Have a question for Lennard? Please email him at email@example.com to be included in Technical FAQ.
I know you’re a big e-bike rider these days, so I’m looking for some advice.
On my normal bikes, I can easily clean and apply chain lube by leaning the bike against a wall and spinning the cranks backward. On the e-bike, the chain doesn’t move when you rotate the cranks backward. Any tips?
Indeed, the chainring on a mid-motor e-bike does not rotate when you rotate the crank backward. That makes it a challenge to wipe down and lubricate the chain. There is a simple workaround for some e-bikes, and for others, it remains a PITA.
If the e-bike motor is not geared down, meaning that the chainring rotates at the same rate as the cranks when pedaling, there is a simple workaround, namely to jam a stick in it. You take a short (2-inches long) stick or chunk of a dowel rod around ¾-inch in diameter and jam it into a hole in the chainring or spider so that when you rotate the crank backward, the chainring turns with it. If this works, you can wipe down and lube the chain in the same way you do on a regular bike.
However, this stick fix doesn’t work on many e-bikes. My current e-bike is one of those. It has a Bosch Gen 2 Performance Speed motor, which takes a tiny, 18-tooth chainring because the chainring rotates 2.5 times with each crank rotation. The chainring is first of all too small to jam a stick into, and, more importantly, because the chainring rotates so much faster than the crankarm does, you can’t jam a stick in it to make one move with the other.
Even though my wife’s e-bike has a mid-motor whose chainring rotates at the same rate as the chainring, jamming a stick in it still doesn’t work, because the chainring is too close to the motor. There simply is not enough space between the chainring and motor to allow the stick to overlap it enough to stay in place.
So, while I used to wipe off and lube the chain on my regular bikes after practically every ride, it is such a pain to do it on (either of our) e-bikes that I do it a lot less. I either have to do any wiping and lubing with the chain not moving, or I have to put the (heavy) bike in a bike stand. With the chain not moving, it’s hard to wipe the jockey wheels, and lubing the chain is laborious, dripping a single drip on each chain roller, one at a time. In the bike stand, I have to pedal it forward while wiping the chain and jockey wheels and lubing the chain, which of course causes the rear wheel to spin, making it hard to hold the lube bottle steady and can easily knock the bark off of my knuckles with the spokes or tire knobs.
Because of their extra torque, power, and weight, e-bikes wear chains faster than regular bikes do. Combine that with the extra hassle required to maintain the chain, and it’s a recipe for rapidly frying chains. In an attempt to mitigate it, I use Squirt E-Bike Chain Wax. Since it’s a pain to wipe the chain down, I want a chain wax to get the grime to fall off rather than be attracted to and adhere to a sticky and oily chain. And Squirt is very slippery (it won VeloNews’ chain lube tests seven years ago), thus slowing wear a bit. I still have to replace chains pretty frequently, because I ride my e-bike a lot.
Recently I tried to find out if using 39-tooth chainrings instead of the more usual 42T would perceptibly affect the efficiency of a tandem timing chain. Failing to find any hard data for tandem drivetrains, I turned to other data in hopes of coaxing out an answer from general patterns. I had to discard older sources such as Keller (1983), Pucket (1999), and Spicer (1999), as they revealed inconsistencies when seen up close, which left me with only the work of Jason Smith at Friction Facts to go by.
Having missed the opportunity to purchase Friction Facts’ reports when they came out, I had to glean the information I could from the graphs published with your article in the Tech FAQ column and in the DiamondBack article “Friction profiles of 1x drivetrains”.
In both plots, the abscissas [horizontal (x) axes] are gear ratio (adimensional) and the ordinates [vertical (y) axes] are drivetrain power loss (Watts).
In his article, Jordan Rapp listed the losses as measured by Jason Smith for cross-chaining. This allowed me to produce the following plots by subtracting the cross-chaining losses (jockey wheel losses remain), where the x-axis of the first plot has gear ratio as before
The x-axis of the second plot is the sum of the inverses of the tooth count of chainring and sprocket. This has the effect of normalizing for chain tension, resulting in an almost perfect superposition of the curves for all four drivetrain cases.
Incidentally, at this point, I was already able to answer my original question. The two black data points in both the graphs above are my estimates for the losses of 39×39 and 42×42 tandem timing setups, adequately extrapolated/interpolated from the general trend visible in these graphs. It turns out that the power loss difference from using 39 x 39 instead of 42 x 42 is very small, only 0.2W or about 0.08 percent for a 250W input.
Applying the same treatment to the VeloNews-sourced data, I was surprised to see that the curves did not come out as neatly overlaid as for the DiamondBack-sourced data. Subtracting the previously used values for cross-chaining and plotting for gear ratio, I obtained this:
Plotting for the sum of the inverses of teeth counts, I derived this graph.
As can be seen:
– the curve for the SRAM FORCE 1x with PC-1170 chain is higher than would be expected;
– the curve for the large ring of the Shimano Ultegra setup with CN-HG701 chain is much flatter than should be expected.
This misalignment is strange as all the data ultimately comes from the same source, Jason Smith!
My guess is that the SRAM curve is not in error and simply reflects the higher friction of the PC-1170 chain (Jason’s reports often showed SRAM’s chains to be rather lossy).
I sincerely cannot make sense of the Ultegra big-ring curve.
In your article, you finished by concluding that the Ultegra 2x setup was more efficient than the SRAM 1x setup. I am afraid this conclusion may have been a side-effect of the lossier SRAM chain and of the strange Ultegra big-ring data.
Here is the response to your question from Friction Facts founder and CeramicSpeed chief technology officer Jason Smith:
Regarding his comment, “In your article, you finished by concluding that the Ultegra 2x setup was more efficient than the SRAM 1x setup. I am afraid this conclusion may have been a side-effect of the lossier SRAM chain and of the strange Ultegra big-ring data.”
Your reader’s observations are both correct, and incorrect. The differences between the 1x and 2x that we found during testing are the results of 1) the general mechanical physics of the 1x vs 2x (cross chaining and ring size) and 2) the other variables, such as the slower SRAM chain and the fact that the SRAM rear derailleur probably had a little bit stronger spring tension (regardless of the clutch), adding more friction.
Therefore, he is correct in that the SRAM chain (and possibly SRAM rear derailleur) most likely added to the delta between the two drivetrains. However, some delta would still exist between the 1x and 2x even if the chains and rear derailleur were equal. Point is, the entire difference is not due to the SRAM chain or SRAM rear derailleur. Some difference is due to the basic 1x vs 2x mechanics.
Jason Smith, CeramicSpeed founder and CTO, Friction Facts
Thanks for your question, Francisco.
Lennard Zinn, our longtime technical writer, joined VeloNews in 1987. He is also a custom frame builder (www.zinncycles.com) and purveyor of non-custom huge bikes (bikeclydesdale.com), a former U.S. national team rider, co-author of “The Haywire Heart,” and author of many bicycle books including “Zinn and the Art of Road Bike Maintenance,” “DVD, as well as “Zinn and the Art of Triathlon Bikes” and “Zinn’s Cycling Primer: Maintenance Tips and Skill Building for Cyclists.” He holds a bachelor’s in physics from Colorado College.
Follow @lennardzinn on Twitter.