The recent ‘I despise everything Ceramicspeed represents’ thread got me thinking about the numerous products aimed at reducing ‘parasitic drag’ from the drivetrain, i.e. ceramic bb’s, hubs, and pulleys, and ‘optimized chains and lube’.
Case in point, BBInfinite claims that a rider using their ceramic bottom bracket averaging 18mph will save 2.02 minutes(?!?) over a 20 mile course vis-a-vis ABEC-7 Pressfit steel bearings. To say that I’m incredulous would be putting it mildly.
Having said that, the measurable differences in drivetrain efficiency seem far more reliable and repeatable than some of the aero claims I see, especially given the myriad variables in rider interaction. As I recall, Friction Facts was pretty well regarded before they were absorbed by CeramicSpeed.
So what do you think? Are ceramic bearings, optimized chains, and wonder lubricants nothing more than a marketing ploy, or are they easily proven to increase your speed, similar to the rolling resistance testing with race tires?
p.s. I totally spent the extra $65 to get the BBInfinite ceramic bearing assembly over a standard one for my Speed Concept
Funny. It isn’t quite a correct conversion for their fastest speed, but if you use the 1 watt power increase/savings is worth 4 seconds over 40k, these guys are claiming a 20 watt savings for their 20k TT. You would have to pack those standard bearings with sand to lose that many watts. Definitely NOT a reputable company.
Reading this It occurs time that we don’t have a general rule of thumb for total drivetrain losses of an average setup. Then you would have a basis to compare with a really well scienced out drivetrain. It seems like that would be a good thing to know.
i cannot imagine why frictionfacts would be deemed to have the same lack of editorial bias as before its purchase by ceramicspeed. you are right to have a healthy skepticism.
you should have the skepticism about what we write on slowtwitch, and you should demand that we be clear with you what our sources of bias would or could be if and when you sense that might be an issue with us.
you should demand to know the editorial policies of other magazines. have you ever wondered whether an article, written under a byline of an author you trust, is true unbiased editorial or whether that was a “native” piece paid for by an advertiser? you should start asking that question.
what i don’t like is when people just assume an author or entity is lying, and label someone a liar and corrupt, whether it’s jason at frictionfacts or me (and i don’t think you’re doing that). but it’s fair to ask for and to require a second step from FF, which is to acknowledge the potential for bias and to deal with it in a straightforward and transparent way.
“Three riders of differing skill levels ride the same 20 mile TT course, once with ABEC-7 steel equipped pressfit bottom brackets, and then again, this time with Ceramitech bearings.”
Wow…way to control the variables and narrow down all the differences in speed to the bottom bracket.
Is the guy from Karbon Speed running their marketing campaign?
Reading this It occurs time that we don’t have a general rule of thumb for total drivetrain losses of an average setup. Then you would have a basis to compare with a really well scienced out drivetrain. It seems like that would be a good thing to know.
I’m too lazy to look for it, but all this stuff has been tested. A couple watts for hub and BB bearings (steel) is a good ballpark. It can be a lot higher due to misadjustment and misalignment.
Funny. It isn’t quite a correct conversion for their fastest speed, but if you use the 1 watt power increase/savings is worth 4 seconds over 40k, these guys are claiming a 20 watt savings for their 20k TT. You would have to pack those standard bearings with sand to lose that many watts. Definitely NOT a reputable company.
Reading this It occurs time that we don’t have a general rule of thumb for total drivetrain losses of an average setup. Then you would have a basis to compare with a really well scienced out drivetrain. It seems like that would be a good thing to know.
Interestingly enough, Ceramicspeed claims 2 minutes over 40k at 32 km/h when using “CeramicSpeed Bearings in your hubs, 11-tooth pulley wheels, and bottom bracket compared to a set of standard bearings”. Evidently, this does not include the OPSW or optimized chain.
Intuitively, it is hard to imagine the same amount of savings just from using a ceramic bottom bracket from BBInfinite, but I suppose my question remains…given the ability for an entity such as Friction Facts to use a reputable protocol to demonstrate this type of savings from reduced drivetrain friction, why are we so much more skeptical of time savings via drivetrain efficiency as opposed to aerodynamics? Is it because there isn’t a clear correlation between that reduced friction and increased speed in the real world?
Reading this It occurs time that we don’t have a general rule of thumb for total drivetrain losses of an average setup. Then you would have a basis to compare with a really well scienced out drivetrain. It seems like that would be a good thing to know.
As a *very *broad rule of thumb, a drivetrain efficiency of 95% is seen as a sensible conservative estimate. So if you’re putting in 200W at the pedals, you should expect to getting at least 190W at the rear hub. With a nice clean, maintained drivetrain you might expect to add another percentage point or two on top of that.
Someone who routinely runs one or more power meters in different locaions may be able to comment on how accurate this might be and how well it scales, but from what I’ve seen it seems to be a good ballpark estimate
Wouldn’t it be very easy to quantify drivetrain losses? All you need is a Powertap Hub and P1 pedals or Vectors. The difference in power is the drivetrain loss. Or would this difference be too small when you account for the measurement error in each of the meters?
