Frame tubing measurements: Initial data set, including several production frames

[Reed Kennedy]

After much faffing about, I releasing my first set of bicycle frame tube wall thickness measurements. You can find them here:

https://docs.google.com/spreadsheets/d/1XeS2nJhBPSDRbqgG1-F2VOI_vK1QZnZmysARDJWyGps/edit?usp=sharing

Included in this first data set are an interesting selection of Rawland, Rivendell, Surly, and other production bicycles. 

I've spent a goodly amount of time with a crashed Centurion Ironman frame (Tange #1, 4130) taking ultrasonic measurements and then cutting up tubes, deburring, carefully sanding away paint, and then re-measuring with a high quality set of Mitutoyo digital calipers. I've also spent a great deal of time comparing other frames against tubing manufacturer and frame manufacturer specifications. I am now confident my measurements are reasonably accurate for bikes made of 4130 steel, which is most bikes. Let's say within 5%, probably better. They should also be fairly representative for bikes made of other alloys, more testing to come.

Note: I am providing these results at the full resolution of the device I used to measure, but please don't get too hung up on hundredths of a millimeter. I recommend rounding up. Or down. Whichever makes you feel better.

Thanks to everyone at the Bicycle Quarterly Un-Meeting last weekend who let me measure your frames, to Eric Keller who originally suggested ultrasonic measurement, to Joshua Horowitz who provided the crashed Centurion, and Charles Vasta who donated a damaged Reynolds 531 frame for testing and calibration.

A very special thanks to Jan Heine, Mark Vande Kamp, Alex Wetmore, and Hahn Rossman for their work in identifying, researching, and teaching about the concept of planing. 

Things to do better going forward: Carry a caliper to measure tubing diameter and a tape measure to measure frame size. If anyone knows any of the folks whose bikes I measured at the Un-Meeting please put me in touch! I'd love to get their frame sizes.

And if you're able to make it to San Francisco with your bike, please drop me a line, I'd love to measure your frame. Even if I've already measured another one of the same type, it would be ideal to confirm those results.

Best,

Reed Kennedy

San Francisco, CA

[Mark Bulgier]

Reed,

I love this!  So cool -- thanks a million.

What do the different shades of green mean in the spreadsheet?  I have a feeling I'm missing something obvious.

I modestly urge you to not skip the "rear" measurements on the TT & DT, because triple butted is a thing...  FYI for newbies, it doesn't mean 3 butts, it just means the two butts are different thickness from each other, so together with the unbutt, there are 3 distinct thicknesses. Ritchey Logic for example.  I'm a fan of TripButt (so-called by no one, ever) because most tubes have more stress and/or fatigue issues at one end than the other, so the butt should be heavier at one end.  Which is almost always the HT end, for TT & DT.

But beggars can't be choosers.  I ain't doing the work, so I complaining!

Mark Bulgier

Seattle

PS, quadruple butting is sort of a thing, but not really.  Same with "Infinity" tubing.  I wouldn't bother taking more than 3 measurements on any tube unless you were motivated to, for instance by curiosity about a "Quad-butted" tubing sticker.

[Reed Kennedy]

Mark, my pleasure! I'm always excited to put hard numbers on something that is otherwise so subjective. Looking forward to helping our planing and tubing conversations continue!

I definitely prefer to measure at three points on each tube (1 cm from front joint, center, 1cm from rear joint) whenever I can. I haven't come across much triple butted tubing, but it most definitely does exist. Even when measuring double butted tubing it gives me a chance to double-check my work, and gives a bit of insight in to the manufacturing tolerances of the tubing. (Of course, here in the real world, wall thicknesses do vary!) Ideally, I'll even measure the center of the tube in one location and then a second location 180° around the tube. 

That said, when I've got a slew of frames waiting to be measured and time is limited I'm not above making compromises. At the BQ Un-Meeting I measured three different times, eventually measuring until I missed the group riding the day-2 route I wanted to take, and I still didn't get everyone I'd hoped to! Tough to choose between excellent measurements of fewer bikes good data about more bikes.

I'm gonna try to bring this device to more bike events where I'm not also trying to go on a ride. Then hopefully I won't have to worry or choose!

Oh, and the spreadsheet colors just show where that value lies in the range of values in that column. The smallest numbers (thinnest walls) are the deepest green. The largest numbers (thickest walls) are pale green or even white. This is an imperfect visualization, as it does not account for either diameter or tube length. Still, I think it's neat, so I left it in.

