twocicle

I think it's usually 3.

FWIW, while our current Di2 2x11 drivetrain (48/30) and my custom "same-side-drive" is ok for the gearing range, we prefer to use a closer set of cassette cogs (we dislike super wide cassette steps which jump +2mph/16rpm). To get that, I am thinking over a move to using XTR Di2 triple FD/RD with a 104/64bcd spider and have 48/36/24 rings along with a 11-23 cassette (or 11-28 for mega climbs would equal a typical 30x34t low end ratio. the 48x11 = 52x12 which is fine IMO). To do triple rings while keeping a narrow Q-Factor (a stoker absolute), it will need some tandem cranks with a left-side crossover/timing. Thinking of Lightning cranks for that setup. DaVinci would also work, but I'm not a fan of square taper and tiny inboard bearing. I've been putting this change off for a while, in hope Shimano would come out with more Di2 triple options, but alas, XTR is the only game so far and not hearing any rumors of Ultegra Di2 triple, etc.

twocicle

Finally pulled the trigger to get the Lightning tandem crankset. I actually ordered the "single bike triple" rear which is supposed to have a Q-Factor of 158mm, the max my stoker can handle. Tim @ Lightning assured me 158mm is the outside-outside crankarm measurement.

Our drive triple spider will be the 104/64bcd version for which I'll need to pick up some new rings. Ultimately I hope to run 48/36/24 chainrings if the XTR triple FD can handle that combo. We'll see. Fingers crossed.

Here's something new, at least I haven't seen any posts about it... Lightning has a 9mm hirth hitch spacer available for widening the cranks if that is ever desired, ie: move to cranks to a mtn bike. Nothing we need, just an option I wasn't aware of.

Bad1

Looks like that was a long day in the saddle for you guys, good effort!

twocicle

FWIW, my setup tinkering on our Dragonfly is pretty much done as of yesterday. Finally! It involved some serious hacking, but in a nutshell...

I swapped out the road Di2 derailleurs for a XTR triple front and a long cage rear, plus now using a set of 104/64BCD "trekking 48/36/26" chainrings on Lightning cranks.

The XTR SGS (long cage) RD can handle the biggest cassettes if needed, but we're only using a 11-28 right now and with a 11-32 on hand should be able to handle most anything. We pretty much live in the 20-24mph cruising speed range, and so with the 36t middle chainring we are now back to using to the tighter gaps of the smaller cogs 12,13,14,15,17.

Synchro crow, time to eat my words...

My initial thoughts about synchro mode were "Bah, who needs it?!" Well we rode with it today and are VERY happy how it works. It's really nice having the best sequence of front & rear gears predetermined and not trying to remember whether or not to shift 2 or three cogs for each chainring shift. I created a specific synchro map for the 11-28 cassette and the trekking chainrings using the Gear Calculator I've mentioned previously. This is another big feature of Di2 and IMO Di2 is light years ahead of mechanical shifting.


Saga of Lightning cranks vs tight Q-Factor...

We had a bit of back and forth with Tim and a couple versions of cranks being shipped/returned/shipped. Initially they shipped their standard tandem cranks that have a ~164mm Q and bonded in pedal washers. We eventually ended up with a custom set with our target 158mm Q-Factor and no bonded in washers (Speedplay axle flanges fit within the carbon surround and so no washers are needed to prevent contact/damage). Removal of the bonded in washers allows us to tweak the pedal stance width a little bit more than otherwise. Aesthetically, the driveside crank arms are not as flared out as the timing side which looks a little strange, but we can live with that. Critical issue was bio-mechanics (Q) and this crankset meets those needs.


XTR Di2 triple FD mount hacking...

Moving to the XTR Di2 triple FD imposed some real headaches due to the wide offset they built into the default/standard mounting bracket system. Even at its inward-most setup, the inner cage plate of FD still mashed up against the outer chainring, or if raised enough, would travel too far outward and significantly interfere with the driveside crank arm. The XTR Di2 triple FD was a bust with the standard Shimano "high mount adapter" clamp.

