coldsetting chainstay calamity? Advice needed

[zac]

Hello

I recently bought an old galaxy 531st frame off ebay. Got it powder coated (for 30quid ) and have been building it up using bits from my old frame.

I have built an old galaxy frame before and the dropouts were at 130ish when i bought it. This one turned out to have dropouts still at 126mm. I should have checked before i got it coated.

So anyway i tried to cold set the rear triangle using sheldon browns plank method http://sheldonbrown.com/frame-spacing.html

This does work and allows you to adjust the dropout spacing by quite small increments. Unfortunately i got carried away and there was that awful moment where i felt the frame suddenly 'give'. This sudden slight give resulted in the dropout spacing moving out by 8mm.

You can seen the resulting dent/slight fold in the drive side chainstay in the pictures below

The reason this happened i think is becuase whoever had the frame before me had manually dented the chainstays near the bridge by the bottom bracket. Not quite sure why they did this but apparently its quite common thing to do to get tyres,mudguards, chainrings etc to fit. So there is actually some denting in both sides of the chainstay.

So my question is .. do i continue to build the frame up ( I havent done the time consuming stuff like the cabling yet) and just ride it and see if its ok?

Or do i cut my losses now and get the tube replaced by Bob Jackson? I would have to get it re-coated- so all in about £100 to do this which is about what i paid for the frame, plus loads of time and hassle.

If i was just going to ride the bike unloaded or on day tours i think i would definitely go with the first option, but the thing is im going fully loaded touring in france for two weeks in August. I dont want my bike to collapse half way through the trip! (or some other trip down the line..)

I am looking for people with lots of experience of riding steel frames in all types of conditions, to say either.. go for it and dont worry too much or... no, the fact that you have actually bent the chainstay is more serious than denting it.. you need a new one.

Thanks for reading

Zac







So.. As you can see in the picture above i now have a bit of a dent/fold in the drive side chainstay

[Brucey]

plenty of frames have bigger dents in the chainstay by design, to allow for chainring clearance. So I doubt you have wrecked your frame; it won't be as strong as it was though.

Hard-brazed 531 tubes can soften considerably local to joints, so between that and the leverage the chainstay was always going to bend just there.

Needless to say what you need when you do this job is a controlled displacement as well as a controlled force. The latter in the absence of the former is not a good thing, as you have discovered. Please note that if respacing from 126 to 135 you will also put the dropouts out of line. You should re-true those too if you want to avoid stressing your axle unduly.

If you want to remove the dent you have made it should be possible to do this without replacing the tube or (if you are lucky) even having the frame refinished. You just need to insert a suitable mandrel into the tube from the BB end, then manipulate the tube from the outside. You may have to make a suitable mandrel for this.

531 tubes can be cold set more than most people expect. In your case I would expect -if it is going to crack at all- the tube to crack whilst being manipulated or after a short period of use.

In the event that a chainstay breaks on tour it can be lashed together using a splint of some kind and then welded later if a better repair is required. If any part of the frame is going to break in use, that one is one of the least likely to cause you to have a nasty accident or anything.

Most steel chainstays have a dents in the RH side. I always expected this to make them more likely to crack. But it seems this doesn't happen. I've seen quite a few aluminium frames crack on the RH chainstay (normally after being gouged by an unshipped chain) but if anything the left side chainstay seems more likely to break in service on steel frames. I can only suppose that the RH chainstay, with its collection of dents etc is much less stiff in bending than the LH one, so doesn't do its share of the work or something.

cheers

[Dynamite_funk]

In the event that a chainstay breaks on tour it can be lashed together using a splint of some kind and then welded later if a better repair is required. If any part of the frame is going to break in use, that one is one of the least likely to cause you to have a nasty accident or anything.

Take some jubilee clips and some spanners, worked for us on our tandem when we were in a bind

[mattsccm]

I have put bigger dents than that in deliberately when making things work that shouldn't.
I wouldn't worry.

[zac]

Thank you all so much for your replies. I have been fretting about it all day and you have set my mind at ease. And I dont have to start from scratch

Good to know that it wont ruin my trip if it does break. I will definitely take some jubilee clips. quite like the idea of riding home with a couple of spanners holding my bike together

Thanks especially Brucey for your comprehensive reply. Just one question about what you say re: dropout alignment. Is this about getting the rear end straight in relation to the rest of the bike? I have been using Sheldon Browns string method for this. Or is it another type of alignment?
Oh and was also wondering what you meant by controlled displacement.. An idea that has caught my attention just now
I found this method using threaded rod which looks like it might offer a bit more control http://seenonthetrain.blogspot.co.uk/20 ... frame.html
might try this if i ever do it again unless you know a better way?

