3.64 vs. 3.91 on your speedo face (pic of overlay)

[KFunk]

OK, I bought a speedometer with a W=1.393 (correct for E21 3.91 LSD, according to records compiled by Curt Ingraham), and found out the W=1.393 just has an overlay over a different face. The face underneath appears to be the same as the one on my old W=1.297 (for a 3.64 diff, according to Curt).

So on the left is the 1.393 speedo with the overlay removed, revealing the face similar to a 1.297. On the right is the overlay in my hand.

Maybe this is is helpful, I don't know. But if you're running a 3.91 diff in your 3.64 car, you can get an idea where you're markings should be.

[Napes]

My wife's '76, which came stock with the 3.91 diff, has that overlay on the face of the speedo, so I was aware of the difference between my '75, with a 3.64, and her car. Going side-by-side down the freeway at around 75-80 mph her rpm are 400 higher than mine. Makes for a little more cockpit noise.

Bob Napier

[KFunk]

Yeah, I assume the speedo I bought came from a '76. I don't understand why it's a W=1.393 (for a 3.91) and not a W=1.39, which would be correct for the 3.90 that came in '76s. I would think the E21 cluster would look very different from a 2002 cluster, but I don't know.

Was Curt saying they put the wrong speedo overlays on the '76 cars, meaning they put one for a 3.91 instead of the 3.90? Either way, guess I'm lucky to have a 1.393 in my car now!

http://www.bmw2002faq.com/component/option,com_forum/Itemid,57/page,viewtopic/t,328497/highlight,speedo+overlay/

Some '76 cars with 3.90 diffs got the wrong speedometers at the factory; that makes the speedo read about 7% high. Same for the odo reading.

I don't know where aa/fd got those numbers, but I do know where they originally came from -- see below.

There are several good reasons for keeping track of the origin of details like this. First, knowing the source helps you judge the credibility. Second, you know who to go to check on errors and provide updates. Finally, you know who to thank for their trouble.

aa/fd's info came from me. I started compiling it about 15 years ago. Here's an update:

3.45:1 diff -- W=1.24 speedo (1602)

3.64:1 diff -- W=1.297 or W=1.300 speedo

3.90:1 diff -- W=1.39 speedo ('76 49-State 2002)

3.91:1 diff -- W=1.393 speedo (E21 320is LSD)

4.10:1 diff -- W=1.4 speedo (1.461 calculated)

4.11:1 diff -- W=1.4 speedo (1.465 calculated)

4.38:1 diff -- W=1.559 speedo (calculated)

Yer welcome.

_________________

Curt Ingraham

1972 2002tii, 1976 2002

Improved 2002 Radiators

[Napes]

Considering that there is only about a 3 mph difference (@ 60 mph) between the 3.64 speedo and the overlay for the 3.90, I cannot image there is much difference between the speedos for the 3.90 and 3.91 setups.

Bob Napier

[KFunk]

Considering that there is only about a 3 mph difference (@ 60 mph) between the 3.64 speedo and the overlay for the 3.90, I cannot image there is much difference between the speedos for the 3.90 and 3.91 setups.

Bob Napier

Nah, there shouldn't be much difference. I'm primarily curious why they would've did it wrong from the factory.

However, I believe someone argued before that there was a real difference between a 3.90 and 3.91 diff. I don't know what numbers they cited, but they're indeed different. Perhaps it might just be in the long run on the odometer.

But for that matter, if they just added an overlay on the '76 2002s, wouldn't the guts and the odometer still be the same as the one on the 3.64 cars? And so the odometers of 76s can have very wrong numbers? and calculating fuel mileage would be wrong too.... I bet they changed a gear as well for the odometer, cuz I don't think they'd leave it that way.

[JohnS]

I have a couple of new 1,393 speedos that I purchased from Wallouth & Nesch a while back. Both of them are dated 7/76 and don't have that overlay. What's the build date stamped on the back of the case of your speedo? Curt may have been on to something. Those speedos with the overlay might just be a rare piece of BMW history. ;-)

PS, Can anyone explain what the little white dots on the face of the speedo mean?

[Napes]

This seems to get curiouser and curiouser. The one that KFunk says is stamped W=1.393 (with the overlay) has the 60 mph mark straight up, which is the same as my wife's '76 (with a 3.90 diff). But, the one purchased from Wallouth & Nesch, also stamped W=1.393, has the 60 mph mark offset to the left, about 3+ mph, the same as on my '75 with no overlay (with a 3.64 diff). Maybe someone has the answer to this apparent discrepancy, because I can't explain it. Maybe the worker at the factory that stamped the wrong VIN on the car in the previous thread ending up working for VDO.

Bob Napier

[JohnS]

I've been running one of those Wallouth & Nesch speedos in my '74 with a 3.91 LSD and I know that it reads correctly. I drive by one of those digital signs that reads out your MPH on the way to work. If you've ever messed around with your speedometer you'll know. It's easy to switch out the rear case. It's probably easier to slap on a new face plate (that fits over the needle) to essentially correct the read-out of a W=1,29 speedo to W= 1,39 than to really recalibrate the thing. Assembly-line engineering at it's best

[PatAllen]

its weird, the overlay is not linear...both read 120mph but the spacing above 50mph is not evenly spaced...strange, this is pure mechanical speedo...

[KFunk]

its weird, the overlay is not linear...both read 120mph but the spacing above 50mph is not evenly spaced...strange, this is pure mechanical speedo...

Yeah, that is weird.... maybe they just ran out of room to put the digits for 120, and said screw it, they won't go that fast anyways.

