why don't we have actual data on Sram axs drivetrain efficiency?

Back by popular demand, the general all-things Road forum!

Moderator: robbosmans

whataboutEee
Posts: 123
Joined: Fri Feb 19, 2021 8:55 pm

by whataboutEee

How is it that we still don't have any actual data on the real numbers of drive train efficiency for sram axs systems? I know that we can make some assumptions about what the smaller cogs/chainrings will mean for watts lost compared to a normal 53/39 setup, but we don't know anything about how the flattop chain factors in. Sram's other chains have done really well on wear tests, and really poorly on friction tests, and if the flat-top chains are similar, the combined friction loss between the chain and smaller cogs could add up to enough watts to be a legit detriment, and not just an academic discussion.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



chicagocyclist
Posts: 10
Joined: Sun Aug 15, 2021 3:40 pm

by chicagocyclist

I'd say in the real world there's no noticeable difference. Heres a couple articles on the topic:

https://www.kogel.cc/blogs/kbba/sram-ax ... n-concerns

https://www.sram.com/en/life/stories/10 ... or-the-win

blaugrana
Posts: 457
Joined: Wed May 24, 2017 9:49 pm

by blaugrana

I wouldn't conclude that there is no noticeable difference, though. Kogel estimate between 1 and 6 W of loss (big window, but it has to be since it's mostly just guessing), but then we have to put that into perspective. The 2015 Friction Facts study that this is partly based on uses 250W of power, but on the other hand, aero testing is done at 45km/h (which unless you are very short and/or extremely aero, requires significantly more power). So to me it seems that it's very possible that one could upgrade to a slightly more aero bike but negate all those gains and then some if also "upgrading" to AXS.

But of course better and more specific testing should be done to have a clear answer.

cajer
Posts: 783
Joined: Sun Jul 14, 2013 1:26 am

by cajer

I would say 3-4 watts from smaller cogs and more cross chaining on average + 2-3 watts from the AXS chain. A few chain testers and silica have come out saying that the AXS chain is super slow. So on average 4-6 watts and if you're at the smaller end of the cassette (smallest cogs give allot more loss and more cross chaining) you're looking at 6-8 watts.

FlatlandClimber
Posts: 2505
Joined: Tue Mar 31, 2020 3:37 pm

by FlatlandClimber

Why more cross chaining?
Shimano and Campagnolo drive trains have 12 cogs on the same width cassette, yet, all the gear ratios for SRAM are larger (e.g. 50:10 > 53/54:11, 50:11 > 53/54:12, 50:12 > 53/54:13).
So at least in this end of the cassette, the SRAM 12s cassette is actually the one with the LEAST cross chaining at a given speed/ cadence.

People like to act like you need the 50:10 like all the time, when in reality 50:11 is already close to 55:12.

The chain is slow, that's the biggest weakness of SRAM 12s. Durable, but slow.
The small cogs are more inefficient, but probably less than people make it (as the 10t cog is rarely used, unless you undersize your CRs).
Crosschaining really should not be a factor here.
Cervelo P5 Disc (2021) 9.1kg
Factor Ostro Gravel (2023) 8.0kg
S-Works SL8 (2023) 6.3kg

*weights are race ready, size 58/L.
Sold: Venge, S5 Disc, Roubaix Team, Open WI.DE, Émonda, Shiv TT, Crux, Aethos, SL7

Clean39T
Posts: 185
Joined: Sat Oct 22, 2016 5:29 am

by Clean39T

If quiet = smooth = fast… SRAM AXS fails miserably. I have been so much happier being back on Campy or Shimano 11 spd. Even ETap 11 is a relief.


Sent from my iPhone using Tapatalk

Rough
Posts: 57
Joined: Sat Jan 30, 2021 1:21 am

by Rough

FlatlandClimber wrote:
Sun Feb 20, 2022 7:19 pm
Why more cross chaining?
Shimano and Campagnolo drive trains have 12 cogs on the same width cassette, yet, all the gear ratios for SRAM are larger (e.g. 50:10 > 53/54:11, 50:11 > 53/54:12, 50:12 > 53/54:13).
So at least in this end of the cassette, the SRAM 12s cassette is actually the one with the LEAST cross chaining at a given speed/ cadence.

People like to act like you need the 50:10 like all the time, when in reality 50:11 is already close to 55:12.

The chain is slow, that's the biggest weakness of SRAM 12s. Durable, but slow.
The small cogs are more inefficient, but probably less than people make it (as the 10t cog is rarely used, unless you undersize your CRs).
Crosschaining really should not be a factor here.
I'm also interested in this topic as I might buy a bike with SRAM Force AXS drivetrain.

https://cdn-ctstaging.pressidium.com/wp ... igures.jpg
this is the table showing the difference in chain efficiency. I don't know if I'm interpreting this correctly but it looks like SRAM Force AXS chain (Red chain seems to be worse than Force) is slow at the beggining but after some use the difference to Shimano Dura Ace seems to be less than 1-0.5 watts?