After much thought about options, I had an idea to try the Shimano "direct mount adapter" + 1 Problem Solvers 26mm offset direct mount adapter. This config allowed me to shave off a few mm from the Problem Solvers square chunk of metal, which pulled the FD in toward the seat tube up until the point where the Shimano direct mount adapter just brushed the seat tube. By reducing the standard direct mount offset, this provided just enough clearance (plus maybe 1mm to spare) for FD adjustment. The FD now shifts through all chainrings without any issue and misses our driveside crankarm by maybe 1mm (no contact under any shift scenario).

Here is a photo of the QBP #
FS1323 adapter of which I shaved down the facing...


Problem Solvers Instructions and photos of how the adapter above meshes with direct mount front derailleurs.


Finally the standard Shimano direct mount adapter...


To note, my adapter hacking was only needed because we insisted on a narrow set of cranks with basically a road triple chainline instead of a mtn chainline (wider cranks & wider chainline). Most tandems with standard tandem cranksets (FSA for example) will have enough chainring offset to work fine with the default XTR triple FD offset.

I'll get to snapping some photos of the new setup ASAP. It looks great and other than the little adapter-adapter direct mount kludge, is fairly standard fair.

mtseymour

Although XTR Di2 allows you to toggle between 2 shift maps, were you happy with one shift map? Did you have any problems with synchro shifting under full power? We're keen to try the road version when it comes out.

BTW, good job with the XTR Di2 front derailleur hack.

twocicle

If I were on your bike, then I might try shifting under full power if that is what you wanted tested Otherwise I would not do that on any bike... single or tandem, ever. I've broken too many mtn bike chains doing that under extreme conditions. That said, shifting worked great under all nominal conditions we encountered on this inaugural XTR synchro ride.

One thing that really pleased me about using the synchro function is that since we have now moved back to a triple chainring setup, remembering crossover points is a lot more messy than when using just a double chainring. Then, another nicety is still being able to force a FD shift whenever desired. Even though this was just our first ride with the synchro turned on, I noticed that I spent a lot less time looking at the gear indicator display or thinking about what multishift combo might be needed on the next FD ring shift.

While these components can only remember 2 synchro maps at a time, the E-Tube software does have a feature to export and import config files. So you can work up as many mappings as you like and then save those as config files for later upload to the bike. I don't foresee any need to have >1 mapping per cassette, so as long as we stick to 2 cassettes max, the 2 mapping limit will be enough.

Our only issue so far may be due to a worn chain. Issue is with shifting toward smaller cogs while in the big chainring only. The shifting to smaller cogs sometimes simply chatters and is slow to drop into the next cog. I have the RD in/out otherwise adjusted perfectly so that in the middle and granny chainrings, the shifting works great in all cogs (where allowed... ie: no excessive cross-chaining).

I think the less precise shifting while only in the big ring is due to the RD cage angulation forward. In this case, the top pulley moves a big further away from the cogs and so that creates a sizable gap between the pulley and cog teeth. A bigger gap = more chain links between the pulley and cog teeth. I have the B-Axle screw adjusted all the way out, which puts the top pulley as close to the cogs as possible, but I would like to see it remain closer yet. The chain length being used is per the standard big ring -> biggest cog, +2 links. This has worked fine with the previous Ultegra Di2 long cage RD. One more little point is that we are still using Ultegra 6800 11spd chain and 6800 cassette, not HGX 11spd.

With a somewhat worn chain being used presently, side-to-side chain slop is apparent and therefore IMO the Achilles heel. I'll try two things to fix this... 1) new chain, and 2) experiment with adding a couple more links in order to keep the RD cage from pulling forward as much while in the big ring.

twocicle

Unique setup features:

1) modified Shimano Direct Mount and Problem Solvers Direct Mount Adapter to pull the FD inboard by ~3mm.

2) timing side Specialized S-Works 130BCD carbon spiders. These spiders were needed to center the spider offset for the somewhat fat Gates timing rings/belt. Spiders are mounted in reverse (inside-out) to get the perfect offset. Rotation direction matches normal driveside, so loads are per directional spec design for the spiders. Lightning provided 2 x 130/74bcd spiders, but the offsets just didn't work out on those because the timing ring was either too close to the crank arms or too far inboard and causing the belt to rub the lower stoker waterbottle cage (frame clearance was not a problem).

