Chainstay dimple tool and a couple questions about it.

[Mtblucas]

I made a fixture today to dimple chainstays to improve tire clearance. Here are some pics of it in action dimpling some old scrap chainstays. I think it turned out pretty good as it seems to work well. I had a couple questions about crimping the stays though, and I was wondering if y'all might be able to chime in.

First off, is there any rule of thumb or suggestions as to how deep of a dimple you make? Is there a point where the strength of the chainstay will be compromised? Will the chainstay still be plenty stiff enough?

Secondly, I'm assuming it's normal for a slight bend to form along the length of the chainstay when you dimple it?? It is very slight, but when I use this thing the chainstay bends maybe a degree or two on the opposite side of the dimple. I figure that is pretty much unavoidable when you bend/dimple the metal on one side of the tube.

Lastly, when using this would it be a good idea to heat the metal or just dimple it cold?

Thanks in advance for any information you can provide!

[rellis]

Definitely cold. Looks good.

[devlin]

If you look at what Kris of Forty Four does with his stays you'll have some of your answers.

[fortyfour]

For the record, I'm using straight gauge .035 wall thickness 4130 for all my stays (Typically .75" O.D.). This is merely to achieve the bends, dimples and forming I want with my stays to maximize stiffness and tire/heal clearances. This material is not as hard as many of the "off-the-shelf" chainstay's which have already been heavily worked to achieve their shape. Mine stiffen up considerably as I take them through these steps. So take care when dimpling off the shelf stays. There's only so far that you can push them. Personally, I try and create my dimples in a very smooth, uniform fashion. It's all about smooth transitions.

Here's a shot of my two dimple dies :



Here's one in use :





The inside of the die is the exact profile of the backside of the chainstay post dimpling. This is key to the setup as it supports the shape the chainstay is / will become as it is dimpled. When you create a dimple or a bend, you are moving material. The material closest to the dimple die is being compressed, while the material furthest away from the dimple die is being stretched. So you need to make a few to understand how far the material will move post dimple and take this into account. I have two separate dimple profiles which key into place on my vice (the one pictured above is a 3" diameter, and a second which is a 5" diameter). I've been preferring the 5" diameter as it creates a more uniform shape and has a wider profile.

[Craig Ryan]

Mtblucas, you may want to relieve the edge of the wood where it leaves the opposite side of the stay. You could end up with a crease there from the abrupt edge.

[Andrew R Stewart]

I'll add that it's not the depth of the dimple that counts but the change in the stay's width. This is measured at the edges of the dimple, or at the top of the two sides of it. Most crimping devices will deepen the dimple to get the top sides and stay width less. More sophisticated crimper's will press across a wider portion of the stay so a less deep but wider dimple results. Andy.

[Mtblucas]

Thanks everybody.

Kris, I really like that setup. I might try something like that in the future. I would probably make it out of wood though, but I'm sure that it will work just fine.

Craig, that is a good idea. There aren't any creases but I don't want to take any chances when I'm working with new chainstays as opposed to scrap. I will sand it down this evening.

[RspinnaKing]

So, here are some questions regarding chainstay "dimpling" versus "flattening" to improve clearance:

As Mr. Stewart pointed out and Mr. Garro illustrated with photos of his setup, what matters is not the "depth of the dimple" but the "change in the stay's width." That being the case, why would one want to use a "dimpling" setup like the chain link employed by the OP rather than a broader implement that presses across the entire height of the stay like Mr. Garro's? I seem to see much more "dimpling" on production bikes, sometimes so extreme that the outside and inside dimples for chainring and tire clearance must nearly touch inside the tube.

It seems to me that in a situation where the manipulations are extreme and especially when they're taking place on both sides of the tube, a "dimpling" setup may be capable of creating more clearance without creases or cracks since the tube has a figure-eight cross-section rather than a flattened oval? However, might it be the case that the actual clearance created by these extreme dimples ("the change in the stay's width") is no greater than what would be possible with a "flattening" setup used carefully and precisely?