I'd like to tell you the build date on mine, but I don't feel like disconnecting the wires and pulling it back out of my car. I'd also like to see if I can compare the gears on each and see if there's a difference in the odometers. But once again, I'm afflicted with this crippling laziness....

[Guest Anonymous]

When I replaced the instrument cluster on my '76 last year, I was suprised to discover that it was stamped W=1.30, not 1.39. I doubt it was replaced by the PO (but who knows). I'm inclined to think that the factory just installed what they had available.

I replaced that speedo with a silver dollar one from a 1600. I notice very little difference between the two (although it could be measured). The difference between 1.39 and 1.393, I would think, is just on paper.

Hugh

[KFunk]

OK, found where john_a described the difference between 3.90 and 3.91:

The 1976 USA 2002 did not normally come with the 3.91 gearset, it was a 3.90. This may seem nit-picky, but there are strength differences to consider.

3.90 = 39:10

3.91 = 43:11

another pair that often get confused are the popular "quicker" gears:

4.10 = 41:10

4.11 = 37:9

These gearset pairs are too close to determine by counting revolutions.

Take the cover off when you change the oil next time, and count teeth.

See if your count matches the stamped numbers on the case.

So, it's a pretty big difference in the teeth on the ring and pinion... probably negligible difference to the speedo. I still wonder if my 1.393 speedo is counting the wrong number of miles on the odo, though (as if it has the gears for a 3.64). I've got to make a quick trip to my parents tonight, so I'll try to compare my miles to google maps calculation.

[KFunk]

Well, I charted out the exact route to my parents on Google Maps, and I got 16.1 miles. I reset the trip odometer, and got just shy of 16 miles. Not too bad. It's a curvy road and the bald 195-60-14s probably aren't the ideal size.

I'm wondering what kind of magnitude of difference there'd be with a W=1.297 odometer, but I can't think through the math right now. I bet that it'd be a bigger difference than 1 tenth of a mile, so I figure they must've changed an odometer gear too....

[Conserv]

In the winter of '77 -- perhaps '78 -- a buddy, in his '77 Capri V-6 and I, in my '76 2002, entered the New York Thruway, at Rochester, at precisely the time a 911 blew past doing around 90 mph. Ever seen a dog race? Right! So we took off, passing the 911, who was then happy -- I suppose, given our willingness to serve as "bear bait" -- to run with us. I placed my pedal to the floor and didn't lift until Syracuse, 90 miles later. My speedometer read 110 mph on the (generally mild) uphills, nearly 120 mph on the downhills, and 115 mph on the relatively flat sections. And, no, I never bothered to look at my tachometer!

 

My conclusions: (1.) my '76 was one of those rare examples that was, by the accidents of manufacturing, perfectly balanced and just plain faster than the average production 2002; and (2.) I was a helluva great driver -- although it's unclear which driver skills were demonstrated by placing my foot to the floor and steering straight (I did turn on my brights and four-way flashers, despite the daylight, and that obviously averted the expected disaster).

 

And then someone invented GPS. And my dreams -- about my car and myself -- were crushed. My speedometer is nearly 10% optimistic. I was actually running between 100 and 109 mph, in line with factory expectations. Why have I subjected you to this long-winded tale of crushing defeat? Because it got me thinking about the great '76 49-state "Speedometer Fiasco"!

 

As we know, 49-state versions of the '76 U.S. cars came with a 3.90 differential (39:10 teeth, NOT 43:11 teeth as in an e21 3.91 differential). And every '76 49-state example for which I've seen, or heard about, the speedometer back stamp had a head unit marked "1.393" (I've yet to see a 1.390 head unit). My car is typical. It was manufactured April 23, 1976. The instrument panel has 3 ink March '76 stamps, 1 ink February '76 stamp on the tachometer, and March '76 mechanically stamped into the back of the "alleged 1.393" speedometer.

 

And I recall reading at least two Roundel notes (shown below) in September and October '76 about the problems with speedometers in 49-state '76 cars. I asked my dealer at the time and the service advisor told me the problem only affected early '76 cars. April '76 is not "early '76" so I erroneously (stupidly?) assumed my car had no issues. The fix for the problem, a problem which I thought I did not have, was not a new head unit, not new gears, but simply a face overlay. I'm still looking to buy one of those overlays!

 

Since my speedometer is marked as a 1.393 -- theoretically correct for a 3.90 -- but still has the error, it would appear that the specific problem was NOT that they installed entirely wrong speedometer heads but, quite simply, they put the 3.64 speedometer gears in many speedometers that were marked “W = 1.393”. (I’d guess the new Walloth & Nesch units are correctly geared!) And it was easier to throw a faceplate overlay at the problem than to open up speedometers and replace the gears. I gather the odometers are off as much as the speedometers. If you have an overlay, it was installed post-factory, generally by the dealers: a cheaper means to correct the incorrect gears installed at the factory.

 

Yes, I know I can have my speedometer re-calibrated, but I really want an overlay to document the '76 49-state Speedometer Fiasco, and BMW's weak attempt at correcting this now-obvious defect. But even an overlay can't help recover the self-confidence I lost when some darned guy/gal invented GPS. Until that time, I was Fangio, I was Moss, I was triumphant!

 

Steve

 

 

[vinntagejoe]

Love the discussion and the thorough insight but you guys are killing me; I've got a 3.91 LSD and use a compact TOM TOM gps unit to track my speed - I know it's lame but someday I'll reach back to this discussion and do the right thing - heh

Edited June 14, 2014 by vinntagejoe