Unfortunately the special S-Works chainring bolts are a bit too short to work with the Race Face Crank Arm Tab Spacers I like to use on the back side of the timing spider tabs. Switched back to using just 2mm chainring spacers/washers.

Items #1 & 2 above are due to special narrow Q-factor crankset needs. Probably non-factors for regular-wide tandem cranksets.

3) 48/36/26 chainrings and chainguard taken from a Shimano Deore XT FC-T781 Crankset. 104/64BCD 4-arm Lightning spider for the driveside. I had some reservations about using the optional chainguard, but stoker and some friends seem to think it looks great so leaving it on and swallowing the few extra grams. It does have some functional purposes such as stoker safety and keeping the chain on in case of over-shift. When messing with FD adjustment, it's definitely a lot easier to see things without the chainguard (installs with 4 simple screws on the backside of the big ring).

4) bike is still sans-aero bars, but ready to mount them (incl Di2 TT shifters). This is why I have the XTR display is on the "wrong" (left) side, as my 3T clipon mounts slip right behind the display and I can still see the display while using the arm rests.

mtseymour

It looks like that you're using the Ultegra ST-6870 road shifters (2x11) with the XTR Di2 (3x11) drivetrain. Did you customize the synchro setting so that only one shifter (eg. R only) handles the upshifts and downshifts? Or did you use both shifters (eg. L for for upshift, and R for downshift)?

The 2nd option is similar to eTap, and remove the likelihood of hitting the wrong paddle (especially when wearing full-finger gloves).

twocicle

These shifters are the "non-branded" Shimano R785 hydraulic version, not the 6870 mechanical brake version. In either case, no I didn't even consider changing the lever behavior. I like the option to shift either FD or RD up/down even while in synchro mode, and so want all four paddles to work as normal in this case.

I've ridden with full gloves in two winters so far and had no problem with hitting the paddles. Even if that did happen occasionally, I probably still would not disable my shifting options.

Pretty sure the Ultegra ST-6870 road shifters are not designated as 2x11. As you know, Di2 doesn't care whether you use the shifters for 1x, 2x or 3x. The shifters, disc calipers and front rotor are the last thing on my upgrade to-do list... once 2017 Dura Ace hydro Di2 hits the streets at a good price point. I'm pretty sure the new DA 160mm rotor will be fine for the front, as the current 180mm Freeza has more power than needed and the DA version is supposed to have a lot higher cooling capacity. The DA 160mm should balance things out nicely and be a touch more aero. Hoping the DA caliper pads are somewhat standard to other models, otherwise if only used on the DA their price point could be stupid high.

Turbotandem

Nice work. Thoughtful, you've been reading a few forums!
Agreed on the synchro shifting.
I have the same shift lag at cog 7 and 8, maybe the same wide gap to the pulley wheel at that point. It would be easy enough for the software to program adjustments at each shift rather than the single adjustment.
Sorry if I missed it in the thread, Are those 55mm deep rims the 50 or 60 from Light Bicycle site? So you like them, are they intended to hold is as your daily riding wheels?
The out-of-production sworks Carbon spider is hard to procure, especially NOS.
I am sure you're aware of the ice rotor alum core melting. Ours on a simple 1 mile 10% grade with a stop sign at the bottom. With organic pads the temp of the rotor and temp of the caliper will not be equal. Fair warning, even more so if your trekking takes you to Europe or other winding steep descents.
Thanks for all the details posted.

twocicle

The carbon rims are working out great. I can't determine how aero/slick they are, but they have been rock solid. I got them with the brake track just in case we wanted to change from disc brakes to rim brakes for various reasons. My initial wheel build info is in post #15 of this thread.

