific part numbers into TSBs (i.e., Service Bulletins) rather than put then in the parts book for all to see.  I found this out when we got a Regal GN into our shop and needed some GN-specific parts.  We got the Buick franchise a year too late, it seems, for us to have gotten that particular TSB.

 

Needed a particular body-to-convertible top rubber seal for a last-gen Riviera convertible executive car our Finance Manager had wangled to buy when it came out of service.  After looking everywhere, it was revealed that that part number was in a TSB, which I had to call a particular person in Buick to get the part number on.  Not fun tracking these things down.  If it had been a Chevy, it would have been in the book.  Just as the speedometer gear charts are.

 

When GM went to Internet-based service information, it seemed a bit unfavorable as there were no hard copies or CDs to consult as archived information.  As things have progressed in 30 years, it is quite nice, provided things can be kept up with, IF one wants to.  One bad thing is that labor times were in a similar format, during a time when techs were paying attention to changes being made, discovering that their times had been cut, suddenly and without notice.

 

As to "getting in the way" on the Interstate, that now happens at the posted speed limit.  Yet there are still lots of trucks which run the normal 65mph or slower.  For the first 5 seconds either side of 60 seconds = 60mph, the speed relationship to time is pretty accurate, as 55 seconds is 65mph and 65 seconds is 55mph.  Past that 5 second mark, the relationship moves too far away without having a calculator or speed chart.

 

The other things is that after, at least in DFW, you get farther away from the metro area, after traffic settles down into a cruise mode, the terrain gets a bit flatter and easier to maintain a constant speed on for 5-10 miles.  When I was starting to do speed calibration and odometer checks, that's what I did.  Now, with GPS, results can be more instantaneous.

 

Speedometers will also have a bit of inaccuracy built into them.  Tires wear and will decrease circumference as they go from 11/32" new tread depth to 2/32" legal tread depth, for example.  As the speed indicator is run by magnetism, the ACSpark Plug specs for speed readings were a bit wide, to me, at something like "5mph slow to 2mph fast" on the speedometer cable-drive "speed checker" tool.

 

In about 1962, Oldsmobile had "an idea" to get around the speedometer gears in the transmission, taking the cable drive off of the lh front wheel speed, mechanically.  That didn't last long.  Back then, I figured that using a signal generator (as in a driven electric generator) to generate voltage to run the speedometer electrically.  Then just using resistors to fine-tune the speed readings.  I think some of the newer aftermarket digital gauge clusters use something like that.

 

End result is that for all of the variables, we CAN have some pretty accurate speedometers and odometers.  Just need to verify how accurate by various means.

 

Enjoy!

NTX5467

[EmTee]

2 hours ago, RiVin said:

The difference in diameter is actually 2.2% and in circumference would be the same 2.2%.

So, it sounds like 2% error due to tire size and that leaves 5% to the speedometer/odometer and the drive gear in the transmission.  5% error for a 1960s speedometer doesn't sound terrible to me.  That's 57 ~ 63 mph at an indicated 60 mph.  A speedometer drive gear change will probably make it worse...

[RivNut]

When I bought my 63 back in 1982, the previous owner had installed a 3.91 rear gear to avoid lugging the engine when he pulled his Airstream trailer.  It took me a number years (before the internet) to find a 3.07 rear gear (Hemmings classifieds.) The previous owner made a plastic template that just set in the speedometer bezel that showed the correct speed of the car - it worked well but who knows how long he and I drove it that way and what my actual odometer reading might be.  But driving a ‘63 with a Dynaflow and 3.91 gears did nothing for your gas mileage.  I didn’t have a tach so who knows what RPMs the engine was turning at 70 - 75 mph. 

[gungeey]

My rear tires are worn to 7/32s and I run them at 28lbs cold. My speedometer is spot on.

[XframeFX]

I accidently wiped the calibration of my Speedo messing with that hairspring for Electro-Cruise. My needle nose pliers were magnetized. Keep magnets away!

 

This Link helped me restore calibration of 60 MPH @ 1000 RPM CCW with my cordless drill in reverse.

The magnetic bar rotor inside the copper cup needs to be saturated with a simple fixture, nothing more.

 

https://www.corvetteforum.com/forums/c3-tech-performance/4100362-tech-tip-how-to-re-magnetize-a-speedometer.html

 

Now, if I could only get the Speedo cable to co-operate at low speeds. I lubed and re-lubed, changed the core, kept ferrules on both ends loose, nothing. It still jumps with noise.

Typical Riviera, have to address the same problem 3X. However, still not remedied.😞

Dakota Digital - LOL!

[telriv]

ALL forget.

A bias ply tire is flatter while it rotates as opposed to a radial tire which appears to be low on air pressure when sitting still.

Maybe the radial will "expand" at speed???

 

Tom T.

