Any one have experience in their speedometer being inaccurate? I have a '66 Newport with original engine and transmission. I downloaded a speedometer app and if my speedometer reads 30 mph the app shows I'm going a few miles an hour slower. The faster I go the larger the gap. If the app says I'm going 55 the speedometer says 65. My buddy has a '67 Mercury Comet and he says his does the same. Were all speedometers this inaccurate back then?
Inaccurate speedometer
- Thread starter Nadine66
- Start date
MrMoparCHP
Old Man with a Hat
Change in tires size is likely the problem, you can change the pinion in the transmission. There are different way to calculate the change.
Alan
Alan
Several reasons. First is speedometers were always built with a lean towards reading faster. Then you have tire changes were we've gone through different sizing standards, combined with changing from bias to bias belted to radials. Somewhere the tire size gets rounded off and may be a fraction of an inch smaller.
Then the speedometer itself wears and usually they read slower as it ages.
So, if it's enough to be worth the effort, you'll have to look at what size the speedo pinion is in the trans now and change it accordingly. The problem with that is they are really all compromises to gearing and tire size, so you can easily make it worse.
There are also speedometer correction adaptors that slow down or speed up the cable to correct the speedo. Not cheap with new cables needed, but they do work.
Or you can just live with it... Figure out what it really is at 30MPH etc. and drive accordingly.
The speedometer in my '67 Newport has an odometer that is dead accurate, yet at 60mph road speed, it reads 10% too high. Tire size is P245/70R-14, which is the same "revs/mile" as the OEM 8.55x14 tires it came with.
By comparison, our '66 Newport which came with 8.55x14 tires was always dead accurate as to speed and distance, even with the later H78x14 BFG bias-belted tires.
The Delco speedometer specs were usually, at 50mph, from indicating 3mph fast to 2mph slow. No specs for odometer readings.
As mentioned, tire sizes have gone through a few size designation spec differences over the past 60 years. Yet all are similar to the same size of the original tires. The issue with radial tires can cloud later years' issues a bit, due to their sidewall deflection, but NOT specifically their "revs/mile" numbers. "Revs/mile" at 45mph is the measuring standard which relates to the tire rolling diameter, and how many revs/mile the speedometer cable makes at the speedometer head.
The reason for the "at 45mph" speed is that older, bias-ply tires would expand with higher rotational speeds. What would be accurate for 30mph would be too slow for 50mph, for example. In a time well before bias-ply tires had "belts" to better-contain the tire's diameter from enlarging due to centrifugal force. Bias-belted tires and radial tires had these "belts".
In the 1990s, the less-expensive GM cars used tire revs to gauge "low tires" without any electronics other than the ABS wheel speed sensors. If the side-to-side differences were too great, the "Low Tire Pressure" warning light would come on, indicating the tire with the lesser rolling diameter, i.e. "low pressure". Thjat was also their "budget" traction control system, too.
The odometer is geared directly from the speedometer cable at the back of the speedometer head. At the rear speed cup. Inside that speed cup is a bar magnet., which makes the front speed cup turn as it does, though magnetism. The clockspring attached to the front speed cup's needle resist's that turning motion. The stronger the magnetism, the more the front cup wants to follow the speed of the rear cup, and vice versa.
So, the odometer calibration follows the speedo cable turns, which relates to the tire diameter and rear axle ratio. Getting the odometer calibration correct is the FIRST step in total unit calibration.
Then, for the speed readings, the speedometer shops had a "zapper" gun which could vary the magnetism in the bar magnet. An inexact science, by observation, as a "zap" could decrease or increase the speed reeding, for the same input speed to the speedometer head by the testing machine. So, a cut and try procedure with the "best" result being variable. Whis is where the Delco specs come in.
In the case of "Certified" speedometers in genuine law enforcement vehicles, they were manufacturer/OEM certified to be within ONE mph, plus or minus, of the indicated speed. Obviously, "close match" production resulted in this, BUT that certification was good between about 30 degrees F and 90 degrees F ambient temp in the car the speedometer was installed into. Which means a special clockspring in the mix, too. I purchased two certified speedometers for my '80 Newport (to get rid of the standard 85mph speedometer) and each one came with the "Calibration Card" in the box.
