Open Source List of Frame Measurements?
 

Hello!

Does anyone know of an open source dataset of frame measurements as described by the manufacture?

To illustrate my idea, I created https://docs.google.com/spreadsheets/d/1Hchy-HQwfRFUSmpGENmRIY1ephZ0pDnnIvk9QaDpYMc/edit?usp=sharing]this workbook [/url] in Google Sheets that contains information found on BMC's website. Basically, I'm looking for a list of bike frames with the associated information in tabular format that's relevant to frame building. Ideally, this dataset would be queryable in Excel or a similar software.

Thanks!

I'm unaware of such a database.

Back when I got serious about riding and later framebuilding in the 1990's I made my own spreadsheet of specs of frames in my size. Take away was that there was very little difference. There were a few outliers with steep angles or slightly longer chain stays but overall there wasn't much variation. I frankly doubt that experienced framebuilders would use such a database should it be available. Most custom frames are custom fit for the intended user and purpose so a database of off the shelf frame specs wouldn't be very useful.

That's a great reply, thanks! I asked because I'd like to get my head around the categorical differences in frame types by looking at geometries over large number of bicycles types. Like was do gravel bikes look like, what do mountain bikes look etc....

I'll keep my eye out, thanks!

I agree with Nessism. Having said that...

Of course someone has done this! It was (cleverly) called the Frame Geometry Project!! Looks like the most recent bike was 2008 Rivendells.

https://www.hiddenfortress.org/geometry/index.html

Wow, this is exactly what I had in mind. Thanks for pointing this out. Wonder how the info was gathered. I'll inspect the backend once I'm near a computer.

On a side note - when you say every frame is different. Is this down to every angle and tube length? I mean, are there no rules of thumb or windows of measure used on a given frame type you consider while designing a frame? For example, do headtube angles fall within 2 degrees of each other for 80% of road bikes?I imagine there are general rules, and that's what I wanted to look for in the data. But this approach might be short sighted!

Thanks again!

I think your theory is valid. The underlying reasoning might not be what you expect though.

You can see how bikes from the '60s compare to bikes from the early 2000's for a specific type of bike. Road racing bikes have gotten steeper because they are more specialized and roads are better. We used to ride road bikes on gravel before we had gravel bikes, they had to be more stable. We cared about comfort then so the bikes were longer (42-43cm CS) and ST/HT angles are shallower to provide some extra flex and room for (gasp!) fat tires like 27x1-1/4. Now race bikes can be short and steep and designers can add chunks of rubber or different carbon layups to provide compliance.

Road bikes can ride really well with head tube angles of 71* to 74*. They will ride differently but both can handle well. In nearly all cases, the same model of road bike from the same manufacturer will have, for example, a 72* HTA with 74* seat tube angle on their smallest bikes and 74* head tube angle and 73* seat tube angle on the largest bikes - not because of handling but because of safety and production process requirements. Chainstay lengths should increase for larger bikes as well. For the largest sizes, keeping chainstays short and HTA steep keeps the large sizes compact so the bike can still fit into the same box as the rest of the sizes - adding a new box to the inventory is costly. Also, they usually use the same fork which is the real compromise. However, the bikes ride predictably within those ranges so it's fine.

The point is that production geometry is as much a result of business and production decisions as bicycle design decisions.

Anyway, enjoy comparing bikes across types and eras!

it's pretty interesting to see what has happened with mtb's in the last 5 years. Used to be less slack head tubes and 100mm travel forks. Now it seems like they all have slack head tubes and 120mm forks minimum. Pretty much all have the same amount of rake though, 45mm is sacrosanct.

Not to mention that road/gravel bikes generally have around 45mm of offset as well.

Now that's insight. This makes me feel dubious about mass produced bicycles, but real about the tension between perfection and continuity. "Why can't the company simply increase the box sizes when the design calls for dimensions larger than a standard box allows?" - "because we all need to eat and keep the lights on."

I'm less interested in the currently popular culture of frame design, than I am more interested in what makes a cross bike tick or a road bike glide. Sounds like I need to be careful with this approach. Thanks again for the insight!

