Help S.Oz.S. My Oz has suffered a serious injury

[Luke]

Hi all,
I seem to have arrived a bit late, but I’ll add my 0.02 Euros
Exactly the same thing happened to me in 2002.
As I have NEVER lowered my seat for technical riding I simply used acrylic sealant (like bath silicone but a bit stronger) and just glued the post into the insert.
A good quick fix.

Only in January of this year the top of my seat post split and the head fell out!
So I cut the post to 5 cm, bonded a suitable piece of solid bar into the end (to stop the tube collapsing) and went at it with a BIG pipe wrench and lots of silicone lube.
Turns out the acrylic sealant is stronger than the original bond between tube and frame, and after a lot of sweating and swearing out came both parts.
I was interested to discover that there is another aluminium part bonded into the carbon.

I measured up and traced a drawing for a new post.
2 things I improved on:

That shoulder is the perfect stress concentrator; I replaced it with a nice long taper.
Inner diameter 26.8 because I happened to have a post that size kicking around.

I discovered a third improvement when the machinist left out the slot

It clamps fine without the slot, as long as your tolerances are respected.

To reply to some of the suggestions/ideas above:

Bigger shoulder = bigger concentration of stress in one zone = bad idea.
Moving the SHOULDER away from where the tube leaves the frame gives the thin part of the tube the opportunity to flex with the post (and at 25mm it WILL) = good idea.
Taller thick section = less flex in the thick section = more flex where the diameters change = bad idea.
The thin seat post will really flex, it will work the tube MUCH more. See above for stress concentrations.

Materials
Stainless + alu + water = battery (Google galvanic corrosion or galvanic couple)

7075 is stiffer, has a higher yield strength BUT the step between yield and rupture is very small. In other words if it bends it will break; not good for a part that you have to preload just to clamp the post…
I had mine made from 2014 alu, lower yield strength and the same rupture. Lower grade alu generally has better fatigue characteristics.
Sometimes it’s better to let the parts work together rather than try to stop the flex with exotic materials.
And it’s a damn sight easier to machine.

If I can work out how to upload files I’ll post the Autocad or PDF file of my insert.
In the mean time PM me for a copy.

I’ve just re-read that; sorry if it sounds a bit authoritative, it wasn’t my intention. Part of my work involves forays into stress evaluation/calculations (albeit in hydro-electric turbines rather than bicycles)

Happy trails
Luke

[shovelon]

Hi all,
I seem to have arrived a bit late, but I’ll add my 0.02 Euros
Exactly the same thing happened to me in 2002.
As I have NEVER lowered my seat for technical riding I simply used acrylic sealant (like bath silicone but a bit stronger) and just glued the post into the insert.
A good quick fix.

Only in January of this year the top of my seat post split and the head fell out!
So I cut the post to 5 cm, bonded a suitable piece of solid bar into the end (to stop the tube collapsing) and went at it with a BIG pipe wrench and lots of silicone lube.
Turns out the acrylic sealant is stronger than the original bond between tube and frame, and after a lot of sweating and swearing out came both parts.
I was interested to discover that there is another aluminium part bonded into the carbon.

I measured up and traced a drawing for a new post.
2 things I improved on:

That shoulder is the perfect stress concentrator; I replaced it with a nice long taper.
Inner diameter 26.8 because I happened to have a post that size kicking around.

I discovered a third improvement when the machinist left out the slot

It clamps fine without the slot, as long as your tolerances are respected.

To reply to some of the suggestions/ideas above:

Bigger shoulder = bigger concentration of stress in one zone = bad idea.
Moving the SHOULDER away from where the tube leaves the frame gives the thin part of the tube the opportunity to flex with the post (and at 25mm it WILL) = good idea.
Taller thick section = less flex in the thick section = more flex where the diameters change = bad idea.
The thin seat post will really flex, it will work the tube MUCH more. See above for stress concentrations.

