Backyard aero testing
I am enjoying some down time after BLST 70.3 and before IMFL. I had a bike fit with TTBIKEFIT and was please with the process. We brought my seat down, brought my pads and extensions back and last was to drop the front end by removing pedestal sun the pads.
Lower front end! Yahoo! I just got faster! After riding on the new fit for a few weeks I came to a few conclusions - the new saddle position was great. But the lower front end was hard to adjust to.
So this past weekend I thought that I would attempt to quantify how much time I would save at a given power output with the lower front end. Mostly to help keep me motivated to keep with the lower position in homes that I would adapt.
I used the Chung graphs via golden cheetah. I tested 3 positions - sitting up on the pursuit bars, with 2.5 cm of spacer under my pads, and without. Sitting up was clearly a killer in all test runs. But in terms of calculated Cda and more basic sped vs watts.
I was surprised however to find that there was no difference between the two aero positions in terms of Cda. I ran so many laps that I wore the pavement down. The software said no difference. Past that, my own power and speed data for the two positions was identical.
In fact, there was only one consistent difference in the data between the lower position and the higher position - my HR was 2 to 3 bpm higher at the same watts in the lower position. Every time. No exceptions.
My guess is that I just don't have the hip and lower back flexibility to truly lower my front surface when I lower the pads. I just end up with a bigger bend in my mid back. Not sure. But I am sure that sitting up is deadly so I will be avoiding that at all costs.
So, till I can get to LA someday to test like a grownup, I'm keeping the spacers. Even if my bike looks cooler without them.
Lower front end! Yahoo! I just got faster! After riding on the new fit for a few weeks I came to a few conclusions - the new saddle position was great. But the lower front end was hard to adjust to.
So this past weekend I thought that I would attempt to quantify how much time I would save at a given power output with the lower front end. Mostly to help keep me motivated to keep with the lower position in homes that I would adapt.
I used the Chung graphs via golden cheetah. I tested 3 positions - sitting up on the pursuit bars, with 2.5 cm of spacer under my pads, and without. Sitting up was clearly a killer in all test runs. But in terms of calculated Cda and more basic sped vs watts.
I was surprised however to find that there was no difference between the two aero positions in terms of Cda. I ran so many laps that I wore the pavement down. The software said no difference. Past that, my own power and speed data for the two positions was identical.
In fact, there was only one consistent difference in the data between the lower position and the higher position - my HR was 2 to 3 bpm higher at the same watts in the lower position. Every time. No exceptions.
My guess is that I just don't have the hip and lower back flexibility to truly lower my front surface when I lower the pads. I just end up with a bigger bend in my mid back. Not sure. But I am sure that sitting up is deadly so I will be avoiding that at all costs.
So, till I can get to LA someday to test like a grownup, I'm keeping the spacers. Even if my bike looks cooler without them.
0
Comments
I still haven't had a chance to lock down formal testing at the Velodrome and will need to do that in August, for sure. Problem is my bike will bouncing in and out of a bike box between IMWI and IMChat camps so not sure when that's going to happen.
I did a 100mi "focused" ride -- not IM pace but not a TT either -- and did 4:18, 227w Pnorm, 218w Pavg, .79 IF. That was with race wheel front, wheelcover, race tires and latex tubes, just ran with my setup from IMSG, and 13cm drop on the front end. Doing a repeat this weekend but will do 112 and have dropped the front end by 1cm, to 14cm drop.
My larger point is that I think it's valuable to consider moving around on the bike to improve your aerodynamics within a set front-end drop. I have my cockpit setup so that I'm a touch stretched out, I think. I then find myself chocking up or down on the bars to accommodate a range of riding situations, with my default being a "falling asleep in the bars" position that I go to when (1) have a clear view of the road in front of me and (2) I'm looking to ride at IM-effort (3) as slippery as possible. I bring my hands forward around the ends of the bars (Di2 with bar-end shifters) and slump my shoulders and drop my head such that my chin is almost on my BTA bottle. I seem to go crazy fasts in this position for the watts I spend.
I just couldn't get my brain to accept that the lower front end wasn't faster so I did research and found Chung and th golden cheetah software. The key to the Chung testing seems to be finding the right course. You need to b able to do multiple laps without braking and without cars. I found a spot and went out Sunday morning early and did the three trials - sitting up, pads high and pads low. Save each trial as its own garmin file. Then import to golden cheetah. Pull up one of the trials and under th drop down menu "view" (I think. I'm at work right now and can't access it). You need to input info on air density and on Crr to get an accurate cda figure. But all I was trying to determine was the delta of moving the pads so even if my crr was off or my air density was off it won't matter because they were held constant across all three tests. Sitting up was in the .31x area and both aero positions were in the .265x area.
Also as a double check, I looked at power and pace for the three trials via garmin. I rode them all super easy. 120w Pavg because I was worried about carrying too much speed at the turnarounds. Sitting up the 120w covered the 2.205 mi loop twice in 15:56. Both aero positions covered it in 14:40 exactly. Except the low position again had a higher he of 116bpm and the higher position was at 114bpm.