Don't get me wrong. A production Specialized/Trek/Giant in the $2-4000 range represents amazing value if a 54-58cm size range bike fits you. They are well designed and engineered and ride really well - they can be great bikes! And the folks who are designing, spec'ing and building them are great folks trying to do the best they can within the parameters they are given. I'm not putting them down at all. It's the same in EVERY industry from shoes to cars.

As the rider and their needs move further from center though, the production world has to introduce compromises. It pains me to see a 4' 8'' rider on a $8000 carbon road bike that I know could fit and handle much better AND be way more interesting.

Totally agree that the good companies are generally doing a great job.

i wish bikeinsights.com and geometrygeeks.bike considering to add what type of tubing the bike listed.

That online XL DB is gone. I think GeometryGeeks supposedly absorbed some of it, but many of the old HiddenFortress geometries on are not GeometryGeeks.

Do you know if the XLS doc exists anywhere else? Can't even find it in WebCrawler

No idea. If I remember correctly, wheel/tire size wasn't a measurement it listed so, it was limited usefulness for framebuilding.

This conversation reminds me of a recent post on the framebuilder facebook group. The person had found out that everyone used 405mm chain stays, built a bike with chainstays that long and the tire hit the seat tube.

As opposed to, find the rider's riding inseam, decide on a % of that you want to be seat post showing. Use the torso/arm length measurements to place the handlebars, decide on a stem length. Make the wheels fit. Done
Okay, I might have glossed over some details, but why do I care what measurements Grant Peterson (or anyone else) made a production bike so it would fit as many people as possible?

My apologies if someone is making a multi-bike run of production bikes, carry on.

Totally agree. Aside from key measurements missing (like wheel size?!), without an objective assessment of the ride qualities, it's just data. And who knows how accurately/correctly the measurements were taken.

Back when i poured over published specs for geometry, specifically steering stuff. I spent hours (pre computer spread sheets) making charts and calculating trail and castor angles (as Bill Boston used it) trying to figure the numbers that made a bike handle right. In time I came to the conclusion that there was so much more then mere numbers that made a bike work well. These days I work within a narrow range of steering geometry for my stuff and look to the rider's fit more.

I'll add that the range of modern (talking from the mid 1970s on to current) bikes' geometries is rather narrow. If you superimposed every frame on top of each other (limited for a "size and style") the results would be a very dense average with variations falling off in numbers fairly quickly. I'll also add that some geometry numbers are driven by regulations and component needs and not fit or handling. Andy

Maybe there is a "most" condition, but quite a few bikes today have less trail and more offset.

There are a number of gravel forks that have changeable offset, 45 or 55. I talked one of the shop mechanics into swapping to 55. He hasn't done it yet.

Seems like there is one with just 55mm offset, but I think most companies are afraid to do that

Well, I would care about that when I'm thinking about whether that bike that Grant is selling would work for me.

Grant should supply you with the geometry. Not sure I trust a excel spreadsheet from some shady internet site.

I should get the subtitle on the forum changed. Most of the suggested topics would be better discussed in more general forums. And usually are.

It seems you would see no situation in which such a spreadsheet ("open source") would be justifiable. Correct?

Yes, the data has to come from Grant, in this case. But bike companies close, sometimes bike companies don't proofread what's in their tables, sometimes measurement standards seem to change from one model year to the next (such as Trek), designs get changed, and sometimes the old data is taken down from the company's public displays. And in those cases, a person considering several bikes that are not current production, might want to make comparisons, say between a 1950s English Club-style bike and a Boulder Randonneur, just for one weird example, which in my brain is quite real.

So for such situations it might be useful to have a stand-alone place to look for such data. We always have to judge for ourselves if the specs are correct and useful.

I just don't care. I see this forum as being about building frames and geometry in the abstract is a side show.

Also, I don't trust a spreadsheet. How was the geometry derived? I find it somewhat difficult to determine the geometry of a built bicycle frame and I don't trust anyone else to measure a bike either. And manufacturers are somewhat untrustworthy as well.

Geometry is part of the fun of building frames. I think it's pretty useful to know what other bikes are using just to get a ballpark, to assess their marketing BS, and to decide what to use. Handling and rider fit are both considerations and both are a bit of a black art.
​​​