Head2Head Sprints

It's Lives!

2011-12-04T21:46:00.000-05:00

It's been some time since I've messed with this project. I've been working on bikes a bit in preparation of selling my SS and setting up my other MTB for a stint on the trainer this winter. Apparently, it's been enough to peek my kid's interest and refresh their memory on this project and a request was made to revive the home grown roller racing system to pass some time in the winter months.

So, I spent a quite a few hours, digging around in boxes, our old laptop and my external hard drive and made a run at getting things rolling.

I set up in the basement first thinking things would be easy and it'd be ideal to have a bike near for some test rides. Unfortunately, nothing ever seems to be easy around our house. Or orderly.

Test rider, anxiously awaiting first go.

Unfortunately, in typical form, I never wrote down any notes on how to hook up the sensors. Which quickly overheated and self destructed. Work moved to the kitchen table where after a full installation of software, registration of DLL's, near complete read of the Lab Jack manual and pillage and re-purposing of two bike computer harnesses I had lying around, things appear to be ready for business.

I am curious how accurate things will be with use of legitimate bicycle computer sensors versus the ones I ordered to specifically meet the specs of the device.

More fun to come.

No progress but, fun planned

2010-01-07T09:19:00.001-05:00

Well, what can I say. I've done nothing.

On the bright side, the event that inspired this project last year is just around the corner. Maybe it will be the motivation I need to finish things off and get some folks together to try it out.

Hell Yes Sprints at the Blind Pig. 9pm, January 19th. Flyer says 7pm. However, doors don't open until 9pm.

Click Here To See The Flyer & Details

Click Here For The Official Website

***This image will soon expire

Waking the Dead

2009-09-22T12:38:00.003-04:00

Spring and Summer came and went, my weekends and free time became devoted to soccer games, family time, yard work and mtb racing. While every weekend until the third week of November is booked on my calendar, I can see some light peeking over the horizon to put a little time back into this project. Work has slowed down for the moment and my thoughts have turned towards something to keep me sain when the air turns ice cold and the options for outdoor activities quickly become shoveling snow, sleding and an occasional snowball fight this winter.

Really not too much left to do. Just need one forkstand and a way to mount them and a couple of sets of rollers to some plywood.

One more important improvement may be to the bikes. Right now they are both running ghetto 52 X 16 gearing. This may prove great for roller racing but, I used one of the bikes quite often this summer to pull my daughter on her tag a long and a 44 or 48 chain ring would be a definite improvement. I've got a spare mtb crank and BB. So, I am thinking of throwing that on one bike and finding a huffy donor somewhere for the other. That should improve the gearing as well as, the ghetto chain lines.

Parts Are In

2009-04-10T09:57:00.003-04:00

Parts are in. Bikes will hopefully be together by Saturday afternoon. I put the bottom bracket and fork back on the white bike last weekend. It turned out pretty sweet. The chrome forks look good. The rear wheel will need to be trued. It's a wobbler.

Getting closer.

Jim, if you are reading this, time to build a forkstand. :)

Transformers

2009-03-29T11:49:00.006-04:00

So, I picked up a couple of vintage schwinns last weekend. A very lightly used, red, approx. 59cm, Sprint and a smaller, yellow, rusty but, still revivable, Le Tour. I got a little excited and forgot to take some before pictures before I started in on their transformation.

Here is the Sprint. Minus gears, cleaned and bearings repacked. I left the brakes on. I am considering using this one around the neighborhood. It was a pretty unmolested bike. Except the threads on one of the crank arms were pretty messed up. I suspect an attempt to do some maintenance with improper tools/technique. It's got it's share of scratches but, it says sprint on it. So, I couldn't justify painting it.

The Le Tour in its new color. Turned out pretty good for a rushed, rattle can paint job. It would have been really nice if I would have been more diligent on the sanding and painted with the garage door closed. But, the kids were helping so, that's my excuse. The fork turned out to be nearly all chrome. I think it's going to look pretty good all together.

This is a before pic of the Le Tour I receive via email before I went out to pick it up. Looks pretty pristine, huh? This one was proof to the fact that pictures are deceiving. Think that it lived outdoors and the tarp behind it was what was covering it up. Steal wool does amazing things with slightly pitted chrome.

Just need to order up some track gears, tires, pedals and maybe some longer sear posts, reassemble and we are in business.

Vintage Delights

2009-03-20T15:17:00.003-04:00

I am on the hunt for a couple of vintage schwinns to convert to fixies and serve as the first generation of roller racing bicycles.

With the weather improving, these nearly all steel gems are popping up here and there on craigslist. Some for reasonable prices.