Let's hear some thoughts, opinions, experiences on the pros and cons of "dimpling" versus "flattening".

[fortyfour]

So, here are some questions regarding chainstay "dimpling" versus "flattening" to improve clearance:

As Mr. Stewart pointed out and Mr. Garro illustrated with photos of his setup, what matters is not the "depth of the dimple" but the "change in the stay's width." That being the case, why would one want to use a "dimpling" setup like the chain link employed by the OP rather than a broader implement that presses across the entire height of the stay like Mr. Garro's? I seem to see much more "dimpling" on production bikes, sometimes so extreme that the outside and inside dimples for chainring and tire clearance must nearly touch inside the tube.

It seems to me that in a situation where the manipulations are extreme and especially when they're taking place on both sides of the tube, a "dimpling" setup may be capable of creating more clearance without creases or cracks since the tube has a figure-eight cross-section rather than a flattened oval? However, might it be the case that the actual clearance created by these extreme dimples ("the change in the stay's width") is no greater than what would be possible with a "flattening" setup used carefully and precisely?

Let's hear some thoughts, opinions, experiences on the pros and cons of "dimpling" versus "flattening".

Are you speaking of my own photos above of my dimpling set up? I'm not seeing any of Garro's setup above unless you are referencing another thread (in which case please add a link so I can see what you're speaking of).

Whatever technique you do use, what my goals are is to create a uniform, smooth transition from dimple or flattening as the tube transitions to the spot where you are doing the forming / flattening / dimpling.

[RspinnaKing]

Mr. Henry, my apologies for the misattribution. I was referencing your photos above and am not sure how I got confused there... I would correct the error in my previous post but it appears the time limit on editing has expired.

My mistake aside, your ideas about creating a "uniform, smooth transition" certainly get to the heart of my questions and make good sense.

[Andrew R Stewart]

R. King- Well actually, ovalizing of stays is the result of this experimentation. This was the shape that the, in my early days, classic tube sets came with unless you asked for all round chain stays. (In fact Reynolds came in dimpled or round/oval/round). The oval shape is what happens when you squeeze a round tube WITHOUT any internal mandrel or external form to get the resulting shape instead of a rectangle or "D" shape. So the cost of just ovalizing is pretty low. Adding a dimple to further take out the high point (at the minor axis) is a small extra step and also low cost. These methods work well and road bike components have evolved around these frame tube shapes.

Since these low cost solutions have worked so well for so many millions of bikes for so many years there hasn't been much need for more costly methods of shape manipulation for steel road bikes. Now that Al and mountain bikes have become a major segment this shape need has been revisited. Hence methods like hydro or oil forming have come about. On relatively soft materials like Al these methods aren't as costly as with higher strength materials like steel, but they still increase the manufacturing cost. Only in the higher levels of mass production do these methods make economic sense.

A small custom shop can invest whatever cost they wish to afford into forms and mandrels to control the shaping of the tubes. You've seen some of these already in this post. I made a set of wooden squeezing blocks many years ago and use them to initially or furtovalizelize stays sometimes. As i ride triple cranks and wider (for road) tires I need to do these easy clearance increasing methods on most of my bikes.

A side point that I mentioned needs to be followed up on. Components and frames have influenced each other since the first rider scooted down the Thed. Thr bike business is full of designs that failed in the market place because of the need to change frame shapes/requirements. The Browing transmission comes to mind. Currently, the belt drive is finding it's place, or lack of place. And in the end all this must fit between the rider's feet.

I don't know if this answers your questions but understand that thousands of fabricator have come before us and some have been a lot smarter the most of us reading this. Pretty much every manor has been explored and what we see today is due to evolution as it happens in the real manufacturing world. Andy.

[fortyfour]

Mr. Henry, my apologies for the misattribution. I was referencing your photos above and am not sure how I got confused there... I would correct the error in my previous post but it appears the time limit on editing has expired.

My mistake aside, your ideas about creating a "uniform, smooth transition" certainly get to the heart of my questions and make good sense.

No harm done. Just wanted to make sure I wasn't missing a reference.