Here's a quick link to the carbon rim model on LB website:
55mm deep carbon 700C 25mm wide road rim clincher U shape tubeless compatible high TG resin surface

I ran the 203mm ICE tech rotor for a year or so with a rear TRP Spyre caliper and no problems for us (mostly used semi-metalic pads). Once moving to the full front/rear hydro brakes we went with 180mm RT99 Freeza rotors. For extreme descents, the rotors did just fine in Mallorca this last April. There are some pretty serious pitches (see post #45) with super tight switchbacks. No issues with the rotors. Some of that success may be due to using the r785/XT calipers and only the semi-metallic pads. There was heat test done by a guy in Colorado that showed the semi-metallic pads helped to keep the rotor temp lower. I think I posted a link to that info somewhere in this forum, so some hard data is there for you to consider. The 2017 Dura Ace 160mm rotor looks interesting for our front brake and that size should balance nicely with the rear 180mm Freeza.

Regarding the S-Works 130bcd spiders for the timing side, I took a chance and . Seller has some 14k positive ratings. These seem legit and came with the correct chainring bolts in what looks to be legit packaging. Also got a S-Works SL seatpost in the same order. Always leery of bogus knockoffs, but I couldn't resist even though my LBS offers a good team discount on Specialized gear. AFAIK, these road spiders are still in production.

Turbotandem

For sure semi metallic pads transfer the heat to the caliper and oil, compared to organic pads. I have trouble with semi metallic squeezing as I use the disc for a lot of drag brake function in the rockies. And I worry about boiling the reservoir so am on organic pads.
I've not seen the sworks spiders road or mountain new. They are an old model with the replaceable crank fitting that's no longer in use by specialized afaik. I've gotten all my sworks spiders eBay both carbon and the 20g more alum.

twocicle

I just checked the Specialized website. The have both the 130 and 110 bcd carbon spiders for sale direct from them. https://www.specialized.com/us/en/co...-spider/116802 . Noticed the photo posted there shows an all carbon spider, even where it interfaces the mounting spline. The two I bought off eBay have a metal spline interface. Not sure if these are an older model or what.

Ok, well if you use the disc for a drag... you should use a solid rotor not the ICE. Try the Hope V2 floating disc (not the newer "saw blade")? I have few I played with, but decided to stick with the Shimano rotors after all. Could work out a good deal if you want them. PM me if interested (anyone).

mtseymour

I hope I don't take your comments out of context because disc brakes are not designed to be "dragged" on a long descent. The proper technique is to modulate F&R brakes (squeeze firmly and release) to control speed and to keep the rotors and calipers from over-heating. This technique applies to rim or disc brakes.

There's been exaggerated posts about "melting" Ice-Tech rotors or calipers when the real culprit was poor technique. One example is a team descending Mt Ventoux by dragging their rear 203mm Ice-Tech rotor. Apparently, their guide suggested that they use their front brake sparingly in case their rear rotor over-heated. Not surprisingly, their rear Ice-Tech rotor overheated and became warped. They were able to finish their descent into Malaucène once the rotor cooled and they modulated the F & R brake normally. To put it into perspective, not even motorcycles or cars should "drag" their brakes on a long descent.

Turbotandem

Of course one modulates, but there is only so much heat dissipation capacity between the two brakes on steep or long descents; both rotors get plenty hot. I am one of those several people who has melted an ice rotor, so I am not just quoting/exaggerating, and I can validate that it was not due to poor technique, thank you very much. In my case, I was thrilled with the 200mm ice rotor for a 9,000 mile, 350,000 foot tandem season; including plenty of "drag braking" to scrub speed on long descents, or in the rain, etc. This is serving a 310# team including the bike and without any touring packs. Where the rotor actually failed was a simple 4-minute, 0.7 mile descent on 10-15% grade with an unfortunate stop sign at the bottom: https://www.strava.com/activities/16...ysis/1073/1192. Pictures have been posted of the resulting liquefied/melted/softened alum core and potato-chipped rotor. IMHO a single rotor ought endure a .7mi four-minute descent. I was indeed thinking in those short 4 minutes not to risk running the stop sign (cross traffic), so scrub some speed on the back and then brake hard with the front when I got near the intersection. The back failed within a few feet of the stop sign. Modulating or not, that's not enough minutes to cool the rotor significantly. With organic pads, that melted alum means the rotor approached 1,200 degrees F.

