The Death of 1x Drivetrains Revisited

Just revisiting this VeloNews article with maybe a little different nuance. In case you missed it, here is the money shot:

This takes me way, WAY back to when I first got the cross-chaining dataset from Jason. Then as now there was some banter back and forth but the thing that stuck in my head was, ‘Hey, bigger gears are more efficient in a chain driven system.’ When I say bigger gears I mean physically bigger cogs and chainrings. Add up the teeth on the cog plus the teeth on the gear. More teeth=bigger=more efficient. And not just a little bit, either. Materially more efficient.

Look at the chart above. Consider the red 48x18 circle and the blue 53x19 circle. That’s the gear combination that the chain line was adjusted to for this data set…so both the 48x18 data point & the 53x19 data point are taken in a condition that minimized cross-chaining friction losses. There is about 2W difference between those two points…for pretty much the same gear ratio. 2W from the drive train might not sound like much but you’ll pay CeramicSpeed $500 for an oversized pulley system that will give you less benefit than that.

I spent many evenings futzing around with a 1x 60t chainring and miche jr franken cassettes trying to get big chainring/big cassettes to work together but could never find a derailleur/shifter pair with enough chainwrap and the right index to get it to work. If anybody has ever heard of such a setup that works or has actually implemented such a thing…let me know: we’ll talk shop.

5 Likes

2W on a 250W base (per the test conditions in the velonews article) is a little less than 1%.

The link below reports a 2% reduction in aeordynamic drag from a 1x vs a 2x, which offsets to some extent the above.

The reason for 1x I think is most compelling in MTB and CX where the drivetrain simplicity makes a dropped chain less likely (and cognitive load easier for shifting). Plus cleans up the MTB cockpit when you have a dropper. And watt penalty (or savings) is less of a concern vs these other benefits.

9 Likes

@DaveWh good points! Add to that drivetrain (in)efficiency varies with rider power linearly while drag varies with the cube of velocity.

The asymmetrical drag/yaw data was interesting! Very cool…

The 2w loss is only in specific gear combinations. The 2w is easily recovered in aero advantage by cleaning up the front derailleur area. As for dropped chain, I’ve dropped my chain far more times on a 2X than the 10s of thousands of miles on a 1X.

On top of that, to put 2w into perspective, it’s about the same as the amount of drag lost/gained by taping or not taping top of aero handlebars.

Other than for the most absolute top athletes who pay their mortgage and kid’s college fund w/ their contracts and winnings, it doesn’t matter.

3 Likes

I don’t understand why they haven’t compared SRAM 1x to SRAM 2x here. That would have been a very good way to see whether some losses are due to e. g. differences in chains and cogs, having a smaller 10-tooth cog and separate that out from differences in the chain line.

1 Like

Of the dozens of people I know who have gone 1x, not a single one of them is concerned with drivetrain “frictional losses”.

Reasons for switching include:

  • It’s something different.
  • It’s cool.
  • It’s simpler.
  • Fewer dropped chains.
4 Likes

What is this “two by” one speaks of? What are these watt things too, while we’re chatting…

4 Likes

@BMAC615, summarize the benefits of 1x as you see them…

@KickrLin I switched to 1x to reduce the opportunity for mechanical failure & it has certainly done that. Plus I can get up the hills in my area well enough that a 2x drive train just wasn’t required.

The downside for me is the occasional dropped chain when I hit a bump at while going fast. So let’s say at the bottom of a hill. In that case I’m in a smaller cog so the tension on the chain is reduced & I’m running a non-clutched rear derailleur.

Within the context of this discussion the folks who are fixating on the 2W figure are probably allowing cognitive dissonance to get the best of them. The difference is much more than 2W. I just picked the 2W figure because that’s where both drive chains had an aligned chain line. As you go faster & faster the mechanical disadvantage of the 1x system becomes more pronounced…which I shouldn’t have to tell folks who have made it this far into the thread because I posted the data at the top.

Just the same, I like to hear the pro argument from die hard 1x proponents…their reading comprehension difficulties notwithstanding! :wink: And I especially enjoyed the 1x wind tunnel data. That was cool.

A seven watt differential in 50/11 vs 46/10 when I’m pushing 1000w+ vs 300w+ in 36/28 vs 46/33 is seriously negligible. It’s not going to be the difference in whether I win or lose a race.

I’ve dropped chains off the front in races that cost me the opportunity to win. Regardless, my life and bank account do not depend on my placing at races.

I rode a 1X 50/11-32 all season two years ago and 50/14-42 all season last year as I took the year off and mostly did easy rides. The 50/11-32 was not enough for road races and training rides, but, plenty for crits. There are lots of Juniors who can smoke me and most Cat 3+ riders on Jr gearing. I wasn’t competing, so, it really didn’t limit me in any significant way. Plus, it was good training Never dropped a chain - not once.

I also ride a fixed gear track bike a lot in my training. Riding 100 miles solo holding 225w for 5 hours on a fixie w/ 48/15 is good for the mind and body. I did BWR on a 1X CX bike w/ 40/11-30 and it was fine.

Benefits are mainly simplicity and my opinion is that, at this point, 2X is superfluous for most riders (including me) just like I ride a 1X MTB and not a 3X. Like most of the decisions I make, the thing that is most important is what is most bad ass. Like my red shoes. They are bad ass and much better than black shoes. Black shoes are for people who don’t know what’s up. 1X is very much more bad ass than 2X.