Conversion should be easy enough. Given they will come apart.
- Pedal adapters to accept 9/16 pedals
- BMX sproket
- Ditch the freewheel for a track hub
- Toss all other non essential parts. Shifters, derailleurs, brakes, reflectors, etc.
- Top it all off with a sweet flat black paint job. Maybe some pinstripping. Or maybe red.

I bet they'll still weigh 30 lbs, in their new, nearly naked, state.

I'll post some pics if I end up aquiring any.

Lil' Bling

2009-03-18T12:33:00.006-04:00

The Opensprints has the great idea to put some LEDs on the wheels of the bikes for an event. There are all sorts of ways to do this. Simple, one light valve stem lights to full out, DIY, programmable setups. I am thinking about something in the middle of the road like the following. Some optional patterns and words but, no need to build or program anything. I've got enough on my plate at the moment. Cost seems pretty reasonable. ~$12. The question is whether they will fly off at high speeds or interfere with the senor pickups on the wheels.

I may just have to stick to good ol' pyrotechnics. ;)

Roller Twins Unite!

2009-03-17T13:45:00.004-04:00

The wife has reported the arrival of my second set of rollers. There was some initial concern that due to the size of the box, it may be another bicycle. I was quick to reassure her that no bicycles were being added to the stable.

At least not at this time. :)

Getting closer by the day.

I've been thinking that a whistle and megaphone may be useful tools at a roller racing event.

Ready for Performance Testing

2009-03-14T17:27:00.004-04:00

Spent a few hours this morning and built out the official beta interface. Figured I might as well get it done since I still had the two bikes hooked up and wouldn't get the chance to do any two up testing during the week.

With my son and daughter providing some hand cranking for pedal power, we got what is called "Acceptance Testing" in the IT world done. Glad I went ahead and did this. Found quite a bit of bugs that needed to be worked out. In the end, the only complaint I have is the progress bar controls have some kind of delay that causes them to reach their maxium after the set up has reached the end of the race. It's not too bad and I may be able to tune it out later.

So, things are ready for some performance testing. I plan on ordering up another set of rollers this weekend. Hopefully, Jim from BIT will get a chance to put together another homebrew fork stand soon. Once that is done, it's time to get a group of people together and beat up on this thing and see how things work out.

Close Enough

2009-03-13T21:54:00.004-04:00

My buddy Lee stopped over and tonight offered up his pedal power services for the cost of a few beers. So, two up testing happened tonight. I am going to chalk this one up as a success. With two up on the setup, the distance matched our bike computers to the 100th. Same behavior as the one up setup. I'd say things are pretty much spot on. I had thought I corrected my math error on the average mph but, unfortunately I hadn't.

After I dropped Lee off, I determined the reason for the difference in the setups calculations of average mph to the bike computers. I've got it damn near spot on now. At least with my bike computer. I have to keep the sample/display rate of the average mph at 3 seconds to get enough time to make up for the variance in time I am using to display the calculation but, I'll take it. I may try 2 seconds tomorrow. 1 second didn't work out as well as I had hoped.

So, looks like it's going to work out. Plan is to get the full interface knocked out by next weekend. One more test and tune and I think we'll be in business.

Who's up for a little roller racing?

Rain delay

2009-03-09T08:49:00.004-04:00

Things are all but, ready to go for running some two up testing. 5 minutes of soldering and hopefully, just 15 minutes of testing to ensure the DAQ & PC are picking up both sensors.

Had every intention to do some testing yesterday but, Michigan weather decided it would drop several inches of rain on the the already super-saturated SE Michigan landscape. So, I spent the remainder of the day attempting to prevent water from entering my basement windows and soaking up the water that did so, we didn't end up with a bunch of wet carpeting to deal on top of it all. The backyards around me are basically a cesspool of standing water and mud. You can't walk across the front or back lawns without sinking in and leaving footprints. Ugh. I need to start playing the lotto.

Looking positive

2009-03-07T19:53:00.006-05:00

Things are looking positive. I wrote the quick test program that I had been dreaming up this week. Then, I soldered up a sensor to a 20ft ethernet cord and taped it to my bike. Jim from BIT was kind enough to lend me a home brew fork stand. I got in a minimal amount of testing before the kids got in bed and had to quit.

After moving the magnets so, the magnet for my bike computer and the magnet for the roller racing set up were on the same spoke. Results started to turn out pretty positive. When converted to miles, the distance in meters output by my program matches the output of the bike computer to the 100th of a mile. Not bad. Pretty exciting actually. If that keeps up and the computer/DAQ/sensors can stay that accurate with a two bike set up. Everything should be golden.

