On Wed, 12 Jun 2019 12:24:40 -0400, Frank Krygowski
wrote:

On 6/12/2019 1:51 AM, John B. wrote:

Some time ago I posted a question about the possible friction losses
when various "gears" are used. Subsequently I have come across a test
of both chain misalignment and sprocket size friction losses.

The test was done with 53-39 front chain rings and a
11-12-13-14-15-17-19-21-28 cassette. The lowest losses were with the
39/17 combination and the highest with the 53/11. Losses with the
53/11 combination were 3.467 times higher than with the 39/17.
The graph states that it is based on the "optimal shift sequence using
the ring-cog combinations with the lowest frictional losses yet
maintaining an acceptable range of final gear ratios."
https://www.ceramicspeed.com/en/cycl...g-size-report/

The second part of the test shows the friction losses with the chain
aligned and using the various ratios of the cassette and chain rings
above.
The lowest losses were with the 53 tooth chain ring and the highest
with the 39 tooth chain ring in all "gears". In other words the 53/ 11
ratio had ~8.2 watts losses while the 39/11 ration had ~8.5 watts. The
53/28 had ~5.6 watts losses and the 39/28 had ~6.2.

So the test was run with an input power of 250 Watts. Typical losses
were about 7.0 to 7.5 Watts. That gives a typical efficiency of 97%, and
that applies to almost all the various chainring and sprocket
combinations. The 53-11 combination drops the efficiency down to 96.6%,
still not too shabby. It's only the weird combinations like 39-11 that
lose significantly more, and even that comes in at 96% efficient.

The important point, though, is that there doesn't seem to be a more
efficient system. Yes, the efficiency would drop if the chain was
extremely grungy and (I assume) worn. But that's easily fixed.

I don't think that the test was done to illustrate the overall
efficiency of the chain drive system but to highlight the fact that
horizontal chain misalignment does reduce efficiency. And, in general
terms, to give some indication of how much.

The second part of the test, that measured the losses depending on
sprocket size was also interesting with all things taken into
consideration showing that large sprockets have lower losses than
smaller sprockets.

As an aside, 250 watts is probably as high, or perhaps higher, than
the usual recreational cyclist normally produces.
--
cheers,

John B.