Materials
Stainless + alu + water = battery (Google galvanic corrosion or galvanic couple)

7075 is stiffer, has a higher yield strength BUT the step between yield and rupture is very small. In other words if it bends it will break; not good for a part that you have to preload just to clamp the post…
I had mine made from 2014 alu, lower yield strength and the same rupture. Lower grade alu generally has better fatigue characteristics.
Sometimes it’s better to let the parts work together rather than try to stop the flex with exotic materials.
And it’s a damn sight easier to machine.

If I can work out how to upload files I’ll post the Autocad or PDF file of my insert.
In the mean time PM me for a copy.

I’ve just re-read that; sorry if it sounds a bit authoritative, it wasn’t my intention. Part of my work involves forays into stress evaluation/calculations (albeit in hydro-electric turbines rather than bicycles)

Happy trails
Luke

So you bonded your seatpost into the frame? Hmmm, I saw an old American classic seatpost with an expander at the bottom. may not be a bad idea to lock it at the bottom too.

You are right on with SS/alum contact and galvanic corrosion. Wonder if there is any coating that can arrest that? Some kind of zinc chromate paste perhaps?

I can agree to a point with your assessment of 7075-T6 having a narrower fatigue band, but 2014 has a higher degree of plasticity which leads to strain hardening to the level of 7075. Upon reaching this level of strain hardening, the yield is much lower than 7075. 7075 has a bigger memory as far as movement is concerned which also makes it less plastic. The biggest issue really is the stress riser positioned right at the stepdown of the 2 diameters.

I certaily do like your choice of 26.8 inner diameter.

Terry

[purple gerbil]

just a thought. but you guy's have access to the bottom of your seat post's.. what if you used a wedge system like in the old stems??

[Spokes]

wow... you guys confuse me. Im a carpenter so im pretty sure anything I make WILL break, but look lovely in varnish . But I also have an OZm frame with the same predicament. So can I ask if and when someone decides which is the best way to remake the insert,can you make two and sell me one, please...

[Thunderchild]

Luke,

That shoulder is the perfect stress concentrator; I replaced it with a nice long taper.
Inner diameter 26.8 because I happened to have a post that size kicking around.

How did you get the taper to fit into the frame; isn't it a constant diameter?   

I will also have the machinist put a radius in the shoulder. 

Thunderchild

[Luke]

The frame is a constant diameter, the whole of the taper sits outside of the frame. It sits the collar a little higher but that’s more help than hindrance (see previous post).

Simon, you’re right about the 2014 work hardening problem. The higher the stress for each loading cycle (bump on the trail)the quicker it will harden.
I don’t think the tube will be deform plastically during it’s use.

I’m just hoping (and I’m not going to calculate it!) that my taper and the long tube keep the stress levels low enough for the work hardening to take a couple of decades to reach critical level.
It will also depend on how much the seat post flexes, which is a function of how much it sticks out and how fat I am.
IMHO a 25mm seat post will rip the tube apart regardless of the alu grade used.

But that’s only based on instinct.

Only time will tell…

[Thunderchild]

A couple of questions for you metallurgist dudes. 

#1  Will it matter if the seat tube is made from a billiet of 7075 or should I insist on bar stock? 

#2 What are the tolerance limits for seat post to seat tube fit?

Thanks again

Thunderchild. 

[kiwi]

The frame is a constant diameter, the whole of the taper sits outside of the frame. It sits the collar a little higher but that’s more help than hindrance (see previous post).

Simon, you’re right about the 2014 work hardening problem. The higher the stress for each loading cycle (bump on the trail)the quicker it will harden.
I don’t think the tube will be deform plastically during it’s use.

I’m just hoping (and I’m not going to calculate it!) that my taper and the long tube keep the stress levels low enough for the work hardening to take a couple of decades to reach critical level.
It will also depend on how much the seat post flexes, which is a function of how much it sticks out and how fat I am.
IMHO a 25mm seat post will rip the tube apart regardless of the alu grade used.

