Santana’s Frame Deflection Test Rig, aka “Otto” (Part 3)

Again, if you haven’t read Part 1 and Part 2, you might want to do that first.

Picking up where we left off in Part 2, with Santana’s old test fixture no longer available and their efforts to successfully use Finite Element Analysis (FEA) to further enhance their tandem frame designs still maturing, Santana decided to adapt Cannondale’s ‘Otto’ frame resistance test fixture for use on tandem frames and fully-assembled bikes.

Cannondale's 'Otto' - Bicycle Frame Resistance Measurement Tool

The Results of Cannondale's Otto Measurements

Just to add some background here, Manfred Otto is known for his work at EFBe Pruftechnik where he is now General Manager.  You can find more info regarding EFBe at their Website.  More specifically, you’ll find that EFBe offers various types of  independent rigidity tests for frames and sell precision test fixtures similar to Cannondale’s (and now Santana’s) ‘Otto’ rigs, aka. the STERW depicted in Part 2.

A more recent photo

A search of the web on EFBe will also reveal an interesting English translation of a 1997 TOUR evaluation of lightweight racing frames. The frames were subjected to some of Manfred Otto’s earlier bicycle frame testing and the article is hosted by Damon Rinard’s mirrored site on; definitely worth a read.  You’ll also find a lot of Cannondale promo material, various other studies and a few articles by cycling enthusiasts with what appear to be technical backgrounds (but who do not appear to be in the cycling industry) who offer commentary on EFBe testing. If you look really hard you can even find Manfred Otto’s commentary on some of those articles and a few others.  It’s especially telling to see where Mr. Otto agreed with the point made in a 2004 posting to Cycleforums where marketing ‘spin’ of EFBe test results was carefully dissected to illustrate how test results can easily be manipulated to create false conclusions.


TG: Without further ado, Bill now shares more about how and why Santana developed their own version of the Cannondale “Otto” frame efficiency jig as a follow-on to their electronic strain gauge testing and more recent efforts to adopt Finite Element Analysis.

BMcR: In the meantime, a German bike testing guru named Manfred Otto, who was aware of Santana’s earlier test jig through conversations I had with him, designed a simpler yet more elegant version of a “pedaling” deflection test bed for Cannondale. After Cannondale named their jig “Otto” — this became a term to describe a frame efficiency test rig that uses a weights plus a specific angle to duplicate the force of a rider standing on a pedal (as if sprinting or climbing).

TG: Bill now offers some specific comments that respond to questions I raised after seeing photos of the Santana ‘Otto’, which included:

  • “The frame doesn’t seem to be the focal point since the vast majority of measured deflection when going from static to loaded would likely come from the “weakest links”, i.e., the fork, rear triangle and wheels/tires.
  • In other words, a bike with a fork like an Alpha Q or Rolf wheels will certainly have more deflection along the centerline of the tandem at that severe angle than the very same frame with a chromo fork and 36h conventionally spoked wheels.
  • It would seem to get just the frame/fork deflection you’d need to lock the rear drop-outs into the fixture and, even then, you’re really only capturing the bottom end, axle-to-axle and not the top tube influence.”

BMcR: The “extreme angle” you noted is based on Manfred Otto’s photographic research of single bike racers.  Cannondale used this rig to evaluate various frame designs for their European Pro Team. Other builders and teams either commissioned their own “Otto” rigs, or copied it.

After years of discussions with Manfred Otto, last year Santana designed and built our own Otto. As opposed to Manfred’s previous Ottos, our device can be used for tandems, and can simultaneously test the pedaling stiffness of forks and wheels. Thus it not only tests the pedaling efficiency of tandem frames, it can also test the relative pedaling efficiency of built tandem bikes. Either way, the test weight we use is 300 pounds. This duplicates the force of a 170 pound captain and a 130 pound stoker—both standing on their left pedal.

In the factory we plan to use the Otto to verify, refine and calibrate our FEA program. As you noted, the device is most instructive when we jig a frame without a rear wheel or fork.

I disagree, however, that the inclusion of the rear wheel and fork (as we were using our Otto in Solvang) is less meaningful because of a “weakest link.” Performance is performance. If you want the “fastest” tandem, ignoring the efficiency of rear wheel or fork would be unwise.