I’ve never run a hub-based PM, so have limited experience, but whilst I wouldn’t necessarily trust the absolute values from each power meter to correlate, I reckon you’d be able to discern (and quantify) improvements in an A vs B type of test protocol.
e.g. run both PMs for a few minutes, switch out bottom brackets and run again, compare data.
You should certainly be able to pick up the ~20W BBInfinite are claiming
Reading this It occurs time that we don’t have a general rule of thumb for total drivetrain losses of an average setup. Then you would have a basis to compare with a really well scienced out drivetrain. It seems like that would be a good thing to know.
As a *very *broad rule of thumb, a drivetrain efficiency of 95% is seen as a sensible conservative estimate. So if you’re putting in 200W at the pedals, you should expect to getting at least 190W at the rear hub. With a nice clean, maintained drivetrain you might expect to add another percentage point or two on top of that.
Someone who routinely runs one or more power meters in different locaions may be able to comment on how accurate this might be and how well it scales, but from what I’ve seen it seems to be a good ballpark estimate
Yea 95% is the general number for a nice clean drivetrain with fairly standard components. There would have to be negative friction to gain the results they are claiming or are comparing a dirty drivetrain with bushing for derailleur jockies.
Wouldn’t it be very easy to quantify drivetrain losses? All you need is a Powertap Hub and P1 pedals or Vectors. The difference in power is the drivetrain loss. Or would this difference be too small when you account for the measurement error in each of the meters?
I doubt seriously that the little difference it might show, would not take that difference number out of the tolerance specification (or combined measurement uncertainty). So no, I don’t think this method would be reliable.
Funny. It isn’t quite a correct conversion for their fastest speed, but if you use the 1 watt power increase/savings is worth 4 seconds over 40k, these guys are claiming a 20 watt savings for their 20k TT. You would have to pack those standard bearings with sand to lose that many watts. Definitely NOT a reputable company.
Reading this It occurs time that we don’t have a general rule of thumb for total drivetrain losses of an average setup. Then you would have a basis to compare with a really well scienced out drivetrain. It seems like that would be a good thing to know.
Interestingly enough, Ceramicspeed claims 2 minutes over 40k at 32 km/h when using “CeramicSpeed Bearings in your hubs, 11-tooth pulley wheels, and bottom bracket compared to a set of standard bearings”. Evidently, this does not include the OPSW or optimized chain.
Intuitively, it is hard to imagine the same amount of savings just from using a ceramic bottom bracket from BBInfinite, but I suppose my question remains…given the ability for an entity such as Friction Facts to use a reputable protocol to demonstrate this type of savings from reduced drivetrain friction, why are we so much more skeptical of time savings via drivetrain efficiency as opposed to aerodynamics? Is it because there isn’t a clear correlation between that reduced friction and increased speed in the real world?
I don’t think there is any question by people on this forum that reduced friction leads to faster times. People get bent out of shape when someone makes dubious claims. For example, a reasonably maintained and lubed chain represents probably a 6-7 watt frictional loss. The absolute best 200$ Treated chain is probably high 3 watt range. A waxed chain using some thing like Molten Speed wax is about low 4 watts. So there are 2-4 watts to be saved in total and the last 1/2 watt will set you back 100$.
I think people harp on frictional reductions because on a watt/dollar basis the top of the line stuff represents the worst $/watt saved ratio in the bike industry. That and the ridiculous claims some unscrupulous companies make.
P.S. That didn’t stop me from spending 450$ on a Ceramic Speed cage and pulley setup for my ETap rear derailleur.
Well, you could reduce the impact of the noise by sticking the bike on a turbo trainer, spinning for a decent amount of time and analysing the discrepancies in total kJ of work from each PM. The raw kJ values from each PM might not necessarily line up, but I reckon you might be able to discern some differences in an A *vs *B test; you’d have to be careful to control for outside variables though, such as chain run-in, seeing as the effect that you’re trying to measure is so small.
Surely someone on the forum must have tried something like this?
As you infer … it is possible that there are device setups that are better at quantifying drive train efficiency rather than timing different riders in the wild.
I roller test tires all the time on my rollers. The longer you ride, the larger the sample size, the narrower the confidence interval of average power. The numbers are fairly repeatable even though one of the bigger problems is probably ensuring equivalent tire inflation. I bet you could suss out a 2 watt difference in drivetrain friction. Send me a set of those Ceramic Speed bearing and I will let you know.
This laboratory finish time results with specific Cr specs for bbs only but not any more relevant specs for the riders are getting closer to the real world if all 3 riders are already super efficient, small and light on heavy bikes at high level with warm temperatures … www.kreuzotter.de
Reading this It occurs time that we don’t have a general rule of thumb for total drivetrain losses of an average setup. Then you would have a basis to compare with a really well scienced out drivetrain. It seems like that would be a good thing to know.
I’m too lazy to look for it, but all this stuff has been tested. A couple watts for hub and BB bearings (steel) is a good ballpark. It can be a lot higher due to misadjustment and misalignment.
2.5% total losses in drivetrain and bearings is the generally accepted “average” for a clean, lubricated drivetrain and cup/cone steel ball bearings (ALL the bearings involved…pedal, BB, wheels), as I recall. That’s 6.25W TOTAL for a rider averaging 250W.
As can be seen…there’s not a lot there to “optimize away”