Anyway, when I see great builds relying on 185 or 160 ice rotors, I'm inclined to give those teams a heads up depending on the kind of riding they do. Twocycle is an equally light team, say 300# with the bike, and fast descender. So they may be OK. Metal pads will help if the calipers are able to keep up with cooling and not boiling the hydraulic fluid. That fluid boils at 500-600. Given my 1,200 degree experience on a short descent, twocycle's set-up is seeking a balance between heat generated on the rotor not to exceed 1,000 degrees or risk melting the alum core, organic or semi-metallic pads to insulate or transfer heat to the caliper, and the cooling capacity of the caliper to stay under 500 degrees or risk boiling the fluid. And at the same time they are seeking to limit the rotor size from 200 down to 160 for aesthetic or aerodynamic reasons reducing the cooling capacity by at least another 25%. If I were balancing that equation, I would say that the aerodynamic advantage of a 160mm rotor over 185 or 200, on a tandem with two bodies and four legs, would be infinitesimally small gains (the frontal area difference between those two rotors is 25sq mm?); but that's me.

It's a tough equation to get right for each team, terrain, and use.

twocicle

We very highly tested the 180mm RT99 rotors in Mallorca where we blasted numerous descents up to 45mph and then slamming on the brakes down to 10-15mph for really tight turns with traffic (bikes and cars) coming up the other side on narrow roads. Usually I'll just let it run and use wind braking (sitting up) to hit a terminal velocity. Not an option on > 10% with tight conditions and oncoming traffic.

There is a little bit of mixing apples vs oranges on the rotor downsizing topic...

The new 2017 DA rotor...


So, I believe moving to the 160mm rotor on the front will likely maintain our current heat generation, while at the same time balancing the braking power front/rear. These hydraulic brakes are really powerful and so I feel a 180mm on the front is too much power available and could needlessly mess up the carbon fork if jammed on harshly to the max. Aero and aesthetics are a minimal bonus and a very minor part of my reasoning. My main interest is in tuning our braking for an optimal balance. The DA rotor might provide that with a 160mm size.

If this rotor performs well on the front in a 160mm size, then that could open up more fork options too. Some forks are spec'd to a max of 160mm. If ENVE comes out with a 367mm A-C thru-axle fork version with > 43mm rake, I'd love to try it out.

Turbotandem

Your results will be interesting. For a skilled descender the most problematic braking case may be when stuck behind traffic, or maybe in the rain when slower speeds are requisite. For my home state of Colorado at 8,000 to 12,000 feet the thin air seems to provide minimal air braking. Sometimes I feel like I am falling out of a plane with no terminal velocity. For less skilled deacenders, like myself, and at high altitudes, braking is more constant. Another tough to predict part of the equation when specing a bike.
FYI, our whisky 9 cross fork (396ac) specs at 185 but clears a 203 rotor. Until we changed to a custom fork. Mathematically for the same deceleration rate, the force on the fork is the same with any rotor, but you're right a larger rotor "can" induce more fork load and faster deceleration if pressed in to action.

waynesulak

Thanks for posting all the details and pictures.

carhillclimb

Thanks twocycle for pointing me to this thread. I had not known about the problem solvers front derailleur mount adapter. Maybe I should have kept the XTR di2 derailleurs.

Any particular reason for using a belt instead of chain for the timing?

And why the specialized spiders on the timing side instead of lightning? My scale indicates that the lightning spiders are light, but I do not have a 130bcd spider. I have a lightning 94bcd spider on my Specialized cranks on my Kona Zone road bike (with a custom 26-46 chainring set using my custom crank bolts that allow a 3 tooth smaller sprocket). And a lightning 74bcd spider that only fits the lightning crank - the bolt circle interferes with the specialized crank arm.

