the last time I posted on this.
I installed a Manitou Tower Expert 100mm travel suspension device on the front of the XXIX. Whew! What a relief to my paws and a major boost to the handling that was! A few more rides revealed no more issues with "The Belt". However; the belt tension is so tight that even with my "throwing" the crank arm as hard as I can backwards, the crank will barely spin one revolution. In contrast, all of my chain driven single speeds spin multiple times around at the crank when doing a similar "throw". What does this prove? Well, nothing except that there is excessive drag in the belt system which is either due to the tension of the belt resisting engaging with the cogs, or the side loading of the bearings. Probably some of both.
This tells me two things. One: The tension necessary to maintain proper belt/cog interface during hard efforts also decreases efficiency of the system overall. Two: Premature wear of bearings is a distinct possibility.
Now, as for "The Belt" working in a mountain bike setting, I give the system a passing grade. Yes- it works. However; one must "count the costs", both real and in a design sense, in regards to using the Gates Carbon Belt Drive.
First off, let me say that Gates is working very hard to minimize the drawbacks, and someday, (perhaps), they will achieve lower costs, better design integration, and more choices in gearing options. (I speak strictly in terms of mountain biking/racing mountain bikes) However; this is now, and Gates is not there today with those goals accomplished.
So, to get a belt drive to work on a reasonably priced single speed, we have this example of the XXIX. A bike with some serious compromises to geometry, (long, even for 29"ers, chain stays, weird drive line), and efficiency that is lowered due to tension requirements of the belt itself. Whether this will also affect future drive train life expectancy is question not yet answered, but signs point to earlier failure, (snapping, crunching sounds from the free hub body already at this early stage.)
Another point to consider is the current cost to invest into the system. You obviously will need a frame that is compatible with a belt, (currently a very limited selection), and you may need to swap out to a different belt/cog combination. This will prove to be quite expensive, as the current costs of cogs is in the neighborhood of $110-$135.00 and belts are around $65.00-$70.00 each. (You can't simply change the length of a belt, you will have to have different lengths to accommodate different cog options.)
Belt drive fans will tell you several "advantages" over chains, but only two things they say really matter. (The rest are not really advantages over chains in single speed applications.) One: Belt drive set ups are lighter than chain drive. Two: Belts and cogs will outlast chains and chain rings in terms of lifespan. This is somewhat offset by the fact that the looming early wear issues on the bearings are still out there, the efficiency of the entire drive train suffers, and obviously, in terms of costs.
Conclusions: The short and simple is this: Belts are not better than chain drive systems. Not in terms of single speed mountain biking. Belt drive has some advantages, but costs, unknown bearing wear issues, design constraints in terms of frames, and the limited availability of cogs and belts in all gearing ranges is holding belt drive back. For now "The Chain" is the obvious winner.
Maybe Gates will achieve the goals they have set forth for themselves and premium, high end single speed mountain bikes with belt drive will be something that makes sense. However; today that isn't the case, and most of us will be better served with a chain driven single speed mountain bike.