[CREPELLO]

The threaded rod method linked to above looks appealing for the level of control, but I would question it's ability to control the balance resetting between DS and NDS. Essentially, it still has to overcome the difference in stiffness induced by the double-dent of the DS that will make that side bend that little bit more easily. The author also seemed to be saying that you should adjust the control nuts on the rod for each side. Quite honestly, I don't think it would matter if you only adjusted on one side only, because the method has no ability to control between DS and NDS. Or am I missing something?

BTW, I've used the wooden post method, and yes the DS did give much more easily, because of the double dent chainstay.

[reohn2]

So.. As you can see in the picture above i now have a bit of a dent/fold in the drive side chainstay

Are you absolutely sure the dent wasn't there before you started and the DS didn't give a little easier on that side because of it.

The alignment I suspect Brucey means is of the dropouts parallel with each other,so they aren't stressed when clamped onto the locknuts by the qr skewer.They can be checked with piece of timber the same length as the OLN width and cut square on the ends

[georgew]

The alignment I suspect Brucey means is of the dropouts parallel with each other,so they aren't stressed when clamped onto the locknuts by the qr skewer.They can be checked with piece of timber the same length as the OLN width and cut square on the ends

Right and there is a tool for this.

http://www.chainreactioncycles.com/park ... lsrc=aw.ds

[zac]

Hmm ok I reckon i will go for the square cut wood idea to check the dropout alignment. I guess they could be adjusted with a long adjustable spanner or putting the dropout in a vice? Not sure about paying 65 quid for a tool i will use once!

There was a bit of a dint before but i have made it bigger.

Re: the threaded rod method i see what you mean. I guess the advantage of the SB plank method is that you are bending each side of the rear triangle relative to the seat tube.
I was thinking though, if you could put the middle of the threaded rod in a small vice then would this mean you could adjust each nut individually and therfore widen each side of the rear triangle independently? It seems to make sense but it feels like i might be missing something.. dont know.

[Brucey]

an easy way to check the rear dropout alignment is to fit the front wheel inside the rear end of the frame, into one rear dropout only. With a bit of luck the end of the QR skewer nut will sit inside the frame as a pointer and you can check in one plane easily and the other plane with a bit of effort (easiest if you slide the wheel right back in horizontal DOs, I reckon).

You can check each side separately like this. Tempting though it is, don't tweak the wheel to reset the dropout!

A length of M10 studding, a few nuts, washers and a length of tubing can be used to make a resetting tool. You can use a big adjustable spanner, but sometimes the DO bends in the wrong place (ask me how I know.... ).

Obviously it is a good idea to check the front axle is straight before you do this check, else the whole thing is pointless...

cheers

[Mick F]

I tried the threaded rod method, and it was scary to say the least!
I was worried to hell about the rod slipping off and going into orbit.

What I did ........... and it worked a treat, was to lie the frame down on the floor and stand on one chainstay with one foot at the dropout and the other at the chainstay bridge. Then, I tugged upwards at the other dropout. It was surprising how much "tug" you can pull, and you can feel the metal give.

I turned it over, and repeated the other side, then measured, tried again, and fitted the rear wheel to see how I was getting on.
Trial and error, and perseverance, sorted it pretty damned quick in all honesty.

I did this to Barbarella our Mixte utility bike. It was previously a 126mm 6sp derailleur, and now it has a Shimano Nexus 8sp rear wheel at 130mm.

[robc02]

I've coldset two frames by holding the BB shell in a vice - with the jaws holding the ends of the shell - and levering each chainstay individually. Pieces of thin aluminium over the vice jaws help prevent damage to the BB shell faces. Alignment can be checked using the piece of string stretched from each dropout to the headtube method.

If the BB shell tries to move in the vice a helper to hold the headtube steady should fix it.

[zac]

well i ended up aligning the dropouts using the proper tool at a local bike co-op. They were way out so im glad i did it. The tool lets you align them easily in three dimensions and it only took five minutes to do so worth while finding one i reckon even if it means going to a proper bike shop

Also was told that i needed to realign the derailleur hanger and they had the tool for doing that too which was also easy.

Isn't steel brilliant?

Ta once more for all the help reckon im pretty close to riding the thing now

[zac]

BTW the bike Co-op was Pedallers Arms in Leeds. All run by volunteers, they have all the tools, its bike heaven

[mitb]

when I cold-set my frame from 126 to 130 I just wrestled the wheel in and over time it's set itself. It was torture to get it out soon after but is less so now. I did read up on coldsetting before but this just sort of worked so I didn't run the risk of nadgering the bike. Hope I haven't missed something terribly important..!