[RiVin]

After reading and thinking about all of your suggestions, I believe that I have a solid plan.  Ed, since you planted that seed about how you run a 225/70R15 rather than the 225/75R15 or 215/75R15, and I read one of your old posts that said that you rub a chassis rivet with the front tires when cornering hard.  So, I looked at the diameter of the 225/70R15 and it is 27.4" while my current 215/75R15 is 27.7".  This is going in the correct direction for my speedometer problem correction.  So, I'm going with that size tire.  It will give my Riv a lower more bulldog stance for sure, as well as more rubber on the road.  And I am ordering a full set of 4 wheel spacers, 1/4" at a 5 on 5 bolt circle. This will help with the rivet rub and give Riv a wider track and improve the bulldog stance even more.  As for the speedometer gear.......I already ordered the seal kit.  That's all that is going to happen there.  But I will observe and count the teeth on both gears and report back here on how it all worked out.  I feel very confident about it, but if you see a flaw, please let me know soon.  I'm going to wait 1 week before ordering the new tires and spacers.  Thanks you all for all of the advice.

[EmTee]

My only comment is that my '67 came with the later road wheels and 1/4" spacers.  I wound-up replacing them with a set of original '67 wheels.  I wasn't happy with the amount of lug nut thread engagement remaining beyond the spacer.  In the end, if you have a similar concern, you could always install longer studs to restore the lost thread depth.

[XframeFX]

On 9/10/2023 at 6:55 AM, RiVin said:

I'm trying to find out what the tooth count is on the original factory ST400 speedometer driven gear.

A little late but, I stumbled upon this while searching for something else.

Some of my "BJ" ST400 research with the 18 tooth Drive Gear and 3.23 final drive. I imagine you would need a 40T Nylon Driven Gear.  Your 3.07 final drive is highlighted in yellow, left Table, mine in right Table:

[RiVin]

Hey John B.,  Thanks, this is awesome info and right on time.  I just received the speedometer trans leak seal kit. So, I'll by replacing the seals by the end of this week and I'll report back what gears that I find.  Also, I fully agree and expect to find the 40 tooth driven gear by my calculations.  All the best.

[RiVin]

Well gentlemen, I changed the seals in the trans speedometer driven gear housing.  All went well.  Interestingly enough, the driven gear is a white plastic 43 tooth gear.  The drive gear was a steel 18 tooth gear.  If you check out the charts that XframeFX provided, you will see that this does not coincide with what was predicted.  So, now I am wondering if there might have been a 3.23 or other axle in there from the factory.  Any thoughts, please lay them on me.  At least I know now that there is the 43 tooth driven gear in there and if I want to correct for an inaccurate speedo in the future, I'll at least have somewhere to start.  Thanks

[RivNut]

The ratio, if it has not been changed, is stamped into the bottom of the pumpkin.  

[RivNut]

Here are a couple of pictures of the rear end of my 64 after cleaning it up for painting.

This one shows where to find the numbers

 

 

This one shows a 3 over an 07 indicating a 3.07 rear gear.

 

 

Hope this helps.

 

Ed

[RiVin]

Thanks Ed. I already did this weeks ago and it looks exactly like yours.  3.07.  And the revolutions of the drive shaft to rear wheels support that as what's in there.  The problem is that a 43 tooth should not be with a 3.07 axle.  Somebody must have changed one or the other.  The TCI speedo gear calculator online says that I should have a 40 tooth in there, just like XframeFX calculated.  And I agree with his calculation.  If you remember, I said I have approx a 7% discrepancy.  Guess what (43-40)/40 equals.  Approx 7%.  Thanks

[RivNut]

I’d guess that some previous owner had to replace the gears and only a 43 tooth could be found at the time.

[EmTee]

Could the transmission have been swapped?

[RiVin]

I am pretty sure that the transmission has never been changed.  If it has, someone did a ST400/ST400 exact swap.  But there are no signs of that whatsoever.  There is a good chance that Ed is correct and someone put larger tires on the rear and needed to correct for the speedometer inaccuracy.  There is no signs of a rear axle swap either.  Thanks

[RivNut]

I just checked the Team Buick reference page and for 1964, contrary to 1965 which had 3, there was only one ST400 therefore there should only be the one driving gear ratio.

[XframeFX]

22 hours ago, RiVin said:

3.07.  And the revolutions of the drive shaft to rear wheels support that as what's in there.  The problem is that a 43 tooth should not be with a 3.07 axle.  Somebody must have changed one or the

So, if you have positraction, you could do this. Do several axle rotations and divide pinion rotations by that amount for accuracy. 3.07 and 3.23 are too close for just one rotation.

Most G1 Rivieras had a 3.23:1 final drive.

Also, I eluded to Speedo calibration earlier. Should see 60 MPH @ 1000 RPM CCW.

 

 

 

 

 

Edited September 20, 2023 by XframeFX (see edit history)