carhillclimb

Turbotandem writes:
Mathematically for the same deceleration rate, the force on the fork is the same with any rotor,
--
Actually the force of the caliper pushing on the fork is inversely proportional to the diameter of the center of the rotor contact patch. The bigger rotor is acting further out on the fork so more fork is subject to distortion, but the caliper mount should be designed to accommodate the forces over its entire range. This is not saying that it is bad to tune the brake balance with rotor sizing.
---
I noticed Continental GP4000SII in a much earlier part of this thread. The GP4000SII 700x28 is actually about 700x31. I use them on my Kona Zone. They do not fit a lot of my bikes where a GP4Season 700x28 (which is more like a 700x27) does. I have a chart of tires started after I tried to put the GP4000SII 700x28 on my BMC TMR01 based on the GP4Season fitting fine.

twocicle

We have a special alignment case due to this set of cranks being made a bit more narrow than usual. I only used the Specialized S-Works spiders (flipped reverse) to achieve the optimal Gate ring position - roughly centered between the Lightning crank arm and the chainstay. The Lightning triple spiders that they recommend for the timing side, set the Gates rings either too close to the crank arm, or too far inboard to where the belt would rub one of the water bottle cages. I didn't want to try flipping the Lightning spiders because they do not have the flange recess on both sides and so I'm not sure if they would engage well with the crank arms. Weight is pretty much equal so that was not a factor. The Lightning double spiders are not beefy enough for tandem use.

twocicle

We are using the GP4000SII 25mm tires, which when mounted on the 25mm carbon rims measure something like 28.7mm wide @ 107psi. The 28mm GP4000 might be too big to fit in our Whisky fork. Like you said, we also use the 28mm 4-Season on another set of semi-wide rims (23mm) but these only measure 28mm. So the 25mm GP4k win the fat tire test.

I don't have mileage numbers handy at the moment, but the GP4k 25mm has been lasting very well. The rear has worn down the middle and showing some squareness. No threads or cuts showing at all. I'm getting antsy to break out a new tire for the rear and not wait for some failure to occur.

carhillclimb

Worn chains notwithstanding, I have had some success solving index shifting issues by using smaller or larger cassette spacers as needed. On 11 speed cassettes a 2mm freewheel spacer is the smaller spacer, and larger is from a soda can shim as suggested by Alex Wetmore. Mark Stonich has suggested changing the top jockey pulley on index shifting to have no side play, and that also seems to help my cable setups although I have not done that for di2. And for custom cassette ratios like an 18 between your 17 and 19, a dremel to rotate the cogs as needed and add the chain contours also works wonders, as well as many different cassettes so you can pick cogs that might work together better. 105 is your friend for custom because the cogs are individual.

I have not studied chain brand and model vs shifting performance. I can say that the Yaban all-titanium chain and the Recon all-aluminum 11 speed cassette on the sub-11pound ioDupont cable shift road bike is pretty marginal.

Turbotandem

Understood. But I was careful to say "for the same deceleration rate", not for the same force applied to the caliper. If one wants to decelerate, say approaching a stop light going from 20 to zero in 500 feet, the strain on the fork is the same regardless of the rotor size (assuming the caliper is mounted at the same point on both forks with just spacers out to the rotor edge. That is because the caliper pressure on a larger rotor need apply less pressure, thus less force on the larger radius rotor. A small rotor on the other hand has less leverage on the wheel so applies more force for the same deceleration; but in the end applies the SAME force on the fork. If you do the math, the force on the fork works out the same, the radius in a way cancels itself out. That said, if you apply equal braking power at the lever and caliper, you CAN decelerate faster with a larger rotor, such as coming to a stop in 300 feet instead of 500 feet, and that would impart more bending on the fork. Many forks only rate themselves to 185mm, so putting a 203mm rotor on, as we do on mtn tandems, can be over spec based on this thinking; in case one does brake real hard.

twocicle

The XTR RD shifting seems to have settled in somewhat, a wee bit better with a new chain but not by much. The precision may not be quite as precise and immediate as was the Ultegra RD. This is more of a RD issue than chain or cassette.

The solution would be to somehow get the top jockey wheel close to the small cogs, more than what the b screw allows. I can't see a larger jockey wheel being a good choice as that could impact the clearance for the biggest cog. Might look at opening up the b screw linkage by using a little judicious filing.