cruiserandmax

After setting up the bike and tightening the front headset as much as possible (any amount tighter at all causes noticeable resistance in the steering) there is still to much play front to back when rocking with the front brake held. See in video:


I just cleaned and re-used the bearings I found in the headset to begin with. The same play may have been there to begin with- I got the frame by itself to start. The balls measured 5/32. I didn't keep track of which set was from the top/bottom when pulling them- but both bearing cages have the same diameter (~ 37mm).. Is it possible someone had wrong sized bearings in there? And maybe they should be 3/16, and that difference is allowing this amount ot front/back looseness ? There isn't really any up/down play.. I am sort of baffled.

Andrew R Stewart 

A cup and cone bearing set (like a headset) that can't be adjusted to have minimal friction and no slop is generally indicative of the two bearings being misaligned WRT to each other or that their ball tracks are so worn that they are no longer suitable.

With a headset the steerer could be bent. The faces of the bearing seats might not be parallel to each other or coaxial to the steerer. The cups might not be fully seated against these surfaces.

But a common reason for this difficulty in adjusting a headset is that it was reassembled with a retainer ring of balls upside down so the ring is contacting one ball track. Very easy to do and quick to check for.

Many will reassemble these retainer rings with a full complement of loose balls (less one to insure the balls don't ride up on each other). Andy

smd4

Also, note there are more balls in the bottom retainer. Make sure you have the correct set in the correct cup.

cruiserandmax

Thanks for the tips- I did put the more balls retainer in the bottom but it is entirely possible that I placed one or both 'upside-down'... I will open them up and investigate that possibility.. When disassembling the headset originally I had planned to replace with loose balls- but was having such a hard time not fumbling balls all over the place throughout the process that I just went with the cleaned retainer ring balls since the balls themselves after cleaning all looked very smooth and defect-free.. But I'm willing to attempt loose balls again- I still have 60+ grade 25 5/32 balls to work with..

3alarmer 

.
....I watched your video, repeating it about six times. That's kind of a new one on me,
On the video, it looks like the top nut and stem are moving back and forth as you rock the bike.

Usually, when I have a loose headset, the play seems to show up elsewhere.

Do you have enough threads on the fork to get good engagement of the top nut ? And are the threads there in good condition ?
If you're short on fork threading at the top, does it help to leave out one of those keyed washer/spacer things ?


I guess it might be a trick of perception, but when you tighten the top nut, locking it against the threaded top race, using two wrenches, those two guys should remain solidly against one another.

cruiserandmax

So yes! You are super perceptive- I had the top nut loosened in that video. And I had the race tightened to what I thought was "max" tightness (any more and the steering starts binding)... However just revistiting now and there was a little room for another 1-2 quarter turns on the race cap. Doing that, followed by tightening down the top nut- there is no more visible play in the race cap/nut when rocking- and the stem/top nut movement while rocking goes away. However there is still a slight "clunk" noise/feel when rocking... I don't get that 'clunk' when rocking either of two other simialr 90's aluminum frame trek road bikes I have... Maybe the races are just worn to the point of allowing a tiny bit of space- just enough to make me perceive a clunk, but maybe ok to ride on... Meh... Maybe they are worn enough that I could put 3/16 balls in there and get good tightness without play? (only half joking :/)

Kontact

Do you have one of the bearings in upside down?

cruiserandmax

I didn't open and check that yet- I just tightened a little more and locked the top nut after reading the most recent sug- I am heading into the garage to do that now- (thanks)... Will report back.

3alarmer 

.
...yeah, the clunk is not acceptable. Good luck checking the bearings. Just in case, on that sort of headset, I usually install loose bearings by laying in a bead of grease in the top race, and similarly in the bottom race that is pressed into the head tube. The grease is what keeps the loose bearings in place as you fill the races. Then flip the bike upside down in the stand, insert the fork, trapping the lower bearings in place. Then screw on the threaded, upper race from below, so the bearings are supported in that race by gravity, until you make contact with its mating race.

