Who is Timing Bike Races?

[mstompro]

I would really like to get setup to time bike races, specially mountain bike races. At this point I think my idea is to invest in a truss system and mount the antennas overhead and get dog one wet tags for the number plates. Anyone doing bike races can help me out with experiences doing bike races?

[timgrummell]

If you put the tag on the helmet instead it would be closer to the reader.

[Kevin]

It sounds like you are going the right direction for MTB. I do a lot of mountain bike races and that's pretty much my setup. Put the tags on the lower part of the number plate because some riders mount the top tight against the handlebar which can cause interference.

Have you raced, timed or officiated mountain bike races previously? I ask because the trick to scoring isn't so much the equipment but having everything setup properly in the computer and your overall race day plan. I'm scoring a mountain bike race this weekend that will have somewhere around 400 participants but we will have to utilize 3 databases with about 20 races in each database.

[tiberiu.gindu]

I did a couple MTB races. Used tags on the race number, sticked lower and aside, to avoid contact with metal (fork, pipe..).

Not achieved full 100% read rates mainly due to athlets rolling their race number on metal bars or other such "brilliant" ideas : )

Antennas were 45degree to the finish gate ( 4 pieces in the corners of the chute).

* * * * *

On the other side, some guys i saw are using hard plastic tags, fixed with zip-ties under the bike - facing mat antennas. This setup seemed to work just fine - the guys were not doing manual bkp and were even leaving the system unattended sometimes.

[Todd Gould]

I time mostly bike races.  I have timed Road races, Criteriums, Time Trials and a few XC races and 5K's.  I experimented with read rates, distances and speeds.  I have had 40+ finishers cross the finish line is less than 8 seconds at 25 to 30 miles per hour and gotten 100% read rates once I finally settled on a solution and specific inlay.  The best inlay from my testing is the dog bone.  I was able to get a few more feet in read distance with that tag.  However, due to the difficulty in getting only 1000 of these inlays I settled on the alien that Brian uses most of the time.  I have purchased 2000 inlays directly from Brian.  He just makes it so easy. 

I use 2 tags per racer.  One on the helmet and one on the front fork.  I read with 8 antennas.  Four overhead and four on the side of the road.  Both tags are coded with the same number.  This is the best solution to catch every 20+ MPH rider finishing in a large pack.  Obviously manual timing is not conducive to large pack finishes.  So the backup we use is 2 GoPro HD Hero3+ cameras at 1080p/60 fps.  This can establish order if there is a discrepancy in the results.

Here is a GoPro video http://vimeo.com/69127702 of my finish line in 2013.  I got 100% read when compared to the video.  However I had to make a few corrections on order due to a few reads being so close and the MS Windows .net code not allowing millisecond resolution. 

My reusable bike timing chip is my own design and allows the chip to be mounted such that the carbon fork does not interfere with the signal. I use the same tags on the helmet.  I currently have 800 chips I have made to time 400 racers.  Let me know if you are interested and I can go deeper into this subject.  I also have about 800 shoe tags made up using Brian's instructions.

We time 4 to 5 concurrent races in the morning and 4 to 5 in the afternoon.  The races start 5 minutes apart and some races contain multiple categories.  You have to setup a separate race for each .  I can help anyone through this process if needed.   It took multiple discussions with Brian to finally come up with the best way to handle USAC cycling events.   Brian was instrumental in figuring out how to make this work for bike races.  I can not thank him enough for the extra effort he put in over the last 2+ years helping me get it right.  Pre Race setup of your databases is a must.  With  bike races you have people registering for the afternoon races while the morning races are going on.  So good DB swapping is imperative.

You are specifically asking about mountain bike races.   My suggestion is a short truss that is 10 to 20 feet wide. (ours is 30 feet wide to cross large roads).   Put 2+ antennas overhead and 1+ on each side.  See the attached finish line photo for a reference.  Mount one on the fork and one on the helmet. Doing this should give you almost 100% read rates.  With mountain biking you can just go with a single tag on the helmet and overhead antennas and you will still get near 100% accuracy.  Mountain bike finishes have no pack and are usually pretty slow.  You can really manually time a mountain bike race.  :)

My Bike Chip Return bin...