TG: Bill discusses the limitations of the Santana ‘Otto’ and challenges with regard to scoring data collected in their study:

Note: I have at certain places in the following text replaced specific references to other tandem models, brands and specific people who were used as examples in our correspondence with more generic ones by the use of [bracketed text].

BMcR: Like all methods of measurement, the Otto has drawbacks. First, it does not duplicate the up-pull of riders pulling on their handlebars, or the right pedal. Second, the “pedaling” stiffness of bars and stems and front wheel are ignored. Third, the single readout answer is a bit simplistic (but repeatable and meaningful).

At this point the biggest unknown is how to score the pedaling stiffness versus the weight. Even if we eventually come up with a standard equation, [extremely lightweight and/or weaker tandem teams], for instance, would fare better on a lighter and flexier frame than [a heavier and/or stronger team. The lighter and/or weaker team] should pay relatively more attention to the scale than the Otto.

TG: Bill notes his belief that there’s too much emphasis placed on frame weight vs. efficiency and lays the foundation for the value of the deflection measure that comes in the last paragraph of our exchange:

BMcR: As a general observation, weight is too easy to determine. A lot of riders cannot perceive pedaling efficiency (even on a single bike), and confuse efficiency with handling or harshness. Most bike design experts agree that “racing” frames tend to be too light and flexible for their riders. Even though the owners of these bikes would experience better results with a heavier and more efficient frame, they’ll instead choose a too light frame because they have no way of judging frame efficiency.

TG: I noted in my original note to Bill, “I’ll take a wild guess and assume your frames do pretty well on this test given the oversized rear stays, 160mm rear wheel and beefier forks y’all have always used, e.g., the Reynolds Ouzo and presumably the current house-branded replacement.”

Bill responds and — as one might expect — speaks highly of Santana’s materials and designs in subjective terms that I suspect the tandem frame deflection project will attempt to quantify.  The last paragraph, below, is one that bears some degree of attention as it once again makes the point that lightweight for the sake of lightweight could be short-sighted when trying to build the most efficient tandem for a given tandem team.  Frankly, I don’t necessarily disagree with this view either.  I have certainly seen tandem teams riding equipment that is not as well-suited to their size (i.e., mass and power output) or how they actually use their tandems. Of course, this also extends to geometry choices, wheel and tire selection and the like.

BMcR: Will Santana tandems score well?

We know for sure that we use better and stronger grades of Aluminum and Steel than any other tandem builder. The thinner resulting gauges allow us to use larger diameters in critical areas without the weight penalty. Our aluminum and steel downtubes, bottom tubes and chainstays have greater diameters AND more efficient shapes. (Round bottom tubes as well as larger top tubes are relatively inefficient). In Titanium we use the same alloy as other builders, but use tubes with more efficient shapes, diameters and butting ratios. In IsoGrid carbon, we attain greater efficiency because our tube (built with the same plies of carbon as [certain composite tandem frames]) is more efficient due to its internal ribs. Our joint weight is lower due to welding. Additionally, our IsoGrid tubes are double-butted. Our IsoGrid bottom tube is ribbed AND oval shaped AND double butted AND laterally butted.

While lots of builders claim to make a more efficient tandem frame, they make no attempt to substantiate their claims. While an Otto device may seem crude, it is certainly no less meaningful than a scale.

PostScript (P.S.)

Bill and I shared a subsequent exchange of Emails that revisited the potential wild card in testing a broad range of tandems where some of the more high-end racing tandems are purpose / custom built for teams of varying size (i.e., height & weight), power (pro vs. club riders vs. recreational / sport riders) and purposes (race vs. sport).  The following is the question I posed:

TG: “Are y’all also recording the team weights of the tandems that you’re building?   I only ask because frames such as the ones we own were pretty much built for our weight and riding style or that do, in fact, have somewhat low upper weight limits, e.g., I think [several certain tandem designs have] a recommended max team weight at which point a team would be better served by [a more robust tandem].”

BMcR: A [certain brand of tandem] owner who attended our tour in Hawaii claimed [his builder] pushed him to a heavier weight frame because of his weight. [The builder purportedly] wanted to protect both parties from a broken frame (which is entirely reasonable!).

This is NOT the same as saying the heavier frame is somehow less efficient. [Some teams], for instance, [are] not only lighter, [but may also be] weaker. [Another equally light team], because of [their greater] strength, might be faster on a heavier [tandem].

