SRM spider out of true?

What would you do with a recent used SRM purchase (2008 Dura-Ace), where the spider is out of true by about 2.5mm? Two of the spider arms are further out than the other three (~2.5mm at the bolts).

The original owner can work a warranty if needed. Is this something that can be user/shop adjusted at the SRM/crank interface? It's unclear if the spider itself is bent or if there's just an alignment issue.

Before you send the crank back for a warranty replacement:

a) try a different chainring that you know is "true" ...

b) try putting the crank in a different frame (this is unlikely to show anything, but you never know!).

I'd say that if the spider is out of true without a chainring attached then it would be a clear case of a warranty issue as long as the unit was undamaged (ie not crashed).

From what I recall, SRM have very specific torque requirements for the bolts on the chainset, especially for the chainrings. Since I don't have an SRM unit I'd hazzard a guess that this is more for accrurate torque measurement during normal power measurement than it is to prevent weird deformities like this happening.

We actually flipped the bike upside down and rested a pointer on the BB shell. I moved the pointer until it was just barely rubbing chain ring bolts as they passed by. I'd get three taps and two misses every time around. So, the spider is what's misaligned (though it could be the spindle or the spider, or the interface between the two).

I believe that the main reason that a particular torque is spec'd is more for uniformity around the whole chain ring. If the chainring bolts are each at significantly different torque values, or they are all significantly over/under torqued, it can mess up the 0-offset reading.

Dave

I had a DA srm that was wobbly. I brought this to the attention of SRM CO when it went in for a service, and they ended up replacing the whole pm with a new one. Sometimes the machining on the DA arm is not 100% correct and results in a misfit. Apparently it is not the spiders that are out of true, but only the crank arm machining that causes it.

^^^^ Good info. Thanks.