Could you than say that Force AXS flattop chain might be actually quite good as it's durable and would enable you to run a more "pristine" chain? As in less worn?
blaugrana wrote:
Sun Feb 20, 2022 5:21 pm
I wouldn't conclude that there is no noticeable difference, though. Kogel estimate between 1 and 6 W of loss (big window, but it has to be since it's mostly just guessing), but then we have to put that into perspective. The 2015 Friction Facts study that this is partly based on uses 250W of power, but on the other hand, aero testing is done at 45km/h (which unless you are very short and/or extremely aero, requires significantly more power). So to me it seems that it's very possible that one could upgrade to a slightly more aero bike but negate all those gains and then some if also "upgrading" to AXS.

But of course better and more specific testing should be done to have a clear answer.
I've heard that number as well but I think 6 watts is a bit much IMO. At first I was also skeptical of SRAM AXS as I've heard a lot of "horror stories" regarding friction loss but now I'm thinking that it's not that bad. I think gear ratios and smaller chainrings enable you to stay in the big chainring more (people that ride AXS said that). Better cadence, less frontal shifting and staying in the big ring might eliminate most of the mechanical efficiency losses.

But I might be wrong and biased as the bike I'm looking to buy comes with AXS and I'm just trying to justify buying that bike. Would definitely like to hear some opinions regarding this topic.

DaveS
Posts: 3944
Joined: Fri Mar 24, 2006 1:26 pm
Location: Loveland Colorado

by DaveS

The force axs chain is the most quiet I've ever used. I even used it on a Campy drivetrain - way less noise.

Pierre86
Posts: 235
Joined: Mon May 13, 2019 3:53 pm

by Pierre86

FlatlandClimber wrote:
Sun Feb 20, 2022 7:19 pm
The chain is slow, that's the biggest weakness of SRAM 12s. Durable, but slow.
The small cogs are more inefficient, but probably less than people make it (as the 10t cog is rarely used, unless you undersize your CRs).
Crosschaining really should not be a factor here.

Yeah SRAM makes great training chains 😂
https://zerofrictioncycling.com.au/prod ... eed-chain/
I'll probably use an eagle chain for everyday riding and swap to a super record for racing on my campy build.

Axs is basically the slowest option with bad chains and increased drivetrain friction, there's no getting around it.
S6 Evo
S5 Aero

cajer
Posts: 783
Joined: Sun Jul 14, 2013 1:26 am

by cajer

FlatlandClimber wrote:
Sun Feb 20, 2022 7:19 pm
Why more cross chaining?
Shimano and Campagnolo drive trains have 12 cogs on the same width cassette, yet, all the gear ratios for SRAM are larger (e.g. 50:10 > 53/54:11, 50:11 > 53/54:12, 50:12 > 53/54:13).
So at least in this end of the cassette, the SRAM 12s cassette is actually the one with the LEAST cross chaining at a given speed/ cadence.

People like to act like you need the 50:10 like all the time, when in reality 50:11 is already close to 55:12.

The chain is slow, that's the biggest weakness of SRAM 12s. Durable, but slow.
The small cogs are more inefficient, but probably less than people make it (as the 10t cog is rarely used, unless you undersize your CRs).
Crosschaining really should not be a factor here.
52/48 (standard shimano/sram big chainrings) is ~8% which is the gap which is about the difference between one cog in the cassette. So this means you need to be shifter one cog further to the right to acheive the same gear ratio meaning more cross chaining.

FlatlandClimber
Posts: 2505
Joined: Tue Mar 31, 2020 3:37 pm

by FlatlandClimber

cajer wrote:
Mon Feb 21, 2022 7:17 am
FlatlandClimber wrote:
Sun Feb 20, 2022 7:19 pm
Why more cross chaining?
Shimano and Campagnolo drive trains have 12 cogs on the same width cassette, yet, all the gear ratios for SRAM are larger (e.g. 50:10 > 53/54:11, 50:11 > 53/54:12, 50:12 > 53/54:13).
So at least in this end of the cassette, the SRAM 12s cassette is actually the one with the LEAST cross chaining at a given speed/ cadence.

People like to act like you need the 50:10 like all the time, when in reality 50:11 is already close to 55:12.

The chain is slow, that's the biggest weakness of SRAM 12s. Durable, but slow.
The small cogs are more inefficient, but probably less than people make it (as the 10t cog is rarely used, unless you undersize your CRs).
Crosschaining really should not be a factor here.
52/48 (standard shimano/sram big chainrings) is ~8% which is the gap which is about the difference between one cog in the cassette. So this means you need to be shifter one cog further to the right to acheive the same gear ratio meaning more cross chaining.
How?
There are 12 cogs on each, and the cogs are about equally spaced for about the same width (we are talking give or take 2mm in total over the whole cassette).
The FIRST COG on the Shimano is an 11, and on the SRAM its a 10. So you are "equally far to the right" with 54:11 as you are with 50:10.
The same goes for all subsequent pairs of gears (54:12 and 50:11, 54:13 and 50:12). The only difference is, that the SRAM gear is always larger (50:10 is 55:11). The same goes for the 48 and 52 (although the 48:12 and 52:13 are the same size), and 46 and 50 chainrings.
At least at the outmost cogs (10 and 11 for SRAM vs 11 and 12 for Shimano), SRAM offers the bigger gear, meaning you are likely to use it less frequently (say you would shift from 50:11 to 50:10 on a SRAM at 95 RPM, then you are already going 55.5kph when you need to shift into 50:10).