Now that I have a process down there are lots of things that I want to test. Body position within given bike fit like coach r says is at the top of the list. Clothing, bottle on the down tube. Glasses vs visor. Anything that I can change out in 5 min or so.
I'm not sure that I can trust getting all the variables held constant from one day to the next. Tire pressure, air density etc. my hunch is that these can move results from one day to the next. So testing my wheel cover would be tough.
If you try the software and have questions, let me know.
Did another test today and got a result so odd that I question the testing method:
I tested a regular cage and bottle on the down tube of my P5/6. I tested as carefully as I could and was as precise as I could be. Golden Cheetah testing indicated hat I had a lower CDA with the cage and bottle. Huh.
Naive testing showed both tests took almost identical time 4:52 with the bottle, 4:53 without. 61watts avg power with the bottle, 62 without. Still the naive testing did confirm what the computer model suggested. But I have to think that the whole exercise could have been impacted by a wind gust or something. Because we all know that the tooth fairy isn't real and that a bottle on the down tube slows you down.
right?
I think that next weekend i will do it again and double the distance of each trial.
Is it time to start training for IMFL yet????
After spending a bit of time reading the references, it occurs to me that a simpler method would be to compare coasting down a large hill a number of times keeping everything the same except for the thing you are testing. If you started the coast at the same point at the same speed, all you would need to do is compare the maximum speed (faster being more aero). I would think 5 repeats should do it.
Maybe I am missing something here?
BTW, as you can probably guess, I have a very long hill 10 mins away without any cross streets or lights that starts steep and shallows out at the end.
And here is the link to prove it. http://www.slowtwitch.com/mainheadings/techctr/waterbottles.html
I am in the process of getting a new TT bike, so I will do some tests in a few weeks and report back as well.
I was looking at tririg's 2014 Kona bike photos yesterday, and there were maybe one or two pros, tops, who didn't have a bottle on the down tube or seattube. I was really surprised. Although I think Cobb's findings are probably outdated - the tubeshapes of modern bikes are just so much more aero than the ti frame test bike - I'm started to be persuaded by how overwhelming this choice is. If only to have one more hydration source onboard.
Peter- a stopwatch would be the way to go. Takes out garmin error. You wouldn't have a calculated cda when you were done but that might not really matter at the end of the day.
The big takeaway for me so far is just how damaging sitting up is. It really drives home that I need a fit and fitness that will allow me to stay in the aero bars the whole time. Even at relatively slow speeds it made a difference in my tests. Specifically, riding 120w AP while sitting up I averaged 16.6 mph on my closed 4.4mi course. In aero, exact same AP and NP I was 18.1 mph. That is a killer.
Next test will be climbing a steep hill at 8 or 10mph. My guess is that when the hill turns up high enough and you are going slow enough sitting up won't matter as much and it will be easier to produce the needed power. Heart rate should tell the story there. I have a hill in mind and will give it a go Thursday morning.
I used a small traffic circle right outside my neighborhood. It is on an incline so I can testa range of speeds without having to touch the brakes. No traffic early on a sunday morning.
I did 3 5min tests each of 16 loops. The first was a "control" test. 3 bottles, me sitting up. Sitting up always tests poorly. Then I retested in aero with 3 bottles and lastly I took the rear cage off.
I imported the trials into golden cheetah. I left the air density and Crr in their default settings. Getting more precise here will help me get a better handle on what I actually Cda is but right now that is not of much concern. I am simply looking to determine of setting A is more or less aero than setting B. So as long as I hold the other variables constant I should be good.
Sitting up tested out as a Cda of .3147 which is more or less in line with other tests. In aero with bottles in all 3 positions tested as .249. much better. Stay in aero. The last trial was without the cage behind my seat. Cda tested as .22.
This is interesting to me. I want to test this again. My hypothesis is that I sit very far forward on my saddle and the post might be sticking the bottle out too far behind me and standing it up too much. I will try good old fashioned zip ties next time and see if I can tighten that gap. Because the difference between .249 and .22 is a meaningful one. and if I can keep 3 bottles and lower my Cda it will be worth the effort.
Next test will be to confirm the results of this test and also try a zip tie position on the rear cage to get it tighter to me.
I have nothing to add other than to let you know I am following along and find this pretty interesting. I too tried the scatter plot method and got nothing usefull. I've been looking for a local course to apply the Chung method with, but so far no luck. Keep us posted on your findings. Thanks for posting.
The course that I am using now is like a traffic circle roundabout on an incline. So I just start and stop on the top. Also, I try and ride the same line as closely as I can for each circle.
By way of example, here is my training peaks data for a couple of runs: The first is me sitting up in the pursuit bars:http://tpks.ws/whjq
The second is me in aero with the bottle behind my seat removed:http://tpks.ws/qX28
hopefully this will help. you can try importing them into golden cheetah and run the chung package on them. you might not get the exact same CDA #s that I did but the relationship between the two will be clear.