[hubbard47630]

Todd,

I can't see/find your pictures of the finish line set up. It sounds identical to ours. We several trusses ranging from a fixed 10' all the way to a modular one that can reach 40'. For Crit racing we use a high speed camera system but for mtb racing we use our chip system. Did you design your chip holders or where did you get them made?

Thanks
Chris

[Ohio Race Day]

@hubbard,

Here is his finish line set up ..

http://vimeo.com/69127702

[hubbard47630]

I thought he had some up close pictures of the set-up. We're changing our overhead antenna mounts and wanted to see what others were using. At first I wanted to use a clamp attached to a quick release so I could easily switch from one of our truss finish lines to using the tripod mounts. Truth is, we are really using the tripods now so may end up removing the detachable quick release heads.

[Todd Gould]

The Truss we have is an expensive lighting stage truss rated for a huge amount of weight.  Therefore it uses tubes that are really overkill for the 4 light antennas we place upon it.  I used these brackets made for stage lighting:
DJ Clamp
I leave these clamps on the 4 overhead antennas.  I leave the 4 side antennas on their respective C-Channel.  2 antennas on 1 horizontal bar with the tripod mount in the middle.
This makes antenna setup quicker.  what takes time is assembling 30 feet of truss, mounting the FX9500 in the middle of it and running cables to the antennas. I made my own 20 foot cables using very thick cable and a math formula to calculate the correct length.  The long cables go to the side antennas and the short ones to the truss antennas.

[jonjkinzer]

Todd Gould I'm wondering how you mount the chip reader in the middle of your truss system.  Could you send a picture or explain it to me.

Thanks,

[Todd Gould]

I have an FX9500. It has mounting tabs on the bottom. I was going to
mount it to a plywood base and then attach the plywood base to the Truss.
However, I fount that I could just use large zip ties on those 4 tabs
around the box truss angle bars and it would mount perfectly flat. Then I
run the 4 short cables to the truss mounted Thinkify antennas. and my long
cables gout out each end of the box truss then to the ground to the side
antennas (2 on each side). I run an extension cord and long CAT6 cable to
the reader. All of the cables are bundled together with zipties inside the
box truss.

We assemble this on a set of chairs on the ground. Then we lift the truss
up onto the vertical supports and crank it up. You can see the bottom of
the FX 9500 in the picture below. We also do not mount the side antennas
one above the other anymore. We have 2 on a crossbar mounted to a single
tripod on each side. They are side by side and about 2 feet apart. If you
look real close, you can see my Bike Timing Chips on the front forks that
the side antennas pick up.

[Joost Huele]

Todd, a few more questions if you will indulge us :)

How are the tags typically mounted to the helmet? The fork mounting is clearly defined, but It seems to me there's about a hundred different ways to mount one on a helmet. Did you find any one method that's better than others?

Have you experienced at all with mounting behind the seatpost? Sort of like this:
http://cdn.media.cyclingnews.com/2011/07/04/1/gilbert_canyon_aeroad_number_holder_600.jpg
If so, i'm curious about the results, obviously.

Ahh, and one more. Will I be able to get away with a lighter solution? Using a 4-port hub and 2 overhead/1 on each side antenna setup, would I realistically be able to get good results? That would make the whole project a lot more feasable for me!

[Todd Gould]

Hi Joost,

Tags on the helmet:  Mounted centered left to right at the very back of the helmet with the RFID tag facing up.  use 2 zip ties to secure to helmet in this area.  Out of hundreds of racers, we never found a helmet that we could not mount the tag to.  The difficult ones are obviously aero helmets.  Have not tried that because for TT's we just use the fork mount since there is never a bunch sprint finish.

