Wednesday, February 24, 2016

Power tamed

Recently I have been spending a lot of time on my Garmin Vector power meter. Here, and on the bike. More like experimenting on the bike and reporting it here. After some weird issues with total drop-outs which cleared up after a battery change the thing that was driving me crazy was getting reliable, balanced readings at the start of a ride. I am happy to report that I have solved the problem.

A word of caution. My advanced age provides me with a depth of understanding typically missing from younger folks. When I say "problem solved" it is with my fingers crossed behind my back. With me?

Bottom line up front: When you move the pedals between bikes, take the receiving bike out for a calibration spin immediately after installation. Do not install the pedals in the evening, put the bike in the car the next morning and drive to the ride start, then attempt the calibration ride. The results will be awful. The fix in that situation is to remove power, wait a bit, then power up as if you just installed the pedals and then do the calibration ride.

The reason this is important to me is that it answers the question of what to do before a race. Get the bike fully calibrated, rack it, and it will be fine coming out of T1.

In order to do a good calibration the head unit should be configured to display left-right power. I do not have this on my main page as it takes up too much space, but I do have it on a separate page.

In the photo the balance data is at the top of the display and reads 87%-13%. I took this during a calibration spin up on a trainer and it shows how whacky the system gets when the pedals are moved and not immediately calibrated. In this case cycling power fixed the problem.

The calibration ride goes like this:

1. Select the correct bike.  The most significant impact of this setting is crank length. On the Edge 800 the sequence is

Menu > wrench > Bike Settings > Bike Profiles > (name of bike)

2. Ride for 1-2 minutes with a cadence of 70 or higher. Best done on a trainer, or on a long, flat road where you can ride without slowing down, turning, etc. The important thing is to maintain a continuous spin with smooth, even pressure on the pedals. It takes a few seconds with cadence above 70 for the display to kick in, and at first the balance numbers may jump around a bit. This can be done on a slight downhill grade as long as a brake drag is added -- spinning with too light a force will result in an incorrect calibration.

3. Stop, unclip from both pedals, rotate the cranks until they are parallel to the ground. Do not move the bike or touch the pedals or cranks. Go to the head unit Power menu and select Calibrate. On the Edge 800 the sequence is

Menu > wrench > Bike Settings > Bike Profiles > (name of bike)  > ANT+ Power > Calibrate

Wait until the head unit says the calibration is complete.

That's it.

The next time you go for a ride on the same bike the calibration should be good enough for anything short of a detailed pedaling drill, providing power was not interrupted. By that I mean the pedal pod connectors were not unplugged and the battery doors were not opened. If you want really precise balance data, repeat steps 2 and 3 before the ride. The good news is that these steps do not need to be performed coming out of T1 as long as they are done just before the bike is racked.

Wednesday, February 17, 2016

Butt and power update

In my last post I shared my revelation concerning Dorian's "push on the handlebars" advice, a solution to my issue with saddle position. To recap, I experimented with tilting my saddle down in front to relieve pressure on sensitive places. The result was a much more comfortable ride, but I kept sliding down to the tip of the saddle. My hunch was that this had more to do with limited hip mobility than gravity, and my revelation was that just a little push on the aero bars would offset this action.

On Friday I was pushing the bike through the parking lot, hand on the saddle, and the angle felt wrong -- too much tilt. Funny how what seems normal one day is anything but the next. Since then I have been fiddling with the tilt, hunting for that sweet spot between a stable platform for riding several hours at a time, and applying minimal pressure to sensitive places. Fore-aft position plays a role and I have not even gone there yet.

