PeteHski

Seems like a lot of effort for very little gain in accuracy. If you just measure the bar drop directly to the floor at the saddle and bar without shims it is going to be very close on any reasonably flat floor. Measuring vertically up at both ends from a 1 or 2 deg slope is not going to make any significant difference to your height measurements. Even with the floor perfectly level I just eyeball the tape measure when measuring up.

The only time a level floor really matters is when projecting a plumb line down from the saddle to measure setback. That part I now fully accept following the earlier discussion. Fortunately the floors around my house are level enough that I'd got complacent and overlooked that potential error.

Unless you are trying to set your bike up in a ploughed field I don't see any point in going to all the trouble of levelling the axles and then projecting a level off the saddle to measure bar drop. I would expect more compound errors using that method than simply measuring the vertical heights directly off the floor at the saddle and bar - even if it is slightly sloped.

Road Fan

Hello Pete, what's the app? I'm a Kickr Snap user.

PeteHski

The App is only relevant for the Kickr Smart Bike i.e. You take a photo of your bike and then it calculates the specific setup parameters on the Kickr Bike to match.

Kontact 

Do whatever you want. My posts aren't based on experience as a carpenter, but as a mechanic and fitter that worked for many years in fit intensive shops and saw first hand how often errors compounded when trying to measure and apply fit data from one bike to another - usually by fairly careful fitters and mechanics. I'm not trying to make dire predictions about the level of your floors as much as illustrate some of the many sources of error and the degree to which they can effect the results you went to great pains to produce.

That includes all the weirdness that happens when people try to take a fit off a bike and apply it to one with dissimilar saddle, hoods or bar. Or when you measure off an incrementally ratcheted seat post to another that doesn't have the same increments. People can be enormously sensitive to very small variations in angle and location. Some people are not - but either way you go to a lot of trouble or expense to get your fit right - there's no point in being sloppy after you have the fit done.

The best fit is when you put your own bike in a trainer, level it with the rider's weight on it (because the front tire sinks and the rear is supported by the hub), and then make all those modifications to the bike that is actually going to be ridden. This is just like the way a suit is tailored and is the most accurate.


So there's nothing to debate. Do whatever you like. I'm just providing the potential pitfalls and everyone can come up with their own solutions, or ignore my experience entirely.

PeteHski

Well I was big enough to admit that I stuck my foot in it initially regarding measuring saddle setback on a wonky floor. But the rest I suspect you know is correct. You have added quite a lot of smoke and mirrors there to cover yourself without really addressing the specific points I was picking up on. But that's fine. I'm happy enough with what I'm doing.

PeteHski

I admire your OCD here, but I presume this bike is going to be ridden on real-world roads which are not perfectly level. So does it really matter if you do your fit on a very slight slope? One thing I really like about the Kickr Bike is that you can see how your fit works out when climbing varying gradients.

Road Fan

Ok, but what's it called?

I typically put a bike in the Snap or in the old Cyclops, prop up the front wheel to level the TT or level across the axles (top of the QRs), then fix the front wheel as well as possible. I haven't seen where front tire compression makes a significant difference In this process, but perhaps Mrs Road Fan and I are not very picky, nor are we in demanding cycle-centered professions. I use a long bubble level with electronic level built in (the one Craftsman used to sell while under Sears). Usually with this low-cost tool the digital agrees with the bubble. But after this "level-setting" is done, I don't care about the floor, as long as the position the level indicates is adequately consistent from day to day. I think it is sufficiently accurate with respect to an absolute for level (the line supporting the bicycle is tangent to the Earth and therefore the normal at the contact point contains the center of the earth). That would be my definition of "level." While I firmly believe it, I don't believe I need to satisfy it nor to prove that it is satisfied. After using my long bubble and checking for reasonable digital agreement, then from day to day or month to month, I only need to be consistent within the resolution of my level and my protractor. My keys for comfort are saddle pitch angle and height. I can't say my setup rivals or pwns anyone else's in terms of accuracy, but I am just adjusting my bikes and those of my wife, not servicing a racing team, a group ride of a group of top-end customers, or a RAAM competitor. When setting the fit on Mrs. Road Fan's Terry with compact frame, I don't worry about whether the TT is level, and use the wheel to wheel approach instead to get level indicated on my long bubble/digital model.

So for most of us, I don't believe we need the absolute reference point of a level setup, rather, if the comfort in some measure has changed after a long day I need to check how much it may have changed from the previous settings, not start again from first principles.

Kontact 

Cover myself? I was sharing years of experience, some of it from one of the oldest and best fitters in the US. You've been rude and dismissive throughout.

I shouldn't have posted anything.