To try to make this clearer, a stronger and lighter elite racer will need a single frame that scores higher in an Otto test than the solo frame [a recreational / sport cyclist would] need. Because [the recreational cyclist is] weaker, [they] might choose a frame that is lighter.

I will note, however, that this is NOT the point of Otto testing. Otto testing is done to measure pedaling efficiency. Both heavier and STRONGER riders will want a frame that scores well in efficiency.

Most frame builders are happy to sell frames that are too light; as long as they don’t break. In Santana’s case, we have certain “pedaling stiffness” benchmarks that we believe are essential for most couples.   This (and not the risk of a warranty claim) is what causes us to use larger and/or heavier tubes than are possible. Could Santana build lighter frames? Sure, but these would be optimal for WEAKER teams, not for LIGHTER teams.

TG: Finally, because frame stiffness and rigidity beyond the amount needed for durability and control is sometimes called into question by everyone from non-cycling industry hobbiest / engineers to the likes of Bicycle Quarterly’s editor, Jan Heine. in the Summer ’09 edition of Bicycle Quarterly, I wanted to introduce that subject and asked: “As for frame / bike deflection being good or bad, have you had a chance to read through [what BikeThink’s author has published]?

BMcR: I’ve read this and a number of similar (misguided) articles. While a spring stores energy, it is dreadfully inefficient (more energy goes in than comes out). The guys who write these […] “stiffness is bunk” articles can’t figure out where the balance of the energy goes. (The answer is heat.) In any case, a bicycle frame is a very inefficient spring.

(If these guys were right, there are thousands of professional bike designers — including hundreds of Engineers — that have wasted more than a Century of combined effort.)

Post-PostScript (P.P.S.) – Lest anyone think otherwise….

TG: Forgot two things: 1. May I share this info on the test?

BMcR: If you understand it!

TG: 2. Anything exciting from Taipei?

BMcR: Lots.

TG: It will be interesting to see how the data plots over time.

BMcR: I certainly agree.


About TG

I've been around a bit and done a few things, have a couple kids and a few grandkids. I tend to be curmudgeonly, matter-of-fact and not predisposed to self-serving chit-chat. Thankfully, my wife's as nice as can be otherwise we'd have no friends. My interests are somewhat eclectic, but whose aren't?
This entry was posted in Advice & Commentary, Technology & Equip.. Bookmark the permalink.

3 Responses to Santana’s Frame Deflection Test Rig, aka “Otto” (Part 3)

  1. Paul Haussler says:

    Well done Mark. I remember reading Bill McCready’s articles in Bicycling in the ’70’s (Ithink). I thought he was crazy to adopt a non-standard rear hub width of 140 mm. I know better now. I always respect Bill’s opinions and insight. We are still riding a “97 Arriva, with a few upgrades, but the bike has some upgrades.

    I always appreicate your opinions and insight too.

    Paul Haussler,
    Huntington Beach, CA

  2. Wayne Sulak says:

    Great post. I especially appreciate your question regarding the desirability of frame flex. I disagree with Bill but I respect his opinion and experience a great deal. Most importantly – can we expect some data of tandems he has tested?

  3. Tandemfahrer says:

    Mark, I just now stumbled upon this old post following a discussion on the German tandem listserv. I agree with Bill in all points except for one: He keeps calling large top tubes unnecessary. In terms of pedalling efficiency, which is sort of what he is measuring both with the old Tarantula and with the new Otto fixtures, he is certainly correct.

    But what about what I will call ‘anti-whiplash’ stiffness along the top? This is necessary for hard steering maneuvers with a seated stoker. All our three current Santana tandems (two Al, one CroMo) are stiff enough in this regard, but I remember testing the original Dual Moto FS tandem with its extremely low top tube. At least with a male stoker I could not get used to the flexibility of that frame. Others have said the same about test rides on Santana Quads (even though AFAIK these use thicker-wall unbutted tubing for the top tube). I can imagine that touring with a lot of weight on the racks one might also benefit from riding a tandem with more stiffness along the top. This could be easily measured by fixing the frame at the headtube and deflecting it at the rear seatpost.

    Traditionally, Santana has used the same tubing on both road and off-road tandems. It’ll be interesting to see how their new off-road and touring tandems will look like.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s