I also have an update on power. This week's Tantalus Tuesday gave me a chance to experiment with moving the pedals between bikes. Today's spin class should provide an additional data point. On Monday night I moved to pedals to the Merlin. I connected the pods but did not go for a ride. Tuesday morning I was careful not to start spinning until I was on relatively flat ground, then spun up to 70+, as smooth as possible. The L-R balance was around 30-70. Not good. So, I thought that this supported my theory that the first spin has to be done soon after plugging in the pods. I unplugged the pods and shut down the head until, waited a couple minutes, then brought everything back up. Still 30-70. Then I remembered that I had to set the head unit to the other bike; it was still set to Makani Kekoa, which has shorter cranks. After I switched to the Merlin the Edge 800 automatically re-scanned for the power meter. I was at the right place in the menu tree so I did a calibration. I did a brief test ride and balance was 50-50. Overall balance for the ride was 48.6/51.4%, which is as good as it gets. Today I will go through the same drill without unplugging the pods and see what I get. My goal is to have reliable results in a race after the bike has been sitting for hours. I really should not have to spin up, stop, dismount, do a calibration, then go again. That would be crazy.

This Saturday I will be attending a triathlon cycling clinic in Kailua hosted by Ben Williams and led by coach Clayton Hamilton. Should be a lot of fun. I plan on asking a lot of questions about power meters in general and Garmin Vector in particular.

Update #1: The clinic date has changed, now scheduled for March 19.

Update #2: My experiment for yesterday's spin class involved setting the crank angles -- the initial spin at 70+ RPM -- after installing and powering the pedals the night before. The result: better understanding but more tests required. After spinning up with the bike on the trainer I had 100-0%. Totally bad. Seeing that it was not improving and knowing I only had a few minutes to get ready I opted to start over. I pulled the plugs and shut down the Edge 800 head unit, waited a couple of minutes (still not sure how long it takes to fully reset the system), then plugged in the pods and fired up the head unit. The spin-up began the same as earlier, 100-0, but this time I kept at it a little longer and the balance gradually came to around 30-70. I stopped spinning, got off and did a calibration. Gt back on and as I peddled the balance gradually came to 50-50. Based on this and previous results I suspect that the calibration does more than zero out the residual imbalance, which is what the documentation says. I think selecting calibration puts the head unit into a balance period, during which it hunts for 50-50.

What remains? First, I need to do the exact same sequence minus the electrical power cycle. Second, I need to know how to get proper results on race day, without having to do the calibration. I think it is time to put the question to Garmin.

Wednesday, February 10, 2016

Sore butt

I knew that arriving at a comfortable riding position on Makani Kekoa - my new Cervelo P3 time trial bike - would take time, not to mention a great deal of fiddling. I may feel like a kid inside, but my body is well along the path to old age. Flexibility was never my strength, and these days what little I had is rapidly vanishing.

Before last Sunday's long and windy ride I made some changes to my rear hydration setup that required moving the saddle. I planned on using the ride to tweak the saddle back to optimal position. Sure enough, it wasn't long before I felt like my torso was resting on my scrotum. A gradually rising burn foretold days of discomfort unless something was done.

I pulled off at Wailupe to make the adjustment. Too much. Now my ride felt like sliding down a grassy knoll on a piece of cardboard. Another adjustment split the difference and "Ah, relief!"

Back on the highway I began to notice something interesting. Every few minutes my butt would end up too far forward. Stop peddling, stand up just a bit and slide back. Good. Wait. No. Back again.

I developed a theory. It was not gravity causing my butt to slide forward, at least not entirely. The culprit was my tight hips and hamstrings. Every time my foot came up and over the top of the stroke my too tight thighs yanked my hip forward. Not a lot. Just enough to cause my butt to creep forward. Had my scrotum not been as sensitized I might have missed this.

Tuesday afternoon I rode Makani Kekoa again, an easy ride that let me concentrate on this slippage issue. In the midst of my ride I heard Dorian's voice in my head, reminding me to push on the handlebars. Sure enough, that was what I needed. This will be my number one focal point for my upcoming rides, along with ongoing efforts to open my hips and hamstrings.