What the "Certified" speedometer did was allow an officer to pace an offender to observe the offender's actual road speed AND have an accurate measure of speed in the process. No need for radar or electronics. Just a careful pacing by the officer.
Things were different back then and that's how it all worked. Before a time of more exact speed measurements.
CBODY67
By comparison, our '66 Newport which came with 8.55x14 tires was always dead accurate as to speed and distance, even with the later H78x14 BFG bias-belted tires.
The Delco speedometer specs were usually, at 50mph, from indicating 3mph fast to 2mph slow. No specs for odometer readings.
As mentioned, tire sizes have gone through a few size designation spec differences over the past 60 years. Yet all are similar to the same size of the original tires. The issue with radial tires can cloud later years' issues a bit, due to their sidewall deflection, but NOT specifically their "revs/mile" numbers. "Revs/mile" at 45mph is the measuring standard which relates to the tire rolling diameter, and how many revs/mile the speedometer cable makes at the speedometer head.
The reason for the "at 45mph" speed is that older, bias-ply tires would expand with higher rotational speeds. What would be accurate for 30mph would be too slow for 50mph, for example. In a time well before bias-ply tires had "belts" to better-contain the tire's diameter from enlarging due to centrifugal force. Bias-belted tires and radial tires had these "belts".
In the 1990s, the less-expensive GM cars used tire revs to gauge "low tires" without any electronics other than the ABS wheel speed sensors. If the side-to-side differences were too great, the "Low Tire Pressure" warning light would come on, indicating the tire with the lesser rolling diameter, i.e. "low pressure". Thjat was also their "budget" traction control system, too.
The odometer is geared directly from the speedometer cable at the back of the speedometer head. At the rear speed cup. Inside that speed cup is a bar magnet., which makes the front speed cup turn as it does, though magnetism. The clockspring attached to the front speed cup's needle resist's that turning motion. The stronger the magnetism, the more the front cup wants to follow the speed of the rear cup, and vice versa.
So, the odometer calibration follows the speedo cable turns, which relates to the tire diameter and rear axle ratio. Getting the odometer calibration correct is the FIRST step in total unit calibration.
Then, for the speed readings, the speedometer shops had a "zapper" gun which could vary the magnetism in the bar magnet. An inexact science, by observation, as a "zap" could decrease or increase the speed reeding, for the same input speed to the speedometer head by the testing machine. So, a cut and try procedure with the "best" result being variable. Whis is where the Delco specs come in.
In the case of "Certified" speedometers in genuine law enforcement vehicles, they were manufacturer/OEM certified to be within ONE mph, plus or minus, of the indicated speed. Obviously, "close match" production resulted in this, BUT that certification was good between about 30 degrees F and 90 degrees F ambient temp in the car the speedometer was installed into. Which means a special clockspring in the mix, too. I purchased two certified speedometers for my '80 Newport (to get rid of the standard 85mph speedometer) and each one came with the "Calibration Card" in the box.
What the "Certified" speedometer did was allow an officer to pace an offender to observe the offender's actual road speed AND have an accurate measure of speed in the process. No need for radar or electronics. Just a careful pacing by the officer.
Things were different back then and that's how it all worked. Before a time of more exact speed measurements.
CBODY67
Last edited:
Where on earth did you get P245/70R-14 tires?
BFG Advantage T/As, whitewall, which I mounted on 14x6.5 Chrysler station wagon wheels (1965 vintage). I found the tires from the old Coker Tire Company, dealing with a younger Corky Coker, in about 1981. Other than the size, the attraction was a 6-ply rated polyester cord body, no steel belts. At that time, steel-belted radials were deemed to transfer too much harshness into the Chrysler UniBody, resulting in a "louder" ride experience. So I went with that.
