I always thought after you've set the preload, and then tightened your stem bolts, the stem cap is pretty redundant.

But then I used a cheapy plastic top cap for while and I think the headset developed a bit of play. Seems to have improved after I put a metal one back on.

So do you think it's necessary? Surely once tightened, the stem can't move?

no, once you have tightened your stem bolts it should stay tight.

the cap keeps dirt out, and is a handy tool for tightening/adjusting on the trail 🙂

'course it can move, and because of this FACT yes it's necessary IMO

No, it's there to preload the bearings, the stem bolts if tightened correctly hold the stem on.

One of my bikes runs without one

I thought it also stopped nice little puncture holes to your head when you had an off 😉

Off course it does… It looks nice 😀

A friend of mine didn't put a star fangled nut in his fork, he used a massive bolt to pre-load the headset then took it out when he'd tightened the stem, rode the trans-alp with no issues at all. Looks slightly odd though, with a hole straight through the steerer still!

Looks slightly odd though, with a hole straight through the steerer still!

Yeah – that's what mines like I just used a bit of 2×2 & brute force to pre-load it – I couldn't find a SFN at the time 😀

I'm not sure I'd be happy to do something like trans-alp with it like that though as it reduces your ability to do running repairs a bit

reduces your ability to do running repairs a bit

That was my thought, if your headset comes loose you really are screwed. I suppose it's not the end of the world, you can finish the day with a vaguely loose headset, but I wasn't sure.

You can actually preload a headset without a topcap – just lean on the stem, wiggle it about a big until it's all seated and tighten – not ideal since it's hard to get it exactly right but it certainly works in an emergency.

I suppose it's not the end of the world, you can finish the day with a vaguely loose headset,

and when you get back you just have to deal with crushed bearings, and an ovalised headtube. Well worth the weight saving

and when you get back you just have to deal with crushed bearings, and an ovalised headtube

only if you are The Hulk!

I took my SFN and top cap off recently – no problems at Kielder running it like that. Just have a carbon steerer expander plug in your toolkit if you're travelling to races and may need to adjust things.

Well thank you folks, looks like it was just me then.

and an ovalised headtube

how might this come about ? If you stuck the thing in a hydraulic press and gave it some Wilbert it would eventually spread, but I can't see a mechanism for making it out-of-round (until it fails)

I tried it a while back when i got new forks. Used a removable expander thing, tightened up the headset and removed it along with the cap. Stem bolts were extra tight but within a few miles the headset had come loose.

It was on my King headset so no suprises there, yet another let down with the thing.

really Simon, if you put a loose bar through a metal guide, and then force it backwards and forwards + up and down hard over hundred/thousands of cycles (as would happen if it came lose mid long ride) you don't think it will ovalize the guide?

With each wobble/rock it would loosen itself off some more, as the leverage against the stem would increase as it could rock more and more and slide the stem up the steerer, allowing yet more leverage and eventually cracking the bearing and stretching the headset.

Mind you I suppose the steerer could snap first, as it would get very scored as it moved around (as can be seen by the flex allowed by the King headsets and long forks)

This is what can happen,simon. i had a bike in a while back and the splits down the front of the alu headtube were about 15mm long (each)

the guy chuckled and said he would get it welded. he did, and its still going strong-(ish)

you don't think it will ovalize the guide?

I can see it might ovalise the headset, where the movement is taking place, but not the housing. A round thing very tight in a round hole seems to have scant scope for unroundness…

and the splits down the front of the alu headtube were about 15mm long (each)

wow!
1) are your arms like tree trunks ?
b) was the headtube ovalised as well as split ?

I've rounded a headset in an old steel frame, but it was about 15 years old and had about 10 years runnign with a loose headset rattling around in there daily. Frame was flawless when it got replaced. There's only 2 methods of getting the head tube ovalised really, ignoring creep under very very high loads, and thats plastic deformation and wear. It wont wear unless the headset cup is undersized/damaged and it wont plastically deform without mental amounts of force. At least thats my view of it.

yes it will take a lot of force, however if the lever is long enough then the force of slamming a bike down a rocky descent would be enough.

The headtube of the tandem ovalised from riding with a loose headset.

The headtube of the tandem ovalised from riding with a loose headset.

was there any headset left ?

but actually, we're straying from the point that the cap does nothing except during pre loading. It won't stop the other bolts coming loose as its force is perpendicular, and it won't stop the bars turning round if they do

yes it will take a lot of force, however if the lever is long enough then the force of slamming a bike down a rocky descent would be enough.

But that would occur without any play. The play does not create extra force. Ok a smidge due to dynamic loading if the steerer started off at one extreme and was rammed to the other instantly.

The headtube of the tandem ovalised from riding with a loose headset.

Over how long? Damage to the headset cup (crack etc?)?

What happens in my view is instead of the forces being applied smoothly due to the free play there is an impact – the same total force but builds up over a shorter time frame – same sort of effect as having a loose seatbelt.

with a tight headset there is no impact – loads apply smoothly. with a loose headset although the total loads are the same the rate at which the load is applied alters – it happens suddenly as the slack is taken up rather than smoothly.

The tandem – it was loose for a bit and the headset was perfect – just moving in the headtube – headtube alloy, headset steel. Cured with a deep insertion headset and locktite bearing fit

If your new headset develops a bit of play, it is likely that the bearings have just bedded in and re-adjusting the tension will fix it permanently.

If, on the other hand, your headset keeps coming loose, then there is something wrong!

The play does not create extra force. Ok a smidge due to dynamic loading if the steerer started off at one extreme and was rammed to the other instantly.

The principle behind my favourite tool the hammer. And all the variants on the hammer, some of which use repeative small blows to make things fall apart – SDS drills for example.

If the two circles (the head tube and the headset) are properly sat into each other there is less scope for the hammer effect.

But I don't know if a loose headset is actually a problem in the really world 'cos I tighten mine up and keep the top cap in place so I can re-tighten if anything starts feeling loose.

Edit – Actually I think TJ just said that

The difference between a push and a punch. They could have the same force but one is more destructive than the other

and of course the tightness of the top cap has nothing to do with these effects…

Well not provided you pre-tensioned the bearings before you tweaked the stem up, no…

Just thinking, I haven't argued with SFB on here in ages

no need to argue, SFB is clearly an idiot