Uniform, smooth transition is the key IMO with methods of this nature.

[William Chitham]

...furtovalizelize.... Andy.

Whoa there, go easy with the jargon!

[Andrew R Stewart]

Whoa there, go easy with the jargon!

One problem I've had with the spell check that I use is that it will combine words in very strange ways. I try to reread my posts after spell checking but some times miss stuff, like checking your answers on a test before handing it in, all you see is what you want.

Those who don't suffer from dyslexia or other related problems will never know the hassles some of us have. Andy.

[Andrew R Stewart]

There's a parallel thread over on Classic Rendezvous. Andy.

subject- Hammer time! 753 Chainstay Modification for Tire Clearance
classic-rendezvous-lightweight-vinta...oglegroups.com>

[Amaro Bikes]

Uniform, smooth transition is the key...

Which could be applied to almost any other aspect of the building procedure

[steve garro]

So, here are some questions regarding chainstay "dimpling" versus "flattening" to improve clearance:

As Mr. Stewart pointed out and Mr. Garro illustrated with photos of his setup, what matters is not the "depth of the dimple" but the "change in the stay's width." That being the case, why would one want to use a "dimpling" setup like the chain link employed by the OP rather than a broader implement that presses across the entire height of the stay like Mr. Garro's? I seem to see much more "dimpling" on production bikes, sometimes so extreme that the outside and inside dimples for chainring and tire clearance must nearly touch inside the tube.

It seems to me that in a situation where the manipulations are extreme and especially when they're taking place on both sides of the tube, a "dimpling" setup may be capable of creating more clearance without creases or cracks since the tube has a figure-eight cross-section rather than a flattened oval? However, might it be the case that the actual clearance created by these extreme dimples ("the change in the stay's width") is no greater than what would be possible with a "flattening" setup used carefully and precisely?

Let's hear some thoughts, opinions, experiences on the pros and cons of "dimpling" versus "flattening".

My "Blattener?"
That was another thread, but similar to what Kris is doing, but a little less extreme.
- GARRO.

[fortyfour]

This may be useful to this thread. Sometimes I need to add a secondary dimple to the outside profile of the drive side chainstay for chainring clearance. I've been meaning to put this one together for some time, and this latest 29+ project which is a Rohloff build (Rohloff requires 135mm rear spacing, so 73mm BB shell width was a must) had me looking for a little extra room for that drive side chainring to chainstay clearance. Here's it in action:



Here it is assembled/disassembled:



The tool allows itself to phase off of the dimple using the same profile. Window to line up where you want to create the second dimple, and tightening up the shoulder bolt snugs everything tight to lock it in place. Little more work to be done on the profile of the dimple die itself, but you get the idea. What I wanted was to create a tool that can be used once the frame was finished to I can check clearances with parts in hand rather than before the fact. I don't know how others perform this, but it was a hassle to hold the frame and juggle all the other factors. You can do it before hand, but I don't always have all the parts in hand when starting, just drawings, measurements and educated guess work. Now I can just have the frame in a stand, attach this tool, and it's done.

[afwalker]

^fortyfour
Man that is nice! What a great tool. Is there a slight radius on the inside "die" piece, the moving one? The picture makes it look like it has a bit of one.
thanks
andy walker

[fortyfour]

^fortyfour
Is there a slight radius on the inside "die" piece, the moving one?

Hi Andy, see first image below image. In the process of refining the die shape. What ever that shape is, will be what the resulting dimple will look like so I'm taking some time now to get it just to how I want it. So that includes making a lot of small stub chainstay dimple samples, testing the die, refining it and so forth until it's just how I want it. Also gives you the opportunity to see what happens when you push the material or just add a slight shape along with all things in between. It's a bit wide right now, so I'm working on it's overall profile and width - once it's exactly like I want it, I'll work it to 600 grit then polish it:



I cut the samples in half at the dimple to see how things shape up internally/externally paying close attention to everything going on. This second image shows a sample pushed pretty far to see what happens. Still relatively smooth in terms of shape.