correct speedometer drive gear for NP231

[jamespwsullivan]

Well, I already knew to look for a chart to find the correct number of teeth for my rear end ration and (new) diameter tires, but I am finding conflicting information I think.  This is a 1987 vintage AW-4 and NP231 combo and I am going to smaller tires. Lots of changes in process for this truck.  So this should be the long gear for mechanical speedometers, but I think the two charts I've found don't seem to agree.

One is the Novak chart found here:

https://www.novak-adapt.com/catalog/transfer-case-parts/np-speedometer-gears

and the other is from Quadratec:

https://www.quadratec.com/c/reference/jeep-speedometer-gear-tooth-chart

 

I am looking for the proper number of teeth for a 4.10 gearing and 32" tires.

Novak says 34 teeth and Quadratec says 36.

 

Am I missing something?

 

Pat

[gogmorgo]

My tendency would be to trust Novak over Quadratec, but the other option would be to figure it out on your own. Tire sizes are only nominal and there’s enough variability in effective tire diameters between tire manufacturers and the pressures different people are running that you’re only going to get “close enough” based off someone else’s chart anyhow. None of my Jeeps have even had accurate speedometers with the factory setup, and some of them have been out by as much as 10%.

Get the thing put together with the wheels and tires you’re going to run and pressures all set up. Then take it out on a reasonably straight, flat section of highway and use GPS to compare indicated to actual speed, at a few different speeds. Then do the math and figure out what correction factor you’ll need, and pick the new speedometer gear based on that. 

[derf]

On 9/30/2023 at 11:04 AM, jamespwsullivan said:

Well, I already knew to look for a chart to find the correct number of teeth for my rear end ration and (new) diameter tires, but I am finding conflicting information I think.  This is a 1987 vintage AW-4 and NP231 combo and I am going to smaller tires. Lots of changes in process for this truck.  So this should be the long gear for mechanical speedometers, but I think the two charts I've found don't seem to agree.

One is the Novak chart found here:

https://www.novak-adapt.com/catalog/transfer-case-parts/np-speedometer-gears

and the other is from Quadratec:

https://www.quadratec.com/c/reference/jeep-speedometer-gear-tooth-chart

 

I am looking for the proper number of teeth for a 4.10 gearing and 32" tires.

Novak says 34 teeth and Quadratec says 36.

 

Am I missing something?

 

Pat

Knowing exactly which tire you are using will help you figure out which gear you want.  Most 32" tires are not 32" in diameter and that affects which speedo gear you want.

[jamespwsullivan]

@derf Yeah I'm thinking it will be difficult to interpolate between 31" and 32"  The tires, when inflated to recommended pressure, are closer to 31.6"

@gogmorgo Your description of the process makes sense except I don't know the relationship between the correction factor and the number of teeth.  I looked back and found that I had ordered the current speedo gear (much larger tires) from Quadratec and it was within 2 mph at 55-60 actual. So, I may just go that way.  I'd be interested in the math you're referring to though.

Thanks to both for the responses.

[gogmorgo]

I was just working with this on my longbed, although I also just discovered I can’t find winter tires I want to run in a reasonable size anymore and that project has snowballed so far that speedometer calculations have been thrown entirely out the window for the time being
 

But for simple speedometer correction in my case, my speedometer is reading 10% higher than actual speed. So at 100km/h the speedometer indicates 110. The math is the same with MPH but my brain works in metric, as does my speedometer and gps, and speed limits. To confirm that it’s a speedometer ratio error and not a needle that’s slipped on the gauge I’ve also confirmed that when I’m at 50 per GPS my speedometer indicates 55, and at 80 I’m indicating 88.

So 110/100=1.1 That’s our correction ratio. 
So right now I have a 32 tooth speedometer in there. I think. I havent actually pulled it out to confirm, but that’s my memory from when I looked at it ~8 years ago. But assuming it’s a 32 tooth, I would multiply by the correction ratio. 32x1.1=35.2.