We have 2 or 3 helmets with chips mounted at the registration / packet pick up to show them how to mount the RFID tags properly.  Unlike Carbon forks, helmets made of foam and plastic do not interfere with the RFID energy.

I will post a picture of the helmet mount below.

Some racers leave the tag ends of the zip ties sticking straight up instead of cutting them off.  Gives them the bug look.  When I first saw this I offered to cut them off and they said no way man.  These are my antenna.

I have not attempted the seat post / brake post mount due to the extra hardware required.  No reason for it so far as the Fork tags work really well.

You can get away with the 1-2-1 solution if you narrow down the finish line.  Our finish lines are 20 to 30 feet wide and require at least 4 overhead to cover the span.  A better solution if I were in your shoes is to abandon the side read tag and only go with helmet mounts.  That way you can put 4 overhead.  You give up the redundancy of 2 tags per finisher, but you get better read rates overhead because there is no potential blockers to the signal.  In side read a bunch finish other riders can block the view / signal from / to the antenna.  The disadvantage to this is the cost and setup complexity of the overhead truss.

Our next race to time is Fouche Gap.  The finish is a 1 mile 6% gradient climb.  So there are no big group finishes and the fastest rider is only doing about 10 mph.  Therefore we will only do side antennas and a tag on each fork blade ( No helmet tags).  This eliminates the truss and all the time it takes to set that up and break it down.  You can almost hand score a finish like this.

My Mountain bike helmet.  (Ignore the light and the dent in the right side of the helmet...)  That timing chip has been on there for over 18 months.  ( I guess it exceeded my expectations on the durability test.)

[Brian Agee]

Fantastic post Todd! Perfectly explained the possible setups and the nuances of each each one. Great job!

[Joost Huele]

Todd, thanks a million for this writeup! This helps me out enough that I'm feeling confident to go ahead and plan my own setup now.
I love the bug look, good stuff! 

The 4-overhead solution sounds like a great idea, if only because of the reduced clutter on the ground. I don't mind adding a little bit of complexity to the truss setup, I'm sure I can come up with something cool and functional. 

I'm excited, lots more planning to do!

[Tim Irvine]

This is right in line with what I was going to try. Thanks for the leg work!

[Todd Gould]

No problem gents.  Glad to help.  Let me know if you need anything.  USAC Bike races are not easy to time.  Our next race on April 4th will have 10 separate races.  5 in the morning and 5 in the afternoon.

Thanks,

[countycoachmo]

Do you sell these tags for the helmets and if so, how much per tag?  We are doing our first bike race in a month and I am totally lost on set up and timing.  It is a time trial with riders leaving every 60 seconds.  Do i set up a different race for each rider or do I set up a chip start and then finish?  Some are doing a 5k and some are doing a 10k so I know i will need at least 2 races set up.

Joshua Moberg

MOGO Timing, Inc.

772.812.0050

[Todd Gould]

Josh,

Just saw this post.  Sorry for the delay.  How did the race go?

I know it is too late, but i would have setup 2 races and I would have used the capture chip start feature.  The key to making this work is to have your start line and your finish line separated.  As you know time trials have people starting and finishing at the same time so you could easily misread a starter as a finisher if you can not separate the lines.  What we had to do at our last TT was to turn off the finish line reader and only turn it on to read a finisher then right back off again.

I do sell the helmet/fork tags.  I usually brand them to your company logo.    

Let us know how the TT timing went.  Really interested to hear.

[Richard Williamson]

Hi,

I've timed bike events all year with an FX7500 and 4 x antenna. We originally mounted chips on the fork with antenna on the ground on tripods but the read rates were really poor, not much better than 50%! We use Alien tags. 

Recently I've moved to antenna pointing downwards from our truss gantry with chips on the helmet. This worked way better and we achieve around 90-95%%. The limitation seems when groups are passing, even if slowly. Individuals even at a high speed seem OK.

