Chain Falling Apart

Question

I asked this on a cycling forum and sorry but it's sort of long winded. The experiment with the first chain I hot waxed went well and I haven't had any problems. So I thoroughly cleaned another one yesterday that was only a few rides old off the other bike. Then I waxed it this morning. Videos suggest anything from 20 minutes to an hour to allow the hot wax to seep into all the crevices and any air bubbles to disperse. I was dealing with other crises, so it got left in for 40 minutes in the wax (at 100 C because I'm using a double boiler saucepan thing)

Anyways, I fitted the chain which I always find is the most awkward and frustrating bike job because the ends always end up in the wrong place under pedals or they slip out of your hands and fling the joining links. With the chain fitted, I was only out the gate and 20 feet up the road when the chain broke and ended up on the road. Maybe this was due to stiff links because of the wax, and I should have spun the pedals first to work out the stiffness. It seemed one of the outer plates with two pins attached had pulled out of the matching outer plate. After fixing this I took off again with no mishaps. A couple of hours later, the chain broke again, going up a steep hill, again, the pins pulled out of the plates and an adjacent plate had pins on the way out too. So I fixed this. 15 minutes later, chain broke again. By then I had run out of quick links and had to walk 12 miles home.

So does waxing damage the steel of a chain? I have a theory that allowing it to sit in the wax for so long caused it to lose its temper/annealed it. The pins are press fit into the outer plates and presumably rely on friction to stay put. This comes from the force due to the springiness of the steel (It's compressed when the pins are press fit during manufacture) Maybe the heat reduced the elasticity so there was no longer a force? Several more of the pins now appear lose and ready to come out. Just to clarify, the outer plates didn't actually snap and break in two, the pins came out of them. The Chain is a Shimano. I've used four of these before and never had a problem and they never broke. In fact I've never broken a chain or lost links. The only thing that happened in over 20 years is I pedalled backwards once when changing gear and the chain came of the chainrings and the quick link got twisted and came undone. So maybe the links were still a bit stiff from the wax and each time the chain jumped the sprockets or chainwheels and came off, it got twisted and this wrecked the links? There's three quick links in it now and there'll be a fourth if I fix it, but I'll have to add chain because I lost links... The repairs are all over it, not adjacent.

Here's another theory. When in the hot wax, the outer plates expanded a lot more than the pins, allowing wax to get in between pins and plates. That lowered the friction between them making them more prone to sliding out. Alternatively it was the other way around and the pins expanded more than the plates, plastically deforming them so they ended up with slightly bigger holes or at least exerting less force on the pins, again making them looser. Some details on the chain as requested. It's reputably a Shimano 8 speed chain, bought on eBay. IG51 is marked on the links. I didn't push the pins back in again, I removed the ones that had any pins dislocated from the outer plates and any that had pins that seemed to be lose. Then I joined the ends with quick links. The pins from at least one of the plates don't seem to have any "upset" flares like the ones on the section of chain I removed when shortening it during fitting. They also slide fairly easily into the outer plates. So another possibility is that this is a bad chain and some pins weren't hammered/pressed to flare them during manufacture.

Answer 1

I believe your chain is counterfeit. The other answer and the various comments thoroughly debunk the heating claim, so I won't elaborate on that here.

Firstly, as I mentioned in my comment under your post, the holes in the outer plates look suspiciously rough and uneven. I'm not sure what they’re supposed to look like for IG51, but it should be safe to assume they won't look like that--the holes should at least be clean and smooth. Compare against the outer plates on the left side of the following image. Additionally, in your third pic you can see that the inner plates are just plain stamped steel. I'm pretty sure they're supposed to have a flange which supports the roller. See parts B in the image below.

https://www.parktool.com/assets/img/repairhelp/Figure_8-1.jpg

You identified that the faulty chain doesn't have flared pins, which I agree with. In the following image of what's supposedly legitimate IG51 AND in your pics showing "another Shimano chain", you can see that the pins are indeed flared in a four-sided pattern. Good catch.

https://i.ebayimg.com/images/g/0o8AAOSwZqZaPc8q/s-l1600.jpg

In your last two images, there's some other observations to make.

• The faulty chain only has lightly chamfered outer plates, whereas the other one has significant chamfers on both.
• You observed the difference in surface finish due to a lack of nickel plating on the faulty chain. Another good catch.
• It's hard to tell, but it appears as if the faulty chain has a more severe bend in the outer plates compared to the "other Shimano chain".
• The two chains use a different font, and the plates aren't stamped with the same pattern.
• The pins are plain steel with a bad surface finish, whereas a lot of Shimano chains use nice and smooth pins plated with shiny chrome.

Conclusion: I'm pretty sure that chain is fake.

Answer 2

This website presents a calculator for linear thermal expansion. I am not an engineer or physicist, so I am going off whatever high school physics I retained. Hence, please comment if my calculations or interpretations are wrong.

Change in length = Initial length * alpha * Change in temp

Above, alpha is the thermal expansion coefficient. Various numbers are reported for steel. From the site's own data, the lowest reported is 9.9 * 10^-6 m/(m˚C), and the highest is 17.3. Let's conservatively assume 20.

Paraffin wax's melting point is reported as 50-57˚C. I'll assume room temperature is 20˚C. Let's say that you did actually heat the wax to 100˚C.

Plugging the numbers in, I believe that a 1 meter length of chain would expand to 1.0016 meters. The same holds true in inches. Keep in mind that 0.5% is the usual wear limit for 11s chains, i.e. the chain should be replaced once an inch of chain 'stretches' to 1.005". 8s chains can be left to 0.75%. As an aside, for chains near the end of their wear life, I think that sometimes you can lose a roller off the end of the chain when waxing (I believe you can just slot the roller back in).

Thus, I don't think that thermal expansion should have caused the pins to come out of the rollers by itself. Also keep in mind that the chain cooled off, so it would have lost any elongation that the waxing caused. I believe that if you Google the material properties of steel, you'd see that you need very high temperatures to affect the chain's material properties. For example, this site says that the yield strength will start decreasing at 200˚C, but this decrease is temporary. Heating to at least 700˚C may change the crystal structure, but I don't think a slow cooker is capable of that temperature.

If the chain is genuine, I don't know how to account for the failures you described, so I will leave that to others. I have to note that if you bought the chain on eBay, there's a non-zero chance that it could be fake. If you check on forums, there has been an epidemic of fake chains on eBay, Amazon, and Aliexpress during the pandemic. Making a chain is not a simple process, and the fake ones could have much worse materials and tolerances. I think a fake chain could explain the failures you experienced. Even on older chain technology, a genuine chain should not fail in the manner described for a casual cyclist pulling away from a stop. A professional doing a hard start in BMX or MTB could do it - that's well over 1,000W, and that's the sort of force you should need to break a quality chain.