...pretty much. If something looks problematic, I might do the heating and ATF/acetone cycle a couple of times before I put the press on.
I try to heat mostly the cotter, to avoid any problems with possibly discoloring the chrome on the arm.

I'm not sure what this thread is about, but it ain't about cotter pins....:foo:

I've always called them crank cotters bcuz they don't look anything like cotter pins!

Thank you for that perspective. You are right. Glad you said it out loud.

An analogy. I have read dozens of versions of how to build a wheel. Some well written some badly written. Most include unnecessary steps and superstitious nonsense. Every account ever seen of how to build a wheel will result in a wheel. Follow the directions, get a wheel every time. Not every wheel is great, they are all wheels.

Bicycles are are simple machines. Knowledge of how to get cotter pins to work was widespread not that long ago. Doing it with Germanic precision will certainly get the job done. Cannot believe that millions and millions of bikes were assembled with that kind of precision. My brother just sent me photos of some very old Stronglight pins that work and work well. The flats are not flat. They look like a mountain range. And they work. No idea why.

They probably work because the angles are close to the same. No need to over think this - no need for "Germanic precision" (whatever the hell that is). Both flat angles need to match or your cranks won't be clocked right and your pins will loosen because the securing forces aren't distributed across the flat, but rather at a small portion on the flat, causing a gouge. Especially important on a fixed gear where pedal pressure is in both the forward and reverse directions. iab's sharpie trick is the way to go if you want to slowly figure out the correct angle.

Cotters are a very simple way of securing cranks. The more precise the better, but they don't need to be ultra precise. In fact some spindles come quite roughly finished and they work OK. I read something on one of your posts about tapping the cotter. They should not be tapped, but but installed with force, either with a press or by hammering (supporting the crank in this case). If not, they will come loose.

Wow, I'm surprised there is so much fuss about it.

I started working in bike shops in 1978. In that year, I think we had more bikes with cotters than without. So I had to do it a fair bit. We did not have a cotter press because one of the owners was cheap with tools, and his 3rd world ways were good enough.

Take two cotters from the same source so their cut angles are the same. The diameter must be right, as someone mentioned above. Test fit the cotters to see if they expose all the threads. If not, file them down.

To file them, put them in a vise, and maintain the angle by eye. All these methods for ensuring that are impressive, but I guess my eye is good enough. Or maybe the cranks I installed were not perfectly 180º apart but no one ever noticed a problem.

Rest the spindle on a big heavy support. We used a plumbing nipple that was six inches in diameter or so and about 12 inches tall. One person held the bike from the top and lowered it onto the nipple. And by the way, we called the nipple Herman so we didn't have to say the word nipple. So hold the bike on Herman, and the other person bangs the cotter in. Unlike what you said, [MENTION=362926]63rickert[/MENTION], no, not with a light tap but with very vigorous and repeated banging. This banging is forceful enough that if you do not support the spindle with Herman, you will damage the cups. You'll do the damage because the shock to the cotter transfers to the arm which transfers to the spindle which transfers to the bearing balls which transfers to the cups. Use a ball peen hammer or a sledge hammer, not a carpenter's hammer.

At subsequent shops, we had a Park cotter press. It usually worked but not always, especially for removing.

I might have used a VAR press at some point.

I've tried the homemade presses, including the one made out of a motorcycle chain riveting tool from Harbor Freight. They did not press the cotters in tightly enough.

I rarely file cotters. It's rarely necessary. If I'm reusing an old cotter, and the face is damaged, i will file it smooth. And of course if a cotter just won't fit othwise, break out the file.

My tool of choice is a small vise, the kind made for use with a drill press. i have several other cotter tools, including the BikeSmith one, and they all have their pros and cons. The BikeSmith tool is made for Raleigh Sports cranks, which have a chain guard. More racy bikes have thinner cranks and the Bike Smith tool just doesn't work.

My first ten speed had a cottered Nervar crank that always creaked. i replaced the cotters many times, but it always creaked. I eventually concluded that the cotters weren't the issue.

+1! I’ve been working on cottered cranksets for many decades and have yet to file a pin. Instead, I have a bag full and just mix and match until I find a pair that allow the cranks to line up properly. Insert and tighten. Done.

Google "cotter pins" and see what you get. There are a many different styles. The most common are the split cotters. Anyway, your comment proved interesting and caused me to think about it for a bit.

You don't even bang or press them in with a leveraged device?

That’s all part of “insert.”

Yes to the part about gouges. Each test fit that seems to work well enough to attempt riding there is a gouge across the face of the wedge. The pins inserted with LBS Var tools also made a gouge. The LBS gouges are engaging much less of the pin than my attempts but work longer. Long in this case is ten miles.


