I've got an almost new chain, but forgot to oil it before today's ride. It would probably have been fine except for repeated dunkings in muddy water early on. Later in the ride it developed a noticeable squeak. Although I carry oil, I didn't particularly want to add oil on top of filth, and couldn't really be bothered (an ill-judged route led to a fair bit of walking/wading so time and cold were both sub-optimal).

But in hindsight, could stopping to oil the chain have actually got me home sooner? Was I wasting loads of power?

I ask in a general sense, but my setup is Sora 3x9, on a heavy tourer (and not very aero with extra bottles on the forks).

  • The loss is minimal, so long as the chain isn't actually freezing up in spots. You lose more energy worrying about it that you do from the friction. The real danger of an unoiled, dirty chain is chain and sprocket wear. Dec 12 '20 at 21:30
  • 14
    The perceived power loss is much higher than reality - the peloton gains between 10% and 20% of motivational power in an attempt to drop the twat with a squeaky chain :)
    – mattnz
    Dec 12 '20 at 21:50
  • @DanielRHicks I reckon it will need a new chain and sprockets in spring anyway. Winter is filthy round here even if I don't do silly things like taking the tourer down farm tracks.
    – Chris H
    Dec 13 '20 at 8:25
  • @mattnz good point. If I'd been with a group the stops would have been longer and I'd have got the oil out. Without the option of going inside to keep warm, I wanted to keep moving, so my longest stop was 15 minutes in a bus shelter, and a total of maybe 45 minutes in 11 hours
    – Chris H
    Dec 13 '20 at 8:32
  • 1
    @Carel, I am and I did! Estimating audio power is hard, but I reckoned no more than about 1W. As the noise is caused by friction there will also be heating, and that's even harder to estimate. I carry enough stuff without the IR thermometer, though I keep meaning to point that at my brakes on a long downhill
    – Chris H
    Dec 14 '20 at 9:02

The question of how much power was lost has been answered. I would just add this: if the chain is squeaking, it’s got insufficient lubricant. You are presumably having bare metal rub against bare metal. That will cause a lot of wear, in addition to drivetrain friction.

Even though the chain was dirty, I don’t see how adding lube on top of the chain could have worsened the situation. The lube should penetrate to the rollers and probably stop the metal on metal situation. That said, note the discussion in the comments; you could risk importing more dirt into the rollers, which would increase the abrasion rate. The question about which of the two situations is less bad might only be answerable if we experimented, and we'd probably need a microscope to measure the amount of contamination between the rollers and pins. In either case, the chain won't explode if you let this happen once or twice, and take this as a reminder to keep the chain lubricated.

  • 3
    The lube could worsen the situation if in penetrating to the rollers it carried a small amount of grit with it. Fine grit in a lubricating carrier is essentially grinding or polishing paste.
    – R. Chung
    Dec 12 '20 at 21:48
  • 2
    @R.Chung it could. However, if there’s no lube inside the rollers, then you already have metal on metal. Also, in the OP’s situation, I would expect inside the rollers to be pretty gritty already.
    – Weiwen Ng
    Dec 12 '20 at 23:43
  • 1
    I was thinking along similar lines to @R.Chung, and that dirty but dry wouldn't be as bad as abrasive paste, which would then be very hard to get out. I'll give it a good clean this morning before oiling
    – Chris H
    Dec 13 '20 at 8:20

This seems to answer your question:


It suggests a dry chain loses 7W out of 244W, which is 2.9%.

Your % speed loss will depend on your speed and aerodynamics - if you are producing, say, 80W, and cycling at 10mph on a flat road, then you could expect to lose almost that much time (so maybe another 100 seconds over an hour). At higher speeds, say 25mph on a flat road, then almost all of your losses are aerodynamic, so it comes increasingly true that your gains (or losses) are proportional to the square root of power losses, so you could expect to lose perhaps 1.5% of your time, so something like 54 seconds.

Of course stopping to lube isn't a bad thing as it helps avoid that little bit of premature wear to components meaning your chain lasts a few more miles, but if it was raining or whatever, it is probably better to just keep going, as it would just wash off anyway.

  • 4
    That’s not exactly my read of the article. It suggested that (last table) at 250W input, you could expect to lose 15W more with a totally unlubricated chain than a lubricated one (unspecified lube), probably before introducing any contamination. A squeaky chain still has lubricant. The watts lost will be less than 15, but that’s before we consider contamination. Contamination with a light oil lube adds an average of ~4W over a CX race (~1hr). With a squeaky chain, I’d expect the same conditions to produce slightly higher penalties.
    – Weiwen Ng
    Dec 12 '20 at 21:21
  • I'm not referring to that table. The last table shows thst a lubed chain loses 6W and then if totally degreased 20W. I was referring to the third table which shows 6W lost lubricated with a light oil, and 13W in rain. Rain would wash out some of that oil. So a chain in rain still has lubricant, just the same as a squeaky chain. That's why I cited 7W (rained on light oil), not 14W (completely unlubricated)
    – thelawnet
    Dec 13 '20 at 7:53
  • 2
    A few Watts, or about twice my dynamo lights (which were on much of the day) is sufficiently precise for me
    – Chris H
    Dec 13 '20 at 8:33

The amount of power you lose is proportional to the amount of power you're putting in, so there's not a one-size-fits-all answer to this question.

Although the question of which lube is best has been studied extensively, as far as I can tell, these studies don't include unlubed chains (which would be interesting as a control). What those lube tests show is that when riding at 250 W, a chain with a top-quality lubricant will cost you about 5 W, and a poor lube will cost you about 10 W.

I'm not sure if we can extrapolate from there to an unlubed chain, but perhaps that would cost 20 W, or 8% of your power input. That might have been a difference of 2–3 km/h.

For whatever it's worth, the factory lube on chains is usually very good. I always ride new chains without adding lube until they squeak, and then try to strip off all the lube before I add my own.

  • ...so when walking there is no power lost to a noisy chain :)
    – Criggie
    Dec 12 '20 at 21:01
  • Can we have a reference to the first line which states expects us to 'take as read' a statement of fact. I have no idea if the statement is true or not, but do have doubts as to its accuracy.
    – mattnz
    Dec 12 '20 at 21:48
  • 2
    @Criggie - I lose a lot of power in my noisy knees. Dec 12 '20 at 23:51
  • 1
    @mattnz Here's an article on the subject. Chain efficiency is given as a percentage, not an absolute number. This is at CeramicSpeed, which is not a disinterested party, but I believe the work was originally done by Friction Facts, which was independent until CeramicSpeed bought it out.
    – Adam Rice
    Dec 13 '20 at 0:07
  • I disagree with "the factory lube is usually good". Whenever I think that I end up with a rusty squeaky chain after the first wet ride of any length. This is (typically) with something like KMC's cheapest 9-speed, but also other brands
    – Chris H
    Dec 13 '20 at 8:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.