Our events are non-competitive so the speeds are quite low. We have typically 500 riders but these set off in groups of 20 over an hour and return in even small groups. 

I mounted the chip on the back of the helmet on a small piece of correx. This meant the chip was vertical but the reader horizontal, I believe a further improvement would be to mount the chips horizontal to get a better read rate. 

I consider a 2nd chip on each rider but not sure where to mount them, the fork got such low rates I doubt there is much point there. 

I did run a 2nd system connected to a FX7400 with 4 antenna, also mounted on the gantry pointing downwards. I ran this in Motorola Power Session and used it as a back up. It missed 17 reads out of 475, the Fx7500 connected to Brian's software missed more. 

I'm surprised that the newer reader is getting a better result. I've seen this in previous tests (though don't have the data).

Any idea on how to improve the read rates? At present we are doing a lot of manual work to manage the system. 

Cheers, 

Richard 

[Richard Williamson]

[Jon Beverley]

I've not timed any bike races but i've ridden in a few and the chips have either been 

1. mounted on a card cable tied to the handlebars (with a foam spacer on the back to keep away from the metal of the bars). RideLondon used these but they had mat antenna.

2. a tag that fixes around the seatpost and the tag then sticks out the back so it is in clear air.

I would also suggest using dog bone chips for the increased range as the size of the tag isn't really an issue on a bike. These are the ones I use for runners http://www.ebay.co.uk/itm/331686895410 if you want a free sample let me know.

[Richard Williamson]

thanks Jon, I've struggled with anything bike mounted. Perhaps the DogBone's would help though. If you don't mind sending me a sample please, that would be great. 

Do you mind emailing me to exchange details?

richard.w...@velo29.com

[Todd Gould]

Your poor read rate on the forks most likely has to do with how the tags are mounted on the fork.  You have to be a Nazi at the race start to make sure all tags are properly mounted.  The RFID portion of the bike tag has to be out in the open air.  Carbon is worse than aluminum and steel at absorbing the energy emitted by the RFID antennas.  I went through extensive testing on this and is how I came up with the fork/helmet tag I have been successfully using for years.

I have used 3 locations over the years:

Forks: with side mounted antennas

Helmet with overhead antennas

Disposable Bib on their back with foam spacer.

For slow finishes or where you have individual finishers, i do fork tags only.

For Fast finishes (in excess of 20 MPH) and more than 3 or 4 people in a bunch sprint, I put up a truss overhead and put 4 antennas over the road and four on each side of the road for the fork tags.  In this scenario, I have successfully read 30+ finishers crossing the finish line in less than 5 seconds with zero misreads. Video here: https://vimeo.com/69127702

[Juan Botes]

Hi Everyone, it has been awhile since I posted something on the forum, but I have to relay the past few races (MTB & Triathlon) that I have timed. I also the use the tag on the helmet: below is my instruction pamphlet I use at registration to coach the riders how to attach the tags correctly

I found by attaching the tag in the "Mohician hairstyle" format I get the best reads, as a matter fact 100% reads. I have timed two Trans Baviaans races with over a 1000 riders in each and manage to time even the interim check-points with a 100% read. The antenna set-up I use is a 2-antenna overhead structure for a 2 metre wide finish line and 4-antenna overhead structure for 4 metre plus finish line. The reader that I use is an Impinj 4-port, as well as an older FX7400. 

Below is an example of the results:

https://docs.google.com/spreadsheets/d/1azz2uK3980Nn2LyUc7lS4nu_JNPzdMKdyqvdFyj7W_s/edit#gid=1565264524

Regards

Juan

[Richard Williamson]

Hi Juan,

Well done getting the 100% read rates! Nice post.

What's driving the tag orientation? Is it the polarisation of your readers?

I've been mounting my tags in the position you advised not to (left to right). Do you notice a big difference if mounted that way?

Cheers,
Richard.

[rickcummins05]

Cycling Timing Questions:

Trying to figure out the best way to use ART in a Velodrome setting.