Then you’re going to have to pick a gear. In this case 35 may seem like the obvious choice.

 

 

And now my classic overthinking everything kicks in. 

But in some cases if you’re between gear tooth counts you might want to bias yourself towards the higher tooth count, to help account for tire wear. At least in my own logic.

I don’t have a good reference to support that because I haven’t found anywhere else they suggest it as a factor. Statically at least an off-road tire could lose 3% of diameter due to treadwear and still remain perfectly sericeable, which would slow your speedometer down by 3%. Dynamically I don’t know how much treadwear actually affects effective diameter with a loaded tire – I may just be overthinking it… it stands to reason that the tread blocks compress more than the rest of the rubber in a loaded tire so with less tread to squeeze down maybe the effective tire diameter doesn’t actually change all that much over the life of the tire ¯\_(ツ)_/¯ I simply don’t know. But the point I’m trying to make is if you do add a 3% reduction in tire diameter to our theoretical ideal speedometer gear tooth count of 35.2 above, you end up with 36.26, so maybe you’d want a 36-tooth speedometer gear instead of the more obvious 35.

Maybe it’s not a factor because you don’t plan on wearing the tires past 50% for better used resale value. Maybe you want to just keep things simple and go off a chart instead. 😅

[derf]

13 hours ago, jamespwsullivan said:

@derf Yeah I'm thinking it will be difficult to interpolate between 31" and 32"  The tires, when inflated to recommended pressure, are closer to 31.6"

@gogmorgo Your description of the process makes sense except I don't know the relationship between the correction factor and the number of teeth.  I looked back and found that I had ordered the current speedo gear (much larger tires) from Quadratec and it was within 2 mph at 55-60 actual. So, I may just go that way.  I'd be interested in the math you're referring to though.

Thanks to both for the responses.

Inflated diameter is inaccurate for figuring out the speedo gear.  Best bet is to find the published spec for the exact tire you are running.  It should have a revolutions per mile number.  That number is based on the actual rolling diameter of the tire on a vehicle on the road.  Plug it in to this equation:

 

Rev per mile × axle ratio × 13 ÷ 1000 = tooth count.

 

Round down to have your speedo read a hair higher than actual speed.

 

 

 

[gogmorgo]

11 hours ago, derf said:

Inflated diameter is inaccurate for figuring out the speedo gear.  Best bet is to find the published spec for the exact tire you are running.  It should have a revolutions per mile number.  That number is based on the actual rolling diameter of the tire on a vehicle on the road.

 

 

 

Except that number is also irrelevant to precise calculations because different loads and pressures will change the effective radius of the tire. 

Hence an actual calibration procedure, getting the vehicle set up the way it will normally go down the road, comparing to a known accurate speed readout and correcting from there. Everything else is still going to be just an estimate.

Which is totally fine. It’s what manufacturers have to do, which is probably why allowable error on a new vehicle is something like 7% in either direction, if memory serves. But if you’re chasing accuracy you won’t get it by estimating, no matter how close your estimate is going to be.

 

[derf]

1 hour ago, gogmorgo said:

Except that number is also irrelevant to precise calculations because different loads and pressures will change the effective radius of the tire. 

Hence an actual calibration procedure, getting the vehicle set up the way it will normally go down the road, comparing to a known accurate speed readout and correcting from there. Everything else is still going to be just an estimate.

Which is totally fine. It’s what manufacturers have to do, which is probably why allowable error on a new vehicle is something like 7% in either direction, if memory serves. But if you’re chasing accuracy you won’t get it by estimating, no matter how close your estimate is going to be.

 

Using revs per mile with that equation will make you as accurate as the manufacturer is.

 

Granted, that's not precise by any means but it's the best baseline to go with for general use.  You'll generally be within 1MPH of actual speed with that method.  That's assuming, of course, that your mechanical speedometer is anywhere in the vicinity of accurate.  And that's not always a given either.