If a headset is worn, it's worn. Sometimes the loose balls thing will save it a little longer by changing the contact surfaces from the ones worn in using caged bearings. But using larger bearings won't help you out at all.

cruiserandmax

Ok! Solved (I think)! I flipped the top bearing, and then as I was re-tightening I notice something was happening that I didn't notice before- I was doing routine increments of: tighten gradually- rock fore/aft with break applied to check for play- when no play was felt turn handlebars left/right +/- 90 degrees a few times to feel for any binding... This time I noticed on every turn back and forth to check for binding the race cap was loosening a little (for whatever reason). So when I cranked down the top nut I was actually setting a looser bearing race than I was assuming!! After preventing that loosening from happening before tightening the top nut I ended up with no play and nice free handle bar movement!! Phew..

Thanks all again for the leads to solve this!

Kontact

So did you have one or both bearings in wrong, or was it just technique? They only fit one way.

cruiserandmax

The bottom was on the correct way- turning that one around just made the retainer scrape the races. But the top set retainer seemed basically symmetrical (the balls protrude above/below the retainer spine the same) but I tried flipping it any way- then while tightening after that was when I realized the technique thing to make sure not to let it loosen when checking handlebar movement... Though maybe there was something subtle about the shape of the top retainer that I didn't notice and now it's in the 'correct' way? Honestly the whole headset movement feels just as smooth with the top retainer in either orientation.

Andrew R Stewart 

Accessing bearing adjustment with the threaded race/cup/cone loose will never tell you how that bearing will feel when all is properly tightened. While I try to avoid absolutes in my posts here (and in other forums) this is one area that is an absolute. Bearing adjustment check is only a valid try if the lock nuts and threaded race are fully tight against each other to eliminate any play on the threads.

I should have mentioned this before but assumed that this was described in whatever guide you were following and thus you knew this. Andy

SurferRosa

You're using two wrenches, right?

SJX426 

@cruiserandmax - when using loose balls, coat the cut with grease and place the balls in the grease. it will retain the loose balls in the cup. Orient the bike upside down for the bottom cup. and try to "level" the cup as well.

cruiserandmax

Yes. I hold the race cap in place with one while tightening the top nut with another.

Kontact

There is no way that they were symmetrical top to bottom.

cruiserandmax

Thanks for the ref images! Then the install before I flipped it was probably that wrong orientation. The headset is play/clunk-free and turns completely freely. I'll look more closely at the retainer ring(s) before removing them whenever I open a headset next.

Kontact

Just look at how the headset and most non-cartridge systems work: One bearing race is cup shaped, the other cone shaped. If you put the bearing retainer in upside down, the bottom corner of the sheet metal retainer is going to the touching the cup. That's the outer ring you see in the photo I posted labeled "back". In the photo you posted of your bearings, both are oriented to be able to see "top". Top goes into the cup, back gets inserted over the cone.

This is true of headsets, BBs and even some hubs that use retainers (old Schwinns).

cruiserandmax

I've always just thought about 'balls between two surfaces' and never really thought about the 'cone' part of "cup and cone" bearings. That makes complete sense- thanks again!

SurferRosa

When I'm rebuilding a bike, I install the bb, then the headset, but I don't adjust the headset perfectly until wheels with tires and a stem and bars are in the frame. When I adjust one, I straddle the top tube and use two wrenches like you stated. But I go incrementally. I tighten them a little bit to each other, remove the front wheel, and check for smooth steering and play in the fork. If it's too tight, I cannot feel that accurately with the front wheel in the frame. If if it feels like I'm on the right track, I replace the front wheel and continue tightening the two to each other and then repeat the same process by removing the wheel again. If it feels like it's too tight or too loose, I eyeball the adjustable cup and move it accordingly. If I go incrementally, it's a lot easier, and I can perfectly dial it in and get the two pieces super tight against each other.

Mr. 66

When I rebuild a headset I finger tight both cup to the bearings and locknut to cup. I then raise the cup tightening to the locknut until the rotation is free then I chinch the locknut to the cup, or the washer in between. I do this with the wheel not installed.