Internal Cup upper headset bearing won’t go in

[77martin]

I’ve just got Spa Cycles Elan mk2 frame to build up a ‘nice bike’ . The frame and forks are gorgeous I’m really chuffed with them. I got the Chromoly forks as I plan to fit a front rack.

I ordered an FSA Orbit ITA with it and they fitted the cups for me. I’m used to the bearings on headsets just slipping in, I assume that’s why the contact faces are angled, the bottom one does but the top one is pretty tight. Having greased it won’t go all the way in with firm finger pressure - I mean really pushing it. Doing this it goes about flush with the top of the cup but should go another couple of mil. I’m sure I have it the right way up. Having pushed it as far as I could with finger pressure getting it back out was quite hard, I was able to gently lever it out with a dowel but it was pretty well stuck.

I reckon it will go in with a press (I have a headset press and can heath robinson something with a socket), or I can try putting it in the freezer first. The question is should I? Should it be hard to get in or is there something wrong with it being so tight that needs sorting. I don’t want to have a nightmare when it comes time to regrease or change the bearing.

[tim-b]

Check the bearing against the spec here... https://shop.fullspeedahead.com/en/file ... c64bb7b90/

I've had a wrong bearing with a headset; similar looking bearings can vary by 0.2mm and be impossible to fit, e.g. FSA 51.8mm bearing (correct) but supplied with a SHIS IS52 bearing. In my case a retailer fault, not the manufacturer

[rjb]

A photo would help everyone respond.

[77martin]

tim-b thanks for the diagram, the bearing dimensions measure spot on.

I only have a cheap vernier caliper (from a German supermarket ) and am not an engineer… but I measure the inside of the cup to be 41mm exactly along the frame, but maybe 40.9mm across. I guess that’s why it doesn’t want to slip in.

[77martin]

rjb here’s a photo of the bearing about as far in as it wants to go (and another of the frame - excuse the greasy fingerprints all over the head tube - in case you just meant you wanted a look at it )

[6.5_lives_left]

I don't have any special knowledge about headsets, but is it possible that this is supposed to be tight, an "interference fit".

If that is the case, maybe cooling down the bearing by putting it in the freezer, then popping it into to the frame before it warms up will work.
Cooling it down might cause it to shrink in size sufficiently that it will go in without having to force it in place. Once it warms up though, it might be difficult to get it back out again, which could be a good thing if you don't want it to move around when you are using the bike.

[rogerzilla]

They are usually supposed to be able to rock within the cup (and should go in with a smear of grease) to accommodate the flexing of the steerer tube.

[77martin]

6.5_lives_left I tried it in the freezer overnight but it still doesn’t want to go in. I’ve driven bearings into other locations but not had to with a headset before.

rogerzilla thanks that chimes with what I guessed.

Doesn’t seem right to me I’ll ring up Spa in the week and see what they reckon.

Thanks for your help folks

[Brucey]

I suspect quite strongly that what is happening here i that the top cup is a bit tight going in, and the result is that the opening in the cup is no longer round.

I would suggest that the best course of action would be to knock the top cup out, and remachine the frame.

I have written before that I think it might be somewhat better if all press-fits on bikes were replaced with adhesively bonded joints, and if I am correct, this would be a case in point. The first time I did this, I thought it was a bit of a bodge if I'm honest. However, I now think that this might be the only way some parts can be installed without stress, and there would certainly be a lot fewer Al frames with cracked head tubes if this was standard practice. Importantly for the home mechanic, it would replace all those interference fits with sliding fits instead, and situations like this might be avoided. If the headtube is slightly oval it would be nearly impossible to overdo the remachining.

[torrens]

I have written before that I think it might be somewhat better if all press-fits on bikes were replaced with adhesively bonded joints, and if I am correct, this would be a case in point. The first time I did this, I thought it was a bit of a bodge if I'm honest. However, I now think that this might be the only way some parts can be installed without stress, and there would certainly be a lot fewer Al frames with cracked head tubes if this was standard practice. Importantly for the home mechanic, it would replace all those interference fits with sliding fits instead, and situations like this might be avoided

Interesting, this chimes with my own experience.
I’ve had three warranty replacement Trek Madone frames due to BB90 bb failures. The ‘precision net molded’ (sic) sockets, esp. the NDS, would fret oversize over a few thousand miles meaning that the ball races would literally fall out of the frame on disassembly.
A year or so ago when I noticed that the mild interference fit of my NDS bb bearing had become, yet again, a very relaxed slip fit, I decided to use Loctite bearing retention compound, plus activator, on it rather than go through the rigmarole of yet another warranty return. And, so far so good, the bb’s behaving impeccably.
I have a slight reservation concerning eventual bearing replacement; the Loctite MDS advises heating the housing to 200 deg C, eek! It’s a cf frame.
I machined up a test housing out of brass and when heated to ‘only’ 100 deg C I was able to remove the Loctited bearing with a smart whack so, hopefully, all will be fine at bearing replacement time.

[77martin]

Ok thanks Brucey that makes sense to me. Yours and other comments have convinced me that there is something’s not right and I shouldn’t just press the bearing in, which is what I was trying to figure out. I don’t have the tools or naus to remachine it myself so I’ll start by speaking to Spa and see what they recommend.

Thanks!

[Cowsham]

On my quest for a rigid fork for my mtb I found this which is a very good guide to headset bearings.

The sealed bearing types come in various bevelled angles. Some with 45° some with 36° and some with a mixture of both. I never knew that.

https://youtu.be/LNv5ywU896s?feature=shared

[Brucey]

.....I have a slight reservation concerning eventual bearing replacement; the Loctite MDS advises heating the housing to 200 deg C, eek! It’s a cf frame.
I machined up a test housing out of brass and when heated to ‘only’ 100 deg C I was able to remove the Loctited bearing with a smart whack so, hopefully, all will be fine at bearing replacement time.

getting an old adhesively bonded bearing out without heat is nearly always possible. If you want to do this, just pry the seals out where you can, then break the clip up in situ. If you then corrall all the balls to one side, the bearing can be disassembled. This should leave only the outer raceway to deal with. This can be attacked with a grinder and then broken up in situ. Normally the raceways are so hard that they are also brittle, so are somewhat susceptible to this kind of treatment.

[torrens]

Thanks Brucey; I think I’ll need to take a few deep breaths before adopting the above approach!
I installed the bearing using Loctite 641 retention compound coupled with Loctite SF7649 activator. I was rather hoping that, come bearing replacement time, I’ll be able to use a cup type puller combined with a little heat as required supplied by my Aldi digital variable temperature heat gun (a great bit of inexpensive kit btw).
Do you think this approach would be a non starter? If not, do you have any experience/knowledge of at what approx. temp. the Madone’s carbon fibre frame’s finish would be marred and, much more importantly, at what approx. temp. might structural damage begin to occur?

[Brucey]

IME if the paint doesn't survive unscathed, neither does the CF, either. Hence the interest in stone cold methods. BTW the bearings will be much improved by conversion to DFC specification, to the point that, if adequately lubricated, they are likely to last better than the rest of the bike. Conversion to DFC specification is quite possible with the outer race in situ.