But that’s only based on instinct.

Only time will tell…

whats your definition of a "long" taper.If the step is say from a of(using theoretical rounded dimensions for our seat post insert) say 32mm od for the actual collar and say 28mm for the actual peice that goes into the seat tube...

[Luke]

The change in diamaters is spread over 10mm, I've glued the tube in so that the taper doesn't contact the frame, in all there's about 35mm of tube outside the frame.
If the glue's ok the whole lot doesn't slip down and work its way into the frame.

Billet, bar, it's all hard to machine in 7075. The clamp will stop it splitting along its length, so it doesn't make much difference. Setting up a non-round part in a lathe to then turn it down is just more of a pain.

Tolerance - I used H8 which works without a slot. I haven't got the tables nearby to tell you + - how much that is (on lunch); any machinist will be able to find them.

Considering the multitude of seatpost manufacturers, the volume at which they produce and the varying quality assurance standards the world over, I'm not sure tolerancing without having measured the seat tube you're going to use is that useful.

After all, that's why there's a slot, to make up for mass production tolerance problems.

Happy machining
Luke

[shovelon]

The change in diamaters is spread over 10mm, I've glued the tube in so that the taper doesn't contact the frame, in all there's about 35mm of tube outside the frame.
If the glue's ok the whole lot doesn't slip down and work its way into the frame.


Happy machining
Luke

Hmmm, 10mm taper. That's a good idea.   

That gives me an idea of my own. I can put a road clamp on my Thompson seat post, and add a wedgelock at the bottom of the seatpost. The screw would run up thru the post and exit on top of the seat plate. Then I reduce all the other clamping points to make up for the added mass. That may just length the life of the clamp area.

[purple gerbil]

just a thought. but you guy's have access to the bottom of your seat post's.. what if you used a wedge system like in the old stems??

terry!! great mind's think alike.... but didn't i say this some day's ago.. c,mon keep up..

[shovelon]

just a thought. but you guy's have access to the bottom of your seat post's.. what if you used a wedge system like in the old stems??

terry!! great mind's think alike.... but didn't i say this some day's ago.. c,mon keep up..

That's where I got it!

Thanks,
Terry

[Thunderchild]

Update:  The seat tube should be done next week.  I had already purchased the 25.0 seatpost, so I have to stick to that route with the thicker wall seat tube.  It is a Thomson (reinforced on the front and back on the inside, so I hope it doesn't flex too much.  I decided to stick with the design with a shoulder with a full radius and no slot for the seat post as the machinist is going to do a tight slip fit with the seat tube.   

I forgot to ask what epoxy glue others have used to bond the seat tube into the frame.  Seems like a GM type of epoxy was mentioned in the past.  Chime in if you have used any.  Once I remove the old tube, surface prep with IPA is sufficient, right? 

Thanks
Thunderchild

[Luke]

I forgot to ask what epoxy glue others have used to bond the seat tube into the frame.  Seems like a GM type of epoxy was mentioned in the past.  Chime in if you have used any.  Once I remove the old tube, surface prep with IPA is sufficient, right? 

I used Scotchweld 490, an epoxy adhesive from 3M. But only because they were throwing away tubes of the stuff at work after it failed the mass spectrometry test for Shuttle flight qualification.
Any LONG cure two part epoxy adhesive should do the trick, there's so much contacting surface compared to the load that It doesn't have to be the flashiest stuff.

Just a thought, but it's too late for me: perhaps it would be wise to bond it with a glue that has a known and available release agent, just in case you have to repeat the whole ordeal...

Surface prep: I rubbed my bits down with sandpaper and degreased with acetone. What's IPA? (I'm so ignorant)
Good luck
Luke

[Thunderchild]

IPA = isopropyl alcohol.  I like the idea of acetone.  I think it is a better all around solvent that evaporates well.  A glue with a release agent would be nice; I will post any that I find. 

Thunderchild