I was a little peeved that my mph calculation only seemed roughly accurate. However, I now realize I have a math error in the conversion from kph to mph. Hopefully, that will correct things and bring the measurement a little closer.

The following is a screenshot of the test program and a couple of shots of the set up on my still messy from a ride in the 70 degree weather yesterday, 1989, vintage Cannondale. Man, you should hear the sound the set up is making right now. The hop in my rear wheel vibrates the rollers on the wood that I have put down so, the roller base doesn't punch through my sweat catching mat. The hollow base of the home brew fork stand resonates the vibration. At 60mph, you'd think I was running a chainsaw in the basement.

I bet my wife appreciates me publicizing our messy basement. Sorry babe. :)

More testing to take place tomorrow.

Need to get a move on

2009-03-05T08:13:00.002-05:00

The Project Manager in me is feeling the need for some more progress. Nearly a week of March is gone. I've done a little more thinking in my 37mile commute to and from work. Really need to focus on determining if the sensor/DAQ is going to work out. I am hoping I can leverage the magnet from my bike computer for testing. 50% chance it'll work. If not, I am going to have rig up a way to mount the magnets I have to the spokes for testing. Which means more time before I'll know anything.

Going to shoot for another round of getting code to pick up signals from the DAQ tonight. Fingers crossed, I'll get that going.

A little progress

2009-03-03T09:44:00.004-05:00

Figured out a better stopwatch & did some prototyping with it. Also wrote some pseudo code for how I expect the interface to be built with the new stopwatch and how I believe I'll need to write the to interface with the DAQ.

Next step is to write up a simply interface to monitor the DAQ and actually hook the sensor up to the bike. That way, I'll have a better idea if this will work and an easy way to determine whether senors are live and functioning.

The botched Iceman registration took up most of my free time tonight. What a mess.

An education in the DAQ

2009-03-02T08:25:00.002-05:00

So, I spent a few hours becoming familiar with the DAQ last night. As expected, getting things going will be a bit of a struggle. I struggled for 3 hours to get a single signal to be registered in the device. The great thing is the device comes with a piece of software specifically for such experimentation. Another great thing is that I accomplished getting the device to register the signal using one of the sensors I purchased. I learned also sorts of interesting and very frustrating "features" of the device. Most importantly, I didn't smash it with a hammer out of frustration.

I originally attempted to update some sample code to accomplish the method I was expecting to use to count pulses and track elapsed time. Boy, what a miserable failure that was. Of course, I had to pick the most robust, complicated method to attempt this. As my luck always has it, the device does not behave at all as I expected it would. So, my example results incremented wildly without any rhyme or reason.

I gave testing a rest at 11pm since I had to be up at 5am for a session of simulated hill climbing and sprinting in spin class.

I've formulate my new plan of attack. This will provide at minimum 1/100th of a second accuracy in the timing. Not the accuracy I was hoping for but, it'll do. Then, it'll just be a matter of whether the sensors will pick up each wheel rotation, bounce and cause double counts and/or the length of wire intend to use will be too long.

Haven't failed yet. Hopefully, this will be the roughest patch.

Daddy's got a logo

2009-02-26T19:51:00.002-05:00

Check out the official beta release logo at the top of the blog.

The Mind Grinder

2009-02-26T08:13:00.006-05:00

The DAQ arrived yesterday. I spent about an hour and half in some seriously deep thought with my nose buried in my laptop reading through the DAQ manual, forum posts and example code. The goal being to finalize my strategy on how I'll use the device to accurately record & time distance traveled, display progress and with reasonable accuracy, determine the winner and riders' true elapsed times.

Seems like these requirements would be so simple to meet. However, they are not. Well, not for a semi perfectionist, bent on having this set up be more than just remotely accurate. The DAQ is simply an interface with the real world, it won't think for the computer. The main issue is that its output is not event based. Meaning it doesn't alert the computer when something is going on or happened. So, the computer has to monitor the DAQ and it's inputs and perform its own logic on the results. Which by the way may be erased if they aren't monitored quickly enough. So, everything will need to be accomplished in a loop. Yeah, I know, still seems simple but, there are limitations, unruly behaviors and other concerns with the DAQ and sensors that need to be accommodated.

Sensor position has come under fire again. The crank position may not provide accuracy in timing I desire. I think I am going to have to go with a wheel mounted magnet to improve the accuracy of the timing but, not require the setup to deal with as high of a frequency as a roller mounted magnet. Which means some potential magnet modification. More work.

