[drussell]

So, I picked up the Delta CX when Airborne had them marked down a little while ago. It did fine commuting on paved roads/mild offroad. Yesterday I did one of our spring races up here in Canada, and found a huge issue with the Apex (and the Rival and Force seem to be the same/similar design). The cable, after exiting the housing from the seatstay to the derailleur is routed through a groove on the bottom of the RD before it gets locked down with the bolt.

2 minutes into the race, I bunnyhopped a pothole, and the bounce on landing caused the RD to pivot, allowing the cable to come out of the groove and preventing fully functional shifting. It happened again towards the end of the race after hitting some rough ground. If you're in a high gear (11 or 12 tooth ring on the cassette), the RD can actually move over and jam the chain between the cassette and frame (probably because I had my limit screw backed out, as the cable was preventing it from moving too far down during normal operation).

It's an easy field fix without tools, but not something you want, or should ever have to deal with in a race. Anyone else encounter it? Do the next level up have an improved design (eg deeper groove?) to prevent the cable from popping out? The Rival looks like it's pretty much the same, but it's a little hard to tell from the manufacturer photos.

 

[locobaylor]

I have the x9 10 speed der on my cross bike, and above the groove is a bridge-type thing that would prevent what happened to you. that might be an option for you.

 

[drussell]

Thanks - I'll take a look in the next day or so. I wonder if they may have put that on the X9 since it's an MTB RD and rough terrain and bouncing is expected, whereas the Apex is a road component stuck onto a CX bike.

 

[nOOky]

Hmmm I have the same bike and components, and I haven't had that problem. If there's no slack in the cable when it's on the smallest cog allowing the cable to move around, I don't know what could be going on with it without seeing it.

 

[AndreyT]

It's an easy field fix without tools, but not something you want, or should ever have to deal with in a race. Anyone else encounter it? Do the next level up have an improved design (eg deeper groove?) to prevent the cable from popping out?

Well, the limit screw issue is obviously something that is entirely your fault. If wonder if the cable issue will reappear if you set your limit screw properly.

I have OEM Apex RD on a 2011 Specialized TriCross Elite. So, I presume that is the 2011 version of Apex RD, if it ever changed. The cable exits the housing and goes into the groove, and then to the pinch screw, passing above the pinch screw axis. The groove opens downwards. When the cable follows that route it is noticeably bent as it goes through the groove, i.e. the location of the housing exit and the pinch screw is intentionally chosen to make the lateral stiffness of the cable to force it into the groove and keep it there.

I don't see how it can possibly get out of the groove under these circumstances, unless there's an enormous amount of slack in the cable.

Could it be that in your case the cable is pinched below the pinch screw axis (which would be incorrect)? In such configuration, I imagine, the cable would be a lot less encouraged to stay inside the groove.

For example, the following picture (from here) demonstrates a major error in RD installation: the cable is pinched below the screw axis.

View attachment 279315

(So much for Kansas Bob's project bike).

The proper way to pinch the cable is shown here

View attachment 279316

 

[drussell]

Dead to rights on the limit screw. I backed it out to set the tension on the cable, but neglected to put it back in. I don't know if that would solve my problem though, as the second time it happened I was in a middle gear, so it wouldn't have come into play.

The cable is pinched on the upper (outboard) side of the anchor bolt, though - it's pretty apparent that's where it goes (by the groove in the washer/plate).

Cable tension is a possible issue. I suppose first step is to start setup from step one and see how that goes.

I don't see how that helps with my upgraditis, though. I mean, all technical issues can be solved with new parts, can't they?

 

[minras]

Hello,

My Specialized crux has the same problem with apex RD and I have replaced it with Sram 10sp X7, maybe heavier but currently works well without any problems.

 

[6_Myles]

Got a Delta as well, and haven't experienced this in either race or mtb'ing...but I would suspect that you've got a bit too much slack in the cable on the smaller gears. Something to keep an eye out for though!

