So I have been reading all those old Bicycle Guide reviews SpeedofLite has been posting. And, as has been my observation over the years of weighing the occasional bare frame - they vary enormously in weight for similar lugged traditional diameter road bikes.

Clearly, frame size is usually a bigger influence than anything else, but it seems easy to find frames of the same size that vary in weight by more than a pound. Sometimes the lighter tubesets are in heavier frames than the heavier tubeset. So I would love to hear what people that actually braze tubes together think is going on.

Factors I can think of:
1. Obviously, we don't know who's scale is right.
2. Lug thinning, or not. Pricier bikes are often lighter than production bikes with off the shelf lugs.
3. Other fittings. Brake bridges can be hollow tubes or solid castings. Etc.
4. Which end of the tube was trimmed?
5. Amount of brazing material flow.
6. Worn tubing mandrel/dies? (Seems less likely)


So do you buy that an Aelle frame can be obviously lighter than Tange #1, and why? I realize that sometimes there are some tube substitutions, but that really shouldn't be more than 20-30 grams a tube.

Thanks, framebuilders!

Great question. When I cut up my old failed Trek frame I found a piece of brazing rod a few inches long in at least one of the stays (can't remember if it was seat or chain). I guess they were feeding it in through the tube and it got stuck! But it wouldn't account for a great deal of weight. However the stays themselves could account for some of the difference I guess, because the Reynolds/Columbus sticker often only refers to the main tubes.

I've measured tube wall thicknesses that were close to 10% thicker than listed. Spec charts are not always what goes on the production floor and tube decals don't always match up with the actual tubes used. Steerers and stays are often not the same grade as the rest of the frame and are a cost savings "location". Cast lugs and other fittings often weigh more than the stamped and folded examples. And then there's how the frame was weighed, not as in which scales were used but what was included or not. Did the frame still have the fixed cup and headset parts on it??? (Back in the 1970s as I became aware of fancy Itl. bikes I noted the often sub 20lb weight claims. Well, for a 19" framed bike with no tires that's about right. Andy

Consider the frame fittings as well. E.g. investment cast bottom bracket shells and fork crowns sometimes weight considerably more than bulge-formed ones.

Many fork crowns are solid steel. Not hollow. Cinelli bottom bracket shells were thick and heavy. Straight gauge tubes were thick and heavy. Small details all add up.

A long time ago I'd buy Columbus SL and SP tubing in bulk volume to get a better price. SL chain stays varied in weight. I would weigh each one so I could find the ones that weighted about the same to pair them together, What I not perfectly recall is that the variation was between 140 and 170 grams each. I would then use them accordingly to put my lighter customers/students on the lighter chain stays.

Another variable is that stated wall thicknesses of almost any brand of tube and actual measured wall thickness seldom agreed. And of course where the butts are located and where they are supposed to be according to the manufacture was another variable. And how long the butts are on any given tube model could vary considerably and then how much is trimmed off when mitering becomes another variable on frame weight.

Doug's comment about weighing stays/blades is what I have been doing for quite a while. I haven't seen the amount of variation between "the same" stays that he recalls but they do vary. (I suspect the wear tolerance range of tubing dies and mandrels has gotten smaller since the 1970s when Doug and I started doing this stuff).

When I receive tubes from a supplier (not too often given my 1 frame a year production) I do a series of checks. For main frame tubes I mark the butt transition portions with a butt checker and I roll the tube on a flat surface to mark the tube's bowing high point. For stays and blades I weigh them, note this on them and pair them up based on weight. (IIRC blade or stay might vary by a few grams). I wrap some masking tape on the tubes with any data I've discovered then coat the tubes with LPS-3 for storage rust control. Andy (who now can't find his little electronic scale...)

Some years back, when living in the Cleveland area, I visited Joe Bringhelli one day. He helped me greatly back then, inviting me into his home and workshop, teaching me much.

Joe wasn't one to worry about frame weight, but he didn't hold it against me if I did. To a point, one day when asking about the weight of such and such a tube, he handed me a box of the tubes in question, along with a scale, and said do as you like. I sorted through a couple boxes, picking the lightest tubes I could find. One such tube was a Zero down tube, with 31.75 at the head tube and 35mm at the bottom bracket. The "baseball bat" tube. I also picked out a nice light 31.75 Zero top tube. Variation in these tubes was more than you might think. Between the two tubes, I saved about 50 grams. Ha ha. 50 grams = 2 ounces. I still have those tubes. They must be 25 years old now. Gotta get off my butt and get busy...

More than 20g difference between a Cinelli (solid, 158g) crown and Henry James (hollow, but also investment-cast, 126g):


I was a bit surprised that the Cinelli investment cast BB shell (160g) was only 12g heavier than a Nikko bulge-formed shell (148g):

Cinelli's even can weigh less: 147gr..........

I don't have an Italian to weigh but I bet they're even lighter, since the outside diameter is the same and the I.D. is 1 mm larger.

This is a metric BB ( Cinelli SC France ) , Swiss threads, from the days Reynolds 753 was only available in metric .

Another variable is that the sticker on the seattube might mean that the seat tube is that tubeset or it might mean the front triangle or all 9 tubes are but you can't tell until you cut it up. I have also found big blobs of brass in repair projects once I cut them open. Framesaver might account for some difference as well as paint/powder coating. No idea how much.

As a builder would anyone send back a tube that was more than 15% heavier or lighter than the spec? What about if other specs were significantly different? And yes, obviously if the wall thickness is thicker, the tube will be heavier so, you have compounding 'out of spec'-ness.

I have sent back/complained about fork blades that didn't seem right - the supplier sent me new ones immediately.

One place I worked we used a lot of Reynolds "Jack Taylor" oversized tandem blades. We got one where they tapered the wrong end, so the top was too thin, not strong enough, but the bottom was so thick it was almost a solid bar, just a little hole down the middle. You could still see where they stamped REYNOLDS BUTTED, which is supposed to be near the top but it was near the bottom, all scrunched up from the swaging - hilarious! Wish I'd taken a picture but I was too poor back then to afford film or developing.

Like this one?




Not my pic, saved from the internet and I forget where, but the person who posted it said the blobs are definitely brass. WTF? Brazing by dipping in a big cauldron of molten brass, then forgetting to drain out the excess?
Not much of a cost-cutting method if you put $50 worth of brass in each frame. There must be some other explanation. Stiffening, for a sprinter? The mind boggles.

We saw a few of those at Trek, as well, with standard 531 blades. And misprinted 1010B dropouts. And other "Inverted Jenny" type flaws on various parts.

Yes, the one that comes to mind was a mid 80s Trek 720 touring bike. I hit some ice and stuffed the front end into a 'two man' rock that was sitting beside the path. I buckled the DT so I thought I would cut it up and see what was inside! I can't find the picture ATM but one chainstay was half full of brass from the dropout end... I assumed the brazer was trying to show how much he was doing based on how much brass he used instead of units produced.