The rest of this problem going to have to be solved in code and and ample amounts of testing. The plan is now to build out a couple of methods to accomplish the requirements in code. Then bribe some friends to take turns putting the set up through its paces for a full blown test. Hoping to do a little more reading this evening and possibly some coding tomorrow.

Tentative Test and Tune day is May 14th. Anyone up for a little sprinting on rollers. I'll probably only need you to sprint for 20 seconds 50 or 60 times. :)

Maybe I'll just work on a logo for the blog and GUI. That's easy enough and would not require as much thought.

The Cutter

2009-02-25T15:29:00.007-05:00

How inappropriate. :)

If things take off, these bikes have a high probability of being the steeds mounted by competition contestants. Sans brakes and set up with a fixie hub.

Check it out!

I love the way this thing looks. Love, Love, Love....

GUI Sneak Peak

2009-02-24T18:52:00.003-05:00

So here's a sneak peak at the prototype display. I haven't added avg cadence to it yet. The graphic in the center is just a link to a graphic on the computer so, it can be quickly changed. At the moment, it's set to my favorite LBS since I'll be leaning the owner Jim Colflesh to help with putting together the bikes, testing facilities and initial debut location. It's nothing too fancy or pretty at the moment but, it gives you a flavor for what's possible.

Progress bars will track riders simulated race.

Meters will display meters traveled. Updates very quickly.

Stop watch for time. Obviously stops as each rider finishes the configured distance.

Average MPH gets updated every second.

Average Cadence will be added and update with Average MPH.

Still lots to be done but, it'll due for the Beta testing and release.

The second experience

2009-02-24T08:07:00.004-05:00

The second AA Hell Yes Sprints event was held last evening. Unfortunately, attendance was definitely down from the last event. Mostly likely, due to the fact the event was held on a Monday night again and just two weeks from the last. The HYS crew has definitely improved their coordination of the event. Things ran pretty smooth minus a mechanical difficulty at the start.

Given the size of the crowd, I was on the fence whether I'd be strapping myself in and pedalling like a manic but, folded quickly when asked by one of the crew if I was going to participate. I did improve my time by about a 1/2 second. Guess running twice the day prior and attending spin class in the morning before the last event didn't help me out much.

Once the races started up, the crowd got lively and the night quickly turned into a success. Definitely, glad I threw my hat in the ring.

I definitely picked up a few pointers/ideas last evening.
1) All rollers and for stands should be bolted, screwed or somehow adhered to pieces of plywood on the floor. Hopefully, doing this will no mean no need to steady the bikes.
2) Pedal changes really slow the tempo of the races. Some dual sided (toe clips & SPD) pedals like spin bikes have would offer an option for folks that wish to wear cycling shoes and folks that show up in street clothes. Idea compliments of my friend Lee.
3) Riders need to have a view of the monitor. Without a good spotter, it's tough to guage where you are and how your doing during the race. After about 10 seconds in, you really have no idea what's going on, how much longer you have, if you are winning/losing, by how much, etc.
4) Event supervisors/mechanics are a must. Not only to quickly set seat height and help people strap into pedals but, being an owner of a set of rollers, I had forgotten how foreign it is to mount and dismount a bicycle propped 6 addition inches in the air. Lots of toppling over, riding off of the rollers, etc.

My sensors arrived yesterday. They came in fancy static resistant bags with warnings all over them about how static electricity would destroy them. Never been shocked getting on or off my bike but, I am guessing rubbing a rubber tire on a roller at 60mph creates a lot of static. Another item to worry about. Hooray!

A broken bicycle computer was donated to the cause by my buddy Jim. I performed a little exploritory surgery on it with a knife, screw driver and some channel locks. As suspected, the sensor is a simple reed switch. This leaves another solution if the sensors I purchased don't work out. A couple of kits that extend sensors to the rear wheel of bikes may work wonderfully.

I've also done some googling over the past few days. Seems one of the Goldsprints events has had the same issue with roller speed exceeding the capacity of the equipment that I was concerned about. The solution was simple. Put the sensor on the crank arms. While I had considered this previously, I had dismissed it due to the fact that I'd have to zip tie a wires to the bikes (would look ugly & make interchanging bikes impracticle) and it would potentially lead to too much inaccuracy in the distance measurements (approx 7 meters). However, I've since reconsidered and now like this idea. With this strategy, the sensors would only need to deal with 3Hz vs. 130Hz, no chance of 100+mph magnet projectiles, it would make displaying cadence a snap and it would leave a little strategy to the start. Technically, you could get almost a rev before the sensor signals false start and have some momentum built up on go or position the crank just in behind the sensor and make the distance as accurate as possible. Small details.