PeteHski

Kettle meet pot. Looking back here I don't actually think I was rude or dismissive throughout. I simply questioned a few of your comments about how difficult it really is to transfer a setup from a Kickr Bike. I had a bit of brain fade regarding your initial explanation about saddle setback error and admitted it. But then you totally dismissed all the other valid points I raised by swerving your way around them. Correct me if I'm wrong, but the only thing you have really shared is that you don't think most floors are level enough to measure a bike fit adequately.

PeteHski

It's actually embedded in the Wahoo App when you add the Kickr Bike as a sensor. It's of no practical use to you unless you happen to have a Kickr Bike.

Kontact 

What points?

What I shared was a front to back method using simple tools for documenting and using fit data to minimize error - which was the point of me speaking up in the first place. As I've seen professionally, fit data gets screwed up constantly. You now understand the math, the pitfalls of certain methods and I even described a method that I invented so anyone can use it.

You're welcome. Bye.

Kontact 

I missed this excellent question. Why do fits at all? Why not just put your heels on the pedals to set you seat height? Why adjust saddle angle if hills exist? Don't red cleats prevent any knee trouble? Can't you just move your hands around if they get numb? Why ride so far if it is uncomfortable? Stop for a beer.

PeteHski

Yet again you don't really address the point I made. I happen to think fit is very important. But I'm actually asking if you think it is really important to do the fit while taking into account front tyre squish on a trainer? I don't think so.

PeteHski

Well you still haven't explained to me why I can't just measure bar drop vertically from the floor at the bar and saddle, even if the floor might happen to be slightly sloped. You tried to tell me earlier that this would introduce a significant error, but I don't believe it does. Let's say my bars are 1000 mm off perfectly level ground and my saddle is at 1100 mm and my floor is on a 2 deg slope. So I measure vertically at both points, but effectively with a 2 deg offset due to the slope. So applying trig I actually measure 999.4 mm at the bars and 1099.3 mm at the saddle. My true bar drop is 1100 - 1000 = 100 mm. But I measure 1099.3 - 999.4 = 99.9 mm (0.1 mm total error). So where is this large error you asserted earlier? :-

Kontact 

Front tire compression will account for a half degree or more of downward slope. Which means if you wanted a level saddle, it will now be a half degree nose up on the road. People can feel that.

Kontact 

Because you were talking about transferring dimensions off the Kickr, and the Kickr doesn't get leveled. It doesn't have hubs to measure level through. Instead, you put the components where the rider feels they should go and have to measure off true level. You could do a fit on a 3 degree slope and the results would be based true level, not the floor. If you now measure to the floor under the Kickr you are introducing the floor slope to the previously pure data.

It is the same problem as failing to account for front tire squish and then treating the bike as level when it isn't.

Barry2 

Gents,
This thread is now stuffed full of theoretical vs practical.
Please accept you won’t convince each other.

just as I’ve accepted neither of you will use a 48” t-square and remove all floor level issues from the equation.
I will concede, I do have to place my digital level in “inspection” mode (it sets zero to the floor) for leveling saddles.

all the best

Barry

PeteHski

On the contrary, I think your T-square is an excellent idea if your floors are not level. Especially for the saddle setback measurement.

PeteHski

I guess I just don't know any roads that are so perfectly level that this would make any practical difference. I can feel a 0.5 deg change in saddle angle, but only if the saddle angle physically changes relative to the other contact points. If the whole bike is simply going up or down a 0.5 deg slope then I don't feel the need to adjust the saddle angle to compensate. I don't even feel the need to change saddle angle while riding up an 8 deg alpine pass for several hours.

PeteHski

Okay I see what you meant now. So if I happen to set my Kickr Bike up on a 3 degree slope and then measure bar drop to the floor then transfer that number to my road bike on the same slope it will be wrong. Fortunately my floor is actually level within 0.1 deg where my Kickr Bike sits, so it doesn't generate a significant error.

Road Fan

They're also approaching fitting from totally different points of view.

Kontact is a businessman in cycling and needs methods to give correct and defensible results to any and all customers who come to him for service. He may or may not need such accuracy for his own riding, but that isn't what matters to him.

Peteski is a rider and self-fitter who is trying to make his bike feel better. Proving it is accurate is mainly a matter of what feels good and what he is able to think about doing within a time and cost budget.

Both can be correct, but are ultimately they are not comparable, unless one wants to make changes, for some reason.

From my engineer's point of view (whatever that is worth!) Kontact is more likely to be correct in some absolute sense, but that doesn't mean I will spend what it takes to replicate his discipline, professionalism, or level of knowledge. If I need such an accurate fitting, we have an expert here in Ann Arbor.