As it turned out, they were very "power hungry", as to rolling resistance. After the car sat in the driveway for a week, sometimes I'd push it around a bit, but even as strong as I was back then, it did not move easily. By comparison, I could move my '77 Camaro with P225/70R-15 Radial T/As much easier. Which explained why those radial tires caused a slight decrease in fuel economy compared to the prior Kelly-Springfield P225x75R-14 whitewall fabric belt radials that had been on there. BUT as to cornering, the BFGs were very surprisingly great. Still on the car as it sits under the carport.
And yes, that P245/70R-14 size is an exact match for the old H78x14 tire size, as to revs/mile.
CBODY67
As it turned out, they were very "power hungry", as to rolling resistance. After the car sat in the driveway for a week, sometimes I'd push it around a bit, but even as strong as I was back then, it did not move easily. By comparison, I could move my '77 Camaro with P225/70R-15 Radial T/As much easier. Which explained why those radial tires caused a slight decrease in fuel economy compared to the prior Kelly-Springfield P225x75R-14 whitewall fabric belt radials that had been on there. BUT as to cornering, the BFGs were very surprisingly great. Still on the car as it sits under the carport.
And yes, that P245/70R-14 size is an exact match for the old H78x14 tire size, as to revs/mile.
CBODY67
Blusmbl
Active Member
x2. I went from 4.30 to 4.10 gears in my car with the same tire size. The speedo was right on with the 4.30 gears and 30" tires, and I *incorrectly* assumed it had the correct speedo gear in it for that tire size and gear, so I ordered the appropriate one for the 4.10's. Put it in and the speedometer was off by something like 7%. Turns out the original speedo gear was not the correct one for 4.30's, so I took that number and adjusted it by the % error, and purchased that gear instead. It then got me to where the speedo is accurate again.
I think the early transmissions also used a different tooth count, but mine wasn't off by that amount either.
Early, '62-'65, transmissions used a different gear and there isn't as many gears available, so you have to do some compromising. I believe there are only 5 gears and Chrysler used an extra adaptor to get a few combinations. The later transmissions used something like 15 different gears. The drive gear on the output shaft is 8 teeth for early and 13 teeth for later.
The adaptor ratio is .75:1.
I recently found a couple formulas that can help.
For '62-'65, the formula is (8 x rear end ratio x (20168/tire dia) /1001) = number of teeth on speedo gear.
For '66 up, the formula is the same, just using 13 instead of 8. (13 x rear end ratio x (20168/tire dia) /1001) = number of teeth on speedo gear.
BTW, the 20168/tire dia is how many tire revolutions per mile.
I need to do a spread sheet that figures that out!
Last edited:
Interesting formulas! Remember, too, that the odometer is the item which needs to be correct, FIRST, then the speed SECOND.
The way I used to do any speedo gear changes was to look at how many revs/mile the driveshaft did, between the prior and later rear axle ratios or tire sizes. Then compare that to what the car originally had in it, as to drive/driven speedo gears. Working on the percentage of change of the driveshaft revs/mile, then I'd look for OEM speedo gear combinations to get that same percentage of change, such that the revs/mile of the speedometer cable were the same. Which then put things back to where they were originally, as to what the speedo indicated in miles and speed.
In the GM realm of things, Chevy used speedo ratio adapters to cover the holes left with the approx 8 tooth drive gears and matching driven gears. You can get an idea of the number of teeth on the driven gear by looking at how the driven gear plug the speedo cable attaches to is clocked in the tail shaft housing. The hole the driven gear resides is not in the exact middle of the housing, but offset to allow the different sized driven gears (related to the number of teeth on them) to be used. The GM ratio adapters could be less than 1:1 or more than 1:1. Some of the old speedo shops could custom-build them for particular applications, as to what gears were in them. Seeems like Stewart-Warner was an OEM supplier of them?
In looking at the various tire size comparison apps, understand that they are dealing with "as stated" or "theoretically exact" tire dimensions. When the actual dimensions of tires for sale can vary some from those theoretically correct numbers. In this case, revs/mile for radial tires.