5 Likes

Red FTW!! Seriously, I thought everyone knew red shoes were not only bad ass but faster, good for like 50W easily overcoming any 1x drivetrain inefficiencies! :smile:

3 Likes

:rofl: You’re engaging in a lot of whataboutism! But that’s ok. Look, I don’t care if your FTP is 200 or 300. Or if you ride with super fast jrs. I’m interested in what you LIKE about 1x. You don’t have to prove to me that you’re fast or experienced or that you ride with fast folks before your opinion has value.

I think your reasoning is mostly qualitative & that’s ok! The front derailleur/shifter is too complex for you. I agree. One of the reasons I switched some of my bikes to 1x was to eliminate opportunity for error ( which is a more quantifiable approach to your ‘too much complexity’ qualitative outlook). That is definitely true.

Reduced the weight on my MTB by around .5kg I’d call that beneficial. Although that was 3x to 1x

Quantitatively, it’s more aero and aero is everything.

It’s lighter and I need every advantage I can get so I can be the first to the top of the big climb on my Saturday Club ride, so, when I’m coasting down the other side, everyone knows who the big dog is.

It’s less expensive because saving a couple hundred dollars on my $10k bicycle purchase is really important to me.

If I drop my chain on the Sunday Club ride, everyone will grumble and talk for several minutes about how I almost took a whole bunch of people out.

I’d never go back to 2x on my MTB.

For road, 2x is fine, maybe even preferable given the smaller steps between gears.

I’m considering a 1x for a gravel bike - which to me is about the “crossover” point where 1x becomes more attractive.

2 Likes

The thing that I really hate about 2x is the annoying double shifts (e.g. having to shift up/down the cassette when you change front ring - either once or twice depending on what bike and gearing I’m on).

I’m sure that the lost momentum from this = lost watts as well, and if Friction Facts did real-world testing the difference between 1x and 2x would be virtually zero for 99% of riders, certainly at amateur level.

I think 2x is a good choice for those (special snowflakes) that don’t want more than 1t jumps between gears for their type of riding. But for me most of my riding is endurance gravel and road and the benefits and simplicity of 1x is perfect for this.

For road I’m a big proponent of 2x and 1t jumps :smile: I’m all about high cadence and spinning up steep climbs. I almost never drop chains and enjoy the multiple shifts, reminds me of driving stick shift. If it ain’t broke… Also out of sheer vanity there will be no pie plates on my road bike and I would need one for the mountain climbs.

I do prefer 1x on my gravel bike:

  • simple: no multiple shifts so I can focus on my line choice and terrain. I’m new to off-road and take my gravel bike on everything including single track.
  • reliable: no dropped chains or chain rattle with a clutch derailleur. my mechanic warned me against 2x on Open UP, several clients had problems with dropped chains.
  • something different, a new toy, a way to step outside my normal “spinning” style
1 Like

Just kicking the wind tunnel 1x vs 2x data around in my head a little bit more today. I wish they would have done the baseline vs 1x comparison AT 20MPH instead of just the baseline. But I think the fact that they didn’t show such data sort of leads to its own conclusion. Anyhow, it’s interesting that the bike did stall at higher yaw angles at 20MPH, but the +ive/-ive yaw angle asymmetry did persist a little bit!

image

Note how the pronounced drag difference a yaw really collapses when you’re at a more intermediate speed! It’s reasonable to conclude that the 1x/2x drag difference also collapses. It’s probably de minimus at positive yaw angles (you might argue it was de minimus even at 30MPH!).

image

Drag difference collapses at lower speeds because (as we all know) drag varies with the cube of velocity. Drive train frictional losses, however, are linear with speed. As Jason has conclusively demonstrated in the past, drive train efficiency is also linear with rider output.

image

So as you slow down frictional losses start to dominate drag disadvantage. Even though drag disadvantage likely still persists.

If I consider three cases of i.) flat 40k TT, ii.) flat ultra tt, iii.) flat gravel ride which probably average ~27mph, 18mph-20mph, and ~16mph…then use www.machars.net/bikecalc.htm to pick out what gearing would propel me along at that speed with 90rpm cadence it looks like frictional drive train losses (1W loss vs 2x) would be about:

i.) 4.5W
ii.) 2.5W
iii.) 2W

If the course was flat and if I was confident there would be no mechanical issue with the 2x drive train I would be inclined to select 2x in the latter two cases and 1x in the former. But if it’s hilly 40k TT course I would still go with the 2x.

On the other hand! IF I could have infinite choice of chainring & cassette along with a derailleur that would shift it I’d probably pick 1x in all three cases. If you can make the gears big enough the 1x frictional disadvantage turns into an advantage. There just don’t exist the gear/derailleur combination that will do it.

2 Likes

And, of course, to all my ultra-distance competitors reading this: I want to strongly confirm that 2 watts DO NOT MATTER as some on this forum have asserted.

So just forget about them…

1 Like

Interesting Stuff.

My understanding of the aero benefits of 1x is that the saving is 2 Watts at 30mph (around 15 grams of drag for ±5 degrees yaw according to the article linked further up this thread).

If you have a very good CDA of 0.222 (Tom Dumoulin during his Giro winning TT) then you’ll be putting out 375 Watts to carry that speed and be in the 53-13 ratio on 2x or 48-12 on the 1x setup with a cadence of 93 rpm.

In these gears at 250 watts the losses are 11.5 Watts vs 16.2 Watts a difference of 4.7 Watts. Assuming that the drivetrain losses scale linearly with power then the 4.7 Watt difference becomes 7 Watts.

Here’s a plot of the power losses versus speed:

I’d like to see some numbers on Sram AXS…

Mike

2 Likes