Today's testing went as follows:

Left Fork Mount ONLY - Bike Tags from AGEE - all chips Behind fork

FX9500 - 4 antennas - all on infield - @16' off track (for safety)

20-35mph speeds / 24 lap races / 200meter track / averages of 15-22seconds per lap

Most chips were read every lap - but orders weren't always correct.  

When the rider on the inside of the track was second visually they would usually WIN on the computer (.05 seconds difference)

Every so often only ONE rider per pack would be read (picked up) - sometimes an entire group did not register.

Corrections / Solutions????

I realize that antennas overhead is how most of the timers using ART are solving road race finish line issues.  This isn't a great option on a velodrome.

What are the opinions of timers on adding 4 antennas to the outside of the track?  This should improve read rates but will it produce much better results in the order of finishers?  if the winner is in the middle of a pack will the same result be produced as I got today - closest to antenna wins??

Access to below the track is not easy - that is why I'm looking for opinions prior to testing.

My guess is that side antennas only won't be anywhere near as accurate as top antennas with helmet tags as well - but I'm hoping.

Also:  Fork Mount Questions

Some forks are curved inward on the back half which points that back side of the tag towards the frame (inward).  This makes the tag NOT parallel to the antennas which I suspect is producing a 'late" read. Does anyone have a solution for this?

Misreads are very hard to deal with in this setting as each lap happens so quickly.  

As with all questions - thank you very much for taking the time to read and help.

Rick

[Michael Heyburn]

Rick, i'm also trying to time velodrome racing and did many tests and had similar results with misreads. I was using the reusable passive RFID tags that Brian sells.  Fantastic for road and mountain bike races, not so good for track racing.  Most often i was missing reads even if it was the only bike on the track.   I was also mounting the tags on the forks and reading from the infield about three feet from the finish line. Scary, i know, but any farther and i got no reads.  I thought about going overhead but logistically that's tough.    To complicate matters for me even further, i have a cell tower on the velodrome grounds that i know is interfering.   

I'm going to try left side helmet mount and see if that helps. The cell tower probably has me screwed.  Active tags may be the only answer. 

-mike

[Michael Heyburn]

Something that might help Rick, have you tried changing the gain on your antennas?  the beam pattern is 60degrees circular in the default 90dbic setting.  you may be getting riders in the same beam pattern that are several feet apart.  if you go to the max gain, 200dbic, your beam will be around 30 degrees circular.  If you then turn the antenna away from the riders about 45 degrees, you'll put the back edge of the beam on the finish line.  It takes some time to set up but that should fix the riders getting read out of order.   I at least had success with that.  Still missing reads though.   

I'm using the Motorola 9500 reader

-mike

[Raphael Leal]

What kind of antenna are you using?
Linear or circular?

[Ted Finch]

We have timed a few 'sportives' using a 'standard set up. ART, FX9500 and side panel antennas - circular polarised. Panels set high as cyclists had HuTags on wrists. A few tags were put on handlebars. No real problems with this set up.

The finish rate and finish density was low with  individuals or small groups rather than a massive pelaton at 30kph.

Ted
Keep it Simple Timing

[Raphael Leal]

Does anybody know how the tag should be placed based on the linear polarization of the antenna?

[JK Race Timing Systems]

Hi Raphael

You will need to know which way it is polariased , normally your antenna spec will show the polarisation pattern in relation to the orientation of the antenna.

For example, an impinj Threshold antenna is linear polarisation paralell to its length and it;s a slot antenna so If you mount the antenna vertically then it will want to see tags coming at it at a horizontal fashion, confusing I know but that is slot antennae for you.

The easiest way for you to know is just by doing a simple test, that is the beauty of linear antennae, they are very strong in the polarisation that they are designed but crap in the opposite polarisation.

Only use linear polarised antennae if you can really guarantee the polarisation will be always correct.

In the correct polarisation they are at least 30% more efficient than circular antennae

Hope this helps

JK