Need help determining spindle length please
 

I made an earlier post about wanting to convert my old Peugeot UO-8 to a cotterless crank setup and got some good advice. Basically I learned that I can either buy the new French threaded cups available on Amazon.com and a new aftermarket spindle OR the sealed cartridge unit available from Velo Orange so I can use just about any crankset. The big question in my mind now (besides which crank to use) is how to pick the right size spindle. The only advice I found on the internet on determining spindle length is to use whatever spindle length is recommended by the manufacturer of the crankset, but what if I want to use an old Nervar crankset so I can keep the bike French looking? Also to consider is the difference in taper between the Nervar crank which is most likely ISO and the Velo Orange which is JIS. I'm guessing any other NEW tapered spindle is going to be JIS also since it's most likely made in Asia somewhere, right?

Any help would be appreciated again. I'm determined to make this work, and I'd rather not have to break down and buy a new Asian made crank just to get the dimensions right without a lot of trial and error.

hello taguy,

you mention you would like to use "an old Nervar crankset." this might be something in the Star series, either five pin or five arm(?). if so, can pass along one tip. found that the Sugino Mighty three piece bb set is a perfect fit for these both in terms of taper and in chainwheel spacing/chainline. these have the advantage of being fairly easy to find and not too costly; plus they are readily serviceable for the home mechanic. the quality of the metal and rectification is better than that on the Nervar o.e.m. pieces so they wear better and turn smoother. Mighty spindles were widely offered for both two plateau and three plateau gearing so which ever you intend there should be no problem.

here is a diagram & photo of a Star 5-arm two plateau:

Nervar catalog (1974)

Nervar catalog (1974)

Star 5-pin two plateau:

Nervar catalog (1974)

Nervar catalog (1974)

Star 5-pin three plateau:

Nervar catalog (1974)

Nervar catalog (1974)

best wishes with your lion ;)

Thanks. That's very helpful!

If you want to use an old (ISO taper) Nervar crank on a JIS taper spindle it's going to stick out more than a JIS crank would.

Insofar as going by the manufacturers specs for used cranksets worn to varying degrees, I find that dancing in circles with headless chickens is a more reliable method of determining what the optimal spindle length should be. YMMV

The Nervar BBs I've come across have all had between 115-118mm spindles for a double crankset, which is also consistent with Stronglight and TA doubles/BB units. Given the JIS issue, you're probably best going with a 115mm spindle, but as Auchen notes, there's a black art to all of this figuring.

To use an ISO crank with a JIS spindle you need to subtract 9mm from the original ISO spindle length.
This calculation is for symmetrical spindles only.

For example using juvela's link for a Nervar double crank:
Original ISO length is 115mm. 115-9= 106mm

For example using juvela's link for a Nervar triple crank:
Original ISO length is 121mm. 121-9= 112mm

The best thing to do to save time and money is to bring your bike to your LBS. Let them dry fit the bb and crank for compatibility.

It sounds here to me that this "9mm" figure is somehow double of what it should read.

Perhaps a quoted-somewhere figure of "4.5mm" was doubled, intended as the sum of both end's "effective length increase", but 4.5mm by itself is more like the sum of both ends of the typical, roughly 2.5mm effective-length difference per side between JIS and ISO fitments.

So, from my experience and measurements, the overall effective length difference between JIS and ISO spindles is about 5mm, not 9mm.

This difference is easily verified at home by using a caliper (any kind, as long as it locks) set to the taper width at some distance from the end of the spindle, then sliding the caliper on the other spindle's taper and measuring the difference in distance from the end of the spindle.

Typically, I measure the larger taper first, as shown below, and then lock the caliper at this dimension.
The locked caliper is then applied to the smaller-tapered spindle end, with the amount of protruding spindle end noted as the difference in effective length of the one end of the spindle. This difference is then doubled to arrive at a difference of overall effective length between the two spindles.

https://farm8.staticflickr.com/7115/...a5b71450_z.jpg

Note that, for any accurate recording of spindle taper dimensions for use with possible future dimensional comparisons, a caliper like the one shown should be tested for accuracy using a new, Grade-25 ball bearing of known size. This will allow compensation for such caliper's known tendency to read low after extended use, something that is not correctible by using the caliper's ZERO function.

Also note that the size difference between JIS and ISO tapers is of the approximate order of ISO 12.7mm vs. JIS 12.9mm, with pre-cartridge Campag BB's measuring something in the neighborhood of 12.78mm (all Campag cartridge bb's are true ISO).

From Sheldon Brown website:
Square Taper Bottom Bracket Interchangeability
ISO/J.I.S. Interchangeability

"If you install an ISO crank on a J.I.S. spindle, it will sit about 4.5 mm farther out than it would on an ISO spindle of the same length."

To install an ISO crank on a JIS spindle you must subtract 4.5mm from each end, thus subtracting 9mm (4.5mm x2) from the ISO spindle's length.

I'm pretty sure the 4.5 is already the overall amount it will stick out.

I've mixed spindle lengths/types a lot before, just by trial and educated guesswork and have always used the 4.5mm figure as a reference point.

You're not going to see anywhere near 4.5mm per side from switching between ISO and JIS.

3mm per side is what I have seen as the upper limit, and 2mm per side would be more common actually.

Again, these things are best easily confirmed by the installer, using simple ~10-20 dollar calipers that should be in anyone's tool box who mucks around with bikes imo.

And again, the locking feature of most calipers allows one to go back and forth between any two spindles to positively home in on those small differences in the depth of engagement measurement. No rocket science required!

It might be useful to keep in mind that in most cases you'll be looking for possible differences of 0-3mm in engagement depth, or 0.0-0.2mm difference in the thickness of the spindle end, measured at some reference point distance from the end of the spindle (note in the photo above that the caliper jaw's thickness defines the distance from the end of the spindle to where the measurement actually takes place).