So I stand by what I said: LESS CROSSCHAINING at the high end on SRAM for any given speed or cadence.
Cervelo P5 Disc (2021) 9.1kg
Factor Ostro Gravel (2023) 8.0kg
S-Works SL8 (2023) 6.3kg

*weights are race ready, size 58/L.
Sold: Venge, S5 Disc, Roubaix Team, Open WI.DE, Émonda, Shiv TT, Crux, Aethos, SL7

Rough
Posts: 57
Joined: Sat Jan 30, 2021 1:21 am

by Rough

FlatlandClimber wrote:
Mon Feb 21, 2022 7:39 am
How?
There are 12 cogs on each, and the cogs are about equally spaced for about the same width (we are talking give or take 2mm in total over the whole cassette).
The FIRST COG on the Shimano is an 11, and on the SRAM its a 10. So you are "equally far to the right" with 54:11 as you are with 50:10.
The same goes for all subsequent pairs of gears (54:12 and 50:11, 54:13 and 50:12). The only difference is, that the SRAM gear is always larger (50:10 is 55:11). The same goes for the 48 and 52 (although the 48:12 and 52:13 are the same size), and 46 and 50 chainrings.
At least at the outmost cogs (10 and 11 for SRAM vs 11 and 12 for Shimano), SRAM offers the bigger gear, meaning you are likely to use it less frequently (say you would shift from 50:11 to 50:10 on a SRAM at 95 RPM, then you are already going 55.5kph when you need to shift into 50:10).

So I stand by what I said: LESS CROSSCHAINING at the high end on SRAM for any given speed or cadence.
FlatlandClimber from what I read you own/owned some bikes with SRAM AXS drivetrains. Do you think drivetrain efficiency from slower chain is noticable in real life? I know it's impossible to notice 2w of drivetrain loss but what is your opinion on the AXS groupsets in general? Did you experience a lot of reliability issues/dropped chains and such?

FlatlandClimber
Posts: 2505
Joined: Tue Mar 31, 2020 3:37 pm

by FlatlandClimber

3 or 4 total chain drops in 3 years. The 10t is pretty loud, which definitely gives the impression of it being slow. However, I hardly used it though outside of descents and intervals (where I try to hit power over speed, so pedal out of the saddle at over 40kph).
The shifting performance is good, but not quite as good as Shimano, which is faster and feels more controlled.
The gear range is better imo. Efficiency and weight slightly worse. Power meter accuracy is better again.
Cervelo P5 Disc (2021) 9.1kg
Factor Ostro Gravel (2023) 8.0kg
S-Works SL8 (2023) 6.3kg

*weights are race ready, size 58/L.
Sold: Venge, S5 Disc, Roubaix Team, Open WI.DE, Émonda, Shiv TT, Crux, Aethos, SL7

Rough
Posts: 57
Joined: Sat Jan 30, 2021 1:21 am

by Rough

FlatlandClimber wrote:
Mon Feb 21, 2022 10:45 pm
3 or 4 total chain drops in 3 years. The 10t is pretty loud, which definitely gives the impression of it being slow. However, I hardly used it though outside of descents and intervals (where I try to hit power over speed, so pedal out of the saddle at over 40kph).
The shifting performance is good, but not quite as good as Shimano, which is faster and feels more controlled.
The gear range is better imo. Efficiency and weight slightly worse. Power meter accuracy is better again.
Thanks for the reply. Quarq pm is a nice feature.

What about braking performance? Any difference in braking power, feel and modulation?

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



User avatar
eucalyptus
Posts: 614
Joined: Thu Mar 19, 2020 3:51 am
Location: Sweden

by eucalyptus

Rough wrote:
Mon Feb 21, 2022 11:00 pm
FlatlandClimber wrote:
Mon Feb 21, 2022 10:45 pm
3 or 4 total chain drops in 3 years. The 10t is pretty loud, which definitely gives the impression of it being slow. However, I hardly used it though outside of descents and intervals (where I try to hit power over speed, so pedal out of the saddle at over 40kph).
The shifting performance is good, but not quite as good as Shimano, which is faster and feels more controlled.
The gear range is better imo. Efficiency and weight slightly worse. Power meter accuracy is better again.
Thanks for the reply. Quarq pm is a nice feature.

What about braking performance? Any difference in braking power, feel and modulation?
Shimano brakes > SRAM.

I love my SRAM red axs but braking performance/modulation is just shit compared to Shimano. When I borrow a friend's bike with Shimano it is pure joy to use the brakes. I prefer the shifters of AXS, the buttons are soooo smooth but levers are big and chunky with really poor braking performance.

Post Reply