The reason I mentioned "radial tires" is that their outer belts keep them from expanding with increasing speed, unlike the bial-ply tires that could "grow" with wheel speed. Ever notice the slicks on a top fuel dragster as they do their burnouts? Same thing, but probably not to that same degree.
Enjoy!
CBODY67
The way I used to do any speedo gear changes was to look at how many revs/mile the driveshaft did, between the prior and later rear axle ratios or tire sizes. Then compare that to what the car originally had in it, as to drive/driven speedo gears. Working on the percentage of change of the driveshaft revs/mile, then I'd look for OEM speedo gear combinations to get that same percentage of change, such that the revs/mile of the speedometer cable were the same. Which then put things back to where they were originally, as to what the speedo indicated in miles and speed.
In the GM realm of things, Chevy used speedo ratio adapters to cover the holes left with the approx 8 tooth drive gears and matching driven gears. You can get an idea of the number of teeth on the driven gear by looking at how the driven gear plug the speedo cable attaches to is clocked in the tail shaft housing. The hole the driven gear resides is not in the exact middle of the housing, but offset to allow the different sized driven gears (related to the number of teeth on them) to be used. The GM ratio adapters could be less than 1:1 or more than 1:1. Some of the old speedo shops could custom-build them for particular applications, as to what gears were in them. Seeems like Stewart-Warner was an OEM supplier of them?
In looking at the various tire size comparison apps, understand that they are dealing with "as stated" or "theoretically exact" tire dimensions. When the actual dimensions of tires for sale can vary some from those theoretically correct numbers. In this case, revs/mile for radial tires.
The reason I mentioned "radial tires" is that their outer belts keep them from expanding with increasing speed, unlike the bial-ply tires that could "grow" with wheel speed. Ever notice the slicks on a top fuel dragster as they do their burnouts? Same thing, but probably not to that same degree.
Enjoy!
CBODY67
67_300_Convertible
Member
I have a similar issue with my 300. The odometer and speedometer are both off, but by different percentages. I understand I can use the formula to figure out which speedo drive gear I need (guess I will get one of these: https://www.brewersperformance.com/...ts/speedometer/1966-up-speedometer-pinions-2/).
But how does one calibrate the odometer? If that info is in this thread I missed it. TIA
But how does one calibrate the odometer? If that info is in this thread I missed it. TIA
The odometer is directly connected to the drive cable, so there's no adjustment except to change the gear.
The speedometer, on the other hand, can be adjusted. The speedometer works by a spinning magnet and while I've never gotten that deep into it, it's a matter of adjusting the spring that works against the magnet. I don't see a good way of adjusting the speedo at home though.
This is a decent explanation of how a mechanical speedometer works.
The clockspring does offer resistance to the front speed cup, what the speedo needle is attached to, but it is non-adjustable. The police Certified speedometers have a different one that has a slightly different material that offers the same resistance from like 32* F to 100* F (inside the car, obviously) for a +/- variance from the actual road speed (with the spec tire size, of course) of ONE mph. The clockspring's other function is to move the needle back to zero, as speed drops.
Getting the revs/mile of the speedo cable is the FIRST calibration area as that is what EVERYTHING works off of.
It IS possible for both calibrations (speed and distance) to be off by differing amounts, as the speed indication comes from a indirect-mechanical situation. My '67 Newport CE23, with P245/70x14 whitewall BFG Advantage TAs and 3.23 is dead on as to the odometer, but reads 10% fast on the speed reading. Using the milemarker posts on Interstate highways for a 10mile stretch of flat road where it is easy to maintain a set speed. Similarly, you can clock speed with a wrist watch (I had a digital Casio) on the same milemarkers. 60mph = 60 seconds. 65mph = 55 seconds. 55mph = 65 seconds. In the pre-GPS and phone app days, that's how I did it AND it worked for me. 10 miles makes the numbers easier to see, although 5 miles can be used, too.