DAQ comes tomorrow. Can't wait. Last night really got me fired up.

Software prototype built & Hardware on order.

2009-02-22T10:00:00.006-05:00

What's the skinny on where I am at today?

Tentative Project timeline.
Feb. - Acquire Parts.
March - Build & Test.
April - Debut & Party.

The prototype interface is built. Just a few hours invested. All features were relatively easy to accomplish. Gotta love that VB.

I've ordered a windows compatible DAQ (Data Acquisition Device) to power the sensors, count the pulses received from the sensors and interface with my PC. It has two internal counters and the manufacturer claims it has a sample rate that will easily handle the 130Hz/130 rotations a second I've calculated that it will need to handle to keep up with a 15 second 500m sprint on my rollers. After trading some emails, I've also been told by the tech support for the product that the device includes an internal pull up resistor and a capacitor would likely not be required in this application. So, with luck, no soldering circuit boards. Hell Yeah!

Sensors & magnets ordered. http://www.digikey.com/ seems to be the be all end all when it comes to this kind of stuff. After ample amounts of research I determined I needed Hall Effect Unipolar Sensors/Switches. These have an internal switch that remain open until the southern pole of a magnet passes it. At that point, the switch closes and voltage is allowed to pass. They are wired so, I should be able to wire them right into the DAQ. No sweet LEDs like the opensprints set up. :(

Concerns to date...
1) Will my PC running VB.net be fast enough to poll the count of the DAQ often enough to provide for at least 100th of a second accuracy?
2) How hard will the DAQ be to wire up and get interfaced with the GUI I built?
3) Did I order the right sensors? When the rollers are traveling at 130Hz, will the magnet be passing by them so often they won't have a chance to open and send a pulse?
4) Will the magnets be to powerful, too weak, to large, too small?

I should have parts by mid week. Going out of town this weekend. So it'll be some time before I know.

Narrowing down what to buy for sensors and to interface with them was way more work than I thought it would be. I thought I had the problem solved many a times, only to do a little more reading before I ordered and find things weren't going to work out.

My guess on the result of this little experiment at the moment is that I'll struggle through getting things programmed & wired up but, in the end find that a limitation in the DAQ or software will prevent accurate/often reads or counts can be processed for a good racing environment. Moving the senors to the wheels of the bikes may be the potential solution. We shall see.

The idea is born

2009-02-22T09:37:00.008-05:00

Well, it wasn't born as much it was discovered. Call me a copycat. That's fine. You won't and probably can't hurt my feelings.

I attended Hell Yes Sprints in Ann Arbor, Michigan on Feb. 9th, 2008.

http://hellyessprints.blogspot.com/

It was an absolute riot. So, I woke up the next day an couldn't shake the thought. I've got rollers, a bike and some computer and programing knowledge. How hard could it be to get a set up going for myself?

The google search was on. Seems a proper roller racing setup is not really commercially available. There is http://opensprints.org, which seems to be the birth place of the Hell Yes Sprints set up. After some thorough reading, I decided, the setup wasn't for me. No doubt, it's a decent setup and possibly the best out there. I just don't like the idea of soldering my own circuit boards, the software being open source and having Linux be the operating system the "open source" software is the most friendly with.

As usually, I decided to cut my own path into this endeavor.

By the weekend following the event in AA, I had not shaken the bug, at the moment had no "project going" and had been in the hunt for sometime. So, I decided that I'd dust off the old programming skills and put together a prototype user interface. I decided that I'd use VB.Net since everything I own runs on windows and with my experience in VB 6.0, I should be able to whip something up with relative ease. A few hours and diet Pepsi's later, Head2Head Beta was born.

It's based on a few assumptions/requirements.
1) Head2Head means two riders. I envision them facing one another as they battle. This will provide a ripe environment for stare down intimidation, trash talking and taunting. All necessary components of a fun filled competitive event.
2) Configurable distances. While I understand 500m seems to be the "standard" distance, I think more seasoned cyclists would enjoy a 1000m or 1500m battle.
3) A fun filled display. Besides a progress bar of some kind to show the riders making progress to their simulated destination, I am thinking a large stop watch, distance traveled and average kph/mph would make things a little more fun.
4) Countdown. A simple countdown from 5 to start things off. Later, I would like to enhance this to signal false starts.
5) Magnetic Pickups/Sensors. Seems to make the most sense for this application. Works well for bike computers. Why wouldn't it here?
6) Configurable roller sizes. The thought right now is to count the rotations of the rollers to determine distance travelled. Since I only own one set of rollers and will likely be borrowing someone else's if this thing every becomes reality, this will be necessary.

So how hard can this be?