(BTW, don't miss the crazy seat tube on the Nashbar bike. 0.5 mm where it joins the bottom bracket! I triple-checked it. Some interesting comments by Adem, the owner, if you hover over the name cell.)

Best,

Reed

--
You received this message because you are subscribed to the Google Groups "650b" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 650b+uns...@googlegroups.com.
To post to this group, send email to 65...@googlegroups.com.
Visit this group at https://groups.google.com/group/650b.
For more options, visit https://groups.google.com/d/optout.

[Justin, Oakland]

Very nice!
One suggestion:
Separate out:
Brand | Model | Year | Size

I can see wanting to see just 55cm frames or just Surly or whatever. Makes it easy to filter using some of he GDpcs built in tools too!

-J

[John Hawrylak]

Reed

All I can say is WOW.  

Would greatly appreciate a TT & DT measurement of a mid-80's Schwinn with the Columbus Tenax tubing (all Std Diameter), if you ever run across one.  I believe my 88 Voyaguer Tenax (21") is 9-7-9 TT & 10-7-10 DT.  Would love a measurement through.   

Interesting, the RBW Beleriot's DT appears to straight gauge!!

John Hawrylak 

Woodstown NJ

[Adam Paiva]

Super cool. Well done Reed. This is bike nerdery at its finest.

[Reed Kennedy]

On Sat, Sep 15, 2018 at 12:06 AM John Hawrylak <John.H...@verizon.net> wrote:

Interesting, the RBW Beleriot's DT appears to straight gauge!!

Yup, and the top tube on the Hunqapillar as well. Weird, but durable!

Reed 

[David Cummings]

Now the question is, "How do we interpret the data?"  In other words, how do you intend to use it? Will there be a column added for planing or high speed shimmy? ;)

Very nerdy, and i love it!

[Reed Kennedy]

On Sat, Sep 15, 2018 at 9:35 AM David Cummings <flath...@gmail.com> wrote:

Now the question is, "How do we interpret the data?"  In other words, how do you intend to use it? Will there be a column added for planing or high speed shimmy? ;)

Hey David, good question! I don't plan to add subjective information to the spreadsheet I'm maintaining. My goal is to provide people with reliable measurements of frames they own or are considering purchasing. What y'all do with that data once you have it is up to you!

My hope is that when people talk post about how a frame feels they'll be able to list the tubing spec instead of just the make and model. Then we can all begin to make correlations and hopefully understand more about the experiences we share and discuss.

Best,

Reed 

[Justin, Oakland]

I would L O V E to get multiple model years of the various Rivs and see how the tubing changes over time.

-J

[Reed Kennedy]

New measurements up, including some Rivendells and Crusts. Thanks to Ben Miller for making his bikes available!

https://docs.google.com/spreadsheets/d/1XeS2nJhBPSDRbqgG1-F2VOI_vK1QZnZmysARDJWyGps/edit?usp=sharing

If anyone else is going to be around the SF Bay Area and would like their frames measured, please drop me a line.

Best,

Reed Kennedy

San Francisco, CA

[don...@desinc.net]

Did internal frame coatings (boeshield, frame saver) make any difference on readings? Or a different frequency that you filter out from the transducer?

[Reed Kennedy]

On Mon, Sep 17, 2018 at 9:50 PM <don...@desinc.net> wrote:

Did internal frame coatings (boeshield, frame saver) make any difference on readings? Or a different frequency that you filter out from the transducer?-- 

I haven't yet specifically tested for internal frame coatings because it hasn't seemed necessary. This device is intended to measure industrial equipment in situ and in use, so I don't imagine having "stuff" in there is that big a deal. I can confirm that it seems to accurately measure the top of the seat tube even when there's a bunch of grease in there (but no seat post).

I am currently testing the reliability of upper seat tube measurements with the seatpost installed. It's too soon to say anything with confidence, but my initial results seem valid.

The Elcometer PTG6 that I'm using is quite simple to operate and doesn't let you directly meddle with frequency or anything like that. It seems to be intended to be handed off to a lower-wage employee who then wanders around measuring stuff and recording results. The Olympus devices offer much more raw data (and much more complexity). I might have chosen on of those instead, but I found out about them after I'd already bought this one. Probably just as well. I'm happy with the speed and usability of the PTG6.

Best,

Reed 

[John Clay]

Very cool Reed! That will help paint the picture.

Carry on!

John Clay
Tallahassee, Florida