 

[Cyclo-phile]

I think some of the previous responses have it all wrong. Derailleur position is a function of cable tension, so if all of your shift positions align correctly the issue isn't cable tension. The issue was caused because there wasn't enough tension on the derailleur to counteract the force generated by your bunny hop. Therefore, I suspect your real problem is a chain that is too long.

 

[TheRealKTrain]

I think some of the previous responses have it all wrong. Derailleur position is a function of cable tension, so if all of your shift positions align correctly the issue isn't cable tension. The issue was caused because there wasn't enough tension on the derailleur to counteract the force generated by your bunny hop. Therefore, I suspect your real problem is a chain that is too long.

This response makes the most sense.

 

[6_Myles]

Were you cross-chained at the time? Small to small?

 

[drussell]

Were you cross-chained at the time? Small to small?

I don't recall now, but it is possible. If the chain is too long, then Airborne ought to be checking more carefully - I haven't touched that since it came out of the box.

My shift positions were good, with no issues moving throughout the range - I had spent some time the day before making sure of that. Which does bring me back to the limit screw.

Honestly, I haven't been back out in the garage to work on it since my original post. MTB race season started shortly after, so I was getting that bike set up, spending some time on my road bike, and I'm now out with an injury so I'll have to take some time with the Delta this summer and try to figure it out. I appreciate the feedback on places to chase the problem down, and knowing that it doesn't seem to be a common problem tells me that it's probably somewhere in the setup.

 

[Cyclo-phile]

Your first mistake is assuming that a mail order OEM built the bike correctly.

Tech Tuesday - Chain Length Basics - Pinkbike

 

[AndreyT]

The issue was caused because there wasn't enough tension on the derailleur to counteract the force generated by your bunny hop. Therefore, I suspect your real problem is a chain that is too long.

This makes no sense at all. It the derailleur cage tension spring is working properly, then the amount of torque that spring applies to the derailleur cage is almost constant across the entire cage swing range (i.e. for all forward-backward positions of the cage). In other words, chain tension is virtually independent from the chain length, as long as it stays in the operable range (i.e. as long as it is not bumping into the RD "slack pickup" limit).

 

[Cyclo-phile]

This makes no sense at all. It the derailleur cage tension spring is working properly, then the amount of torque that spring applies to the derailleur cage is almost constant across the entire cage swing range (i.e. for all forward-backward positions of the cage). In other words, chain tension is virtually independent from the chain length, as long as it stays in the operable range (i.e. as long as it is not bumping into the RD "slack pickup" limit).

You've never done much mechanical design, have you? The cage is tensioned by a helical torsion spring. The force generated by the spring is directly proportional to the angle of deflection.

 

[AndreyT]

You've never done much mechanical design, have you? The cage is tensioned by a helical torsion spring. The force generated by the spring is directly proportional to the angle of deflection.

Quite the opposite. "Mechanical design" is my middle name.

Helical torsion spring is exactly the point here. The torque, as you said it yourself, is directly proportional to the angle of deflection. However, the absolute difference in torque for the given range of deflection angles depends on how much the spring is already pre-tensioned (i.e. how much it is wound up in its most "relaxed" position).

If the spring is not pre-tensioned at all, the difference is maximal. The more you pre-tension the spring, the smaller the absolute difference becomes (again, for the given fixed range of deflections).

The helical torsion spring in a RD is always pre-tensioned. (Anyone who ever disassembled/reassembled a RD knows that). The reason for that pre-tensioning is, among other things, to reduce the absolute difference in spring torque between different positions of the RD cage. While the torque is indeed not constant (and I never claimed that), the variation that can be produced by changing the chain length is sufficiently small to make the "chain length -> chain tension" relationship essentially negligible.

In fact, in virtually all helical torsion spring applications, it is typically installed in pre-tensioned state. This is done to "offset" the operational range of the spring to a more "constant" area, i.e. to reduce the absolute variation of torque.