Of course, for best accuracy, a flat and straight road works best as to maintaining a very steady speed for a length of distance. No overpasses or underpasses, just "on the ground" roadways, with gentle turns. Make it a "time and distance road rally" situation, with exactness as the ultimate goal.
Enjoy!
CBODY67
Getting the revs/mile of the speedo cable is the FIRST calibration area as that is what EVERYTHING works off of.
It IS possible for both calibrations (speed and distance) to be off by differing amounts, as the speed indication comes from a indirect-mechanical situation. My '67 Newport CE23, with P245/70x14 whitewall BFG Advantage TAs and 3.23 is dead on as to the odometer, but reads 10% fast on the speed reading. Using the milemarker posts on Interstate highways for a 10mile stretch of flat road where it is easy to maintain a set speed. Similarly, you can clock speed with a wrist watch (I had a digital Casio) on the same milemarkers. 60mph = 60 seconds. 65mph = 55 seconds. 55mph = 65 seconds. In the pre-GPS and phone app days, that's how I did it AND it worked for me. 10 miles makes the numbers easier to see, although 5 miles can be used, too.
Of course, for best accuracy, a flat and straight road works best as to maintaining a very steady speed for a length of distance. No overpasses or underpasses, just "on the ground" roadways, with gentle turns. Make it a "time and distance road rally" situation, with exactness as the ultimate goal.
Enjoy!
CBODY67
Since I always want to make things work asap and am not afraid to be a cheap a.... h.ave anybody tried to open the speedo lens and bend the needle a little as close to the center as possible. The bended area would be invisible but you have to know what you are doing.
This of course after checking that all other things are in order first.
That's really not going to work.
First, the needle is just a really lightweight aluminum stamping that would fold up if you tried to bend it. In other words, you'd ruin the needle without much effort.
Next, and the real reason why this won't work. I have to put on my gauge calibration hat here. While I didn't calibrate speedometers, I did calibrate a lot of other stuff for a living for a lot of years.
Speedometers that aren't otherwise broken are going to off by percentage. As an example, lets' say your speedometer is off 20%. At 10MPH, this means your speedo is going to read 12MPH. At 20MPH, your speedo is going to read 24MPH. At 50MPH, your speedo is going to read 60MPH. At 100MPH, your speedo reads 120MPH.
So. let's say you tweak that needle down 10MPH. At 50MPH, it's correct, but at any other speed, it's off. At 20MPH, it's going to read 10MPH. At 100MPH, it's going to read 110MPH.
So, if you want to do it the cheap way, get an app for your phone that gives you GPS speed and using some masking tape, mark 60MPH and 30MPH or whatever speeds you usually drive at.
The better way is to figure out if you have the correct speedo gear in the transmission and fix that. If you think the speedo is off, you may need to send the speedo head out to some shop to repair and calibrate. Sorry, but sometimes old cars aren't cheap to fix small issues, but a speedo gear isn't all that expensive for a '66 or later trans. A little more $$ for a '62-65 and less choices too, but still not that crazy.
If you would bend the needle 10mph down in the case that the speedo shows 20 percent too high reading
it would mean a 2- 4 mph off reading between 30 and 70 mph real speed. Since I drive 99%
of my miles between those speeds it would be ok for me. But you are right , bending the needle requires
some skills and is not without risk. As I wrote everything else must be checked first such as axle ratio ,
tyre diameter and drive pinion . I would think twice before sending a speedometer to "electroshock therapy"..
it would mean a 2- 4 mph off reading between 30 and 70 mph real speed. Since I drive 99%
of my miles between those speeds it would be ok for me. But you are right , bending the needle requires
some skills and is not without risk. As I wrote everything else must be checked first such as axle ratio ,
tyre diameter and drive pinion . I would think twice before sending a speedometer to "electroshock therapy"..
3175375
Old Man with a Hat
Lots of good info here.
I’ll just add that the clock spring doesn’t like being dropped.
I did by accident and had it repaired…
I’ll just add that the clock spring doesn’t like being dropped.
I did by accident and had it repaired…
Similar threads