Have done the sharpie routine more more than a dozen times now. Cannot glean any information that way.

How to engage the two mating flats is the whole problem. There has to be a simple way to do that. Every attempt so far engages the edge of spindle to a line on the cotter. Or at least that is what the damage says to me.

Right. Skip one or two or five steps and pretend that you said something. This is why it is so hard to learn anything in a place like this.

There's a potential cultural element here as well, tied to region. But I know people from one region may not be as sensitive as those to others. per Wikipedia:

A cotter is a pin or wedge passing through a hole to fix parts tightly together. In British usage cotter pin has the same meaning, but in the U.S. it refers to a split pin. Typical applications are in fixing a crank to its crankshaft, as in a bicycle, or a piston rod to a crosshead, as in a steam engine. Wikipedia

It's also really hard to hear with all of the whining going on.

Interesting but not useful. I don’t have your bag of cotters. There is some reason that casual fitting of type described works. I am getting a lot of reports that casual work often does the job. But something has to be right with that casual fit. Everything tried so far is wrong.

That is interesting and useful. All my cranks are old and quite thin. The attempt to use the Bikesmith tool was a disaster and now am sure there is no further need to think about that.

I am working with new pins. No old pins here. They don’t fit at all if used as is.

Yesterday had a forty minute tutorial on how to file and get a flat surface. My teacher was a plumber and luthier. He said I was the worst student he had ever had but by end of session my work was good enough for plumbing or bikes. Still no clue at all what the shape of the cotter pin should be. How steep is the taper? How deep is the cut? No one seems to know.

There is no question my problem is the cotter. Crank comes off.

If it was 1978 and it was a low value bike and there was a container of low value bikes out back waiting to be assembled and I was being paid hourly I would use a sledgehammer.

Many many reports from old mechanics that sledgehammer work is not necessary. Have now inspected sample old cranks on which the cotters lift out with two fingers or a tap with a 6 ounce hammer. Those cotters work. Why do they work?

If I had access to a Var 7 or even a Park it would make sense to learn how those work. In meantime any archived Var tools that exist are in LBS where it is all covid rules and I get to stand outside the door while a young hourly employee makes a hash of the job.

Think of cotters as a game. To win the game, you need as much of the filed face of the cotter to come in contact with the flat face of the spindle as possible. The more surface area of each flat you have in contact, the more chance you'll have of success and thousands of miles of creak-free, loose-free riding. If you're seeing a significant linear gouge, you're not engaging the flats well enough, deep gouge means you're catching the edge of the spindle before the flats can engage nicely, or after you're too far past engaging.

You need to press or punch in the cotter enough to see where you're making contact, then adjust the angle of the cotter so the flats match. Sometimes you'll get lucky and you'll get good engagement without touching a file - it happens! But more often than not, especially with the French stuff I'm wont to work on, I'm filing a good bit of the face.

Best I can do is throw together a picture to visualize it. Blue line is where you're meeting the spindle. You want contact flushly across the face for a long blue line, engaging across a lot of the surface. If your angle isn't steep enough to meet the face, you'll gouge at the blunt end. If you over-file, you'll gouge a line into the threaded end. It's that simple. File things flatter on the end you're gouging, or if you're gouging on the threaded side, you might've over-filed already (it depends on how much blunt end is exposed after a full press!)

Make any more sense? Just make sure you file evenly, flatly, because the spindle face should be perfectly flat also, you want flat to flat.

https://cimg7.ibsrv.net/gimg/bikefor...2ed3f07bb2.png

We saw the cotter pins get shredded when we did not bang or press the cotters in hard. A lot of torque goes into the spindle.

Hmm, maybe we didn't manage to get good surface area unless we banged hard. Perhaps our filing technique was bad enough to require hard banging.

I'll avoid any further innuendo. :D

Reality is, I wish I could get full-surface contact like the far left cotter. But the reality is, I'm usually only getting about 1/2 - 2/3 contact best I can tell. And I'm mashing hard on them at a low setting, generating 2x-3x what a BFH would when I press in.

But when you see a 1mm gouge on the high or low side, you know your angle is off, and the side with the gouge is the "high" side you need to plane down. I normally aim for a wide rub, deep linear gouge is bad.

Reverse threads?





(Sorry; I'm not helping, am I?)

Sorry, that's my dyslexia shining through :D Threads so steep and reverse it looks more like an ez-out!

The humor is definitely needed and doesn't bug me a bit. Keep it coming!