Tuesday, February 9, 2016

Power struggle much improved

In my last post I shared the problems I had encountered with my Garmin Vector pedals. Since then I have done a lot of testing and fiddling, and have learned a lot. The significant points are
  • The pods are a snug fit, so tighten pedals carefully
  • Be extra precise reading torque wrench
  • Equal torque is just as important as tightness
  • Always perform spin-up and calibration before ride
  • Do spin-up right after plugging in pods
In the last two weeks these procedures have yielded 100% satisfaction.

While writing this post I came across this item on the Garmin web site:
What best practices are recommended for achieving consistent, accurate and reliable results from my Vector or Vector S pedals?

My life would have been a lot easier had I known about this page sooner. Oh well. Here is a wrap-up of what I learned on my own:

I did have a burr on one pedal pod and was successful in smoothing it down, but even so both pods are a snug fit on the pedal shafts.  This means that as I tighten the pedals I should expect a lot of drag on the pods -- they will try to rotate with the shaft -- and this will continue for more than a full rotation until the shaft shoulder finally makes contact, at which point the tightening process feels normal. To my experienced mechanic's hands this feels like something is wrong, but it is not. At least that is what I believe to be the case. For now.

I wanted to buy a click-type torque wrench but the only one in stock in Honolulu only works on right-handed threads. Apparently this is not that unusual, as even some models of Snap-On wrenches are left-hand challenged. Parks tools makes one, and I really ought to get one. Anyway, I have my simple, inexpensive Craftsman beam type wrench that does read left and right, but I have to be very careful to position it so that the crow foot does not rub against the pod and I can clearly see the indicator. A nice, smooth pull to 25, spot-on, and that is it. Exactly the same both sides.

The jury is still out on this power-up procedure but here is what I have observed. Say I have a ride with an early start time, and the pedals need to be moved to the bike I want to use, so I move them the night before. Next day I am likely to get erratic power readings. Typically starting at 60-40 balance, then deteriorating to 100-0 (the high-low relation can be either way), then no power at all. Solution: reboot the system. Meaning get off the bike, unplug the pedal pod connectors, power down the head unit, count to ten (slowly), plug in the pods and power up the head unit, then get on and spin for awhile. Ta Da, much better.

So, I do not plug in the pods after a pedal move (after they have been unplugged) until I can proceed with the spin-up drill at the same time. Leave them unplugged until I arrive on location, set up the bike, plug in the pods, go for a short ride to complete the spin-up process. This has been working very well.

In the early edition of the Garmin software the spin-up procedure required pedaling backwards. Somewhere along the way Garmin tossed the back spin method, so I assume that there is a window of opportunity right after a pod plug-in event where normal pedaling is used to set the "installation angles." If this is not done properly the left-right balance will be off, and perhaps even the overall power output. My typical Tuesday morning Tantalus ride presents a challenge here, because the gathering place is on a hill. This morning I started "backwards," a downhill run with brakes on and pedals spinning easily above 70RPM until the power reading appearing on my Edge 800 head unit. I turned around and went uphill, and saw that I had about 45-55 balance. Close, but it could be better. I stopped and did a static calibration -- feet off the pedals, cranks parallel to the ground -- and after that I got 50-50. Garmin recommends doing all that calibration on a trainer; I wish I could afford a house big enough allow me to set up my indoor trainer indoors. You do what you have to do.

The numbers from this morning's ride look good. NP was 141 watts, just over my current FTP of 130, average power 117 and max of 554. I was by no means doing this all-out, only approaching that on the steep sections and shooting for zone 3-4 on the easier parts. Respiration matched these numbers perfectly, so I am confident my FTP value is good -- I was way off a few weeks ago, but that is typical of a new power meter user. All this fiddling with power numbers can throw off Training Peaks' assessment of my workout. For Tantalus I usually get a TSS of 80 but today I got 131, and that was not all-out. Let's wait and see what a few more weeks of data reveals.

This is curious: Garmin Connect shows power balance for this ride as 48% L / 52% R. Training Peaks gives Pwr. Bal. 52.6/47.4%. So which is correct?