Flagtronic in car warning light system

[Steve Davis]

Originally Posted by mmi16

Possibly - does FT have a Timing Function to produce practice/qualifying/race results for each session?

Although their lit says it can provide 'millisecond' timing info, I'm pretty sure that it would NOT be possible to provide an accurate lap time. It might provide information "to the millisecond", but the inherent latency delays would most likely be able to only provide something in the neighborhood of 1 second accuracy, if that. Probably more like 2 or 3 seconds. With some additional work/design they might be able to do better than that by timing internally using the GPS info, but it would still seem a bit far fetched to think that they could get accurate results down to even a tenth of a second. The entire concept of design and use depends on that latency to be able to make it "INexpensive". To reach 'lap time' accuracy would require at least some sort of external 'trigger' tied to the timing line. Like the old light beams or the more recent trigger wires.

IMHO...
Steve, FV80

[BLS]

"To reach 'lap time' accuracy would require at least some sort of external 'trigger' tied to the timing line."

Are close races still decided by "photo" or by the timing system?

The addition of a start/finish line trigger combined with the already in place smarts and communication system is all that is required, plus a few lines of code.

EDIT: Maybe I'm wrong about the communications, haven't been following this closely enough. I'm assuming there is two way comms...

Edit2: Their website lists "Timing and Scoring", but I find no explanation. I'd guess the ability is built in for future expansion.

[Rick Kirchner]

Although their lit says it can provide 'millisecond' timing info, I'm pretty sure that it would NOT be possible to provide an accurate lap time. It might provide information "to the millisecond", but the inherent latency delays would most likely be able to only provide something in the neighborhood of 1 second accuracy, if that. Probably more like 2 or 3 seconds. With some additional work/design they might be able to do better than that by timing internally using the GPS info, but it would still seem a bit far fetched to think that they could get accurate results down to even a tenth of a second. The entire concept of design and use depends on that latency to be able to make it "INexpensive". To reach 'lap time' accuracy would require at least some sort of external 'trigger' tied to the timing line. Like the old light beams or the more recent trigger wires.

IMHO...
Steve, FV80

I'd love to know how you're coming up with these numbers. Latency has nothing to do with accuracy. I helped develop an optical missile tracking system that was accurate within inches - with a 3/4 second latency because of all the processing required to get the accuracy and find the needle in a haystack.

The "trigger' is a GPS defined line. The question is, how much does that line move? The longer the system is on the more accurate it becomes, especially if differential is used. If you use differential to establish the line then the accuracy question becomes one of how accurate the car measurement is. If you are doing this properly, every measurement is time-tagged and then Position data can be extrapolated between two tags. Not as precise as a specific line-crossing trigger, but its possible to create a method that functionally is the same.

[GregAmy]

I am aware of a 10-year contract between SCCA and FT stating no subscriptions will happen. After that...? I guess we'll see.

FT could easily do timing for us*. Of course that would mean it has to be mandatory for all cars, at all races, like AMB is now. But there's a lot that has to happen before these can become mandatory for all racing, such as getting the infrastructure installed at all race tracks; regions cannot easily manage purchase of these systems in addition to setup and breakdown for every race weekend (it's time consuming). Topeka owns their own system and has staff for setup/breakdown at each Super Tour (Road America already has a full system installed for the Runoffs).

At which point, of course, our $500 AMB transponders will be worth nothing...but I have gotten some 20 years use out of my two, for which I paid $150 each, so I can't complain.

*Any of us that have GPS-based data acq systems, even the cheapest ones, knows that lap times between data acq and AMB loop timing are pretty insignificant, especially if you set your S/F accurately. And we'll never "score" by those systems anyway.

[romoman]

I'd love to know how you're coming up with these numbers. Latency has nothing to do with accuracy. I helped develop an optical missile tracking system that was accurate within inches - with a 3/4 second latency because of all the processing required to get the accuracy and find the needle in a haystack.

The "trigger' is a GPS defined line. The question is, how much does that line move? The longer the system is on the more accurate it becomes, especially if differential is used. If you use differential to establish the line then the accuracy question becomes one of how accurate the car measurement is. If you are doing this properly, every measurement is time-tagged and then Position data can be extrapolated between two tags. Not as precise as a specific line-crossing trigger, but its possible to create a method that functionally is the same.

It's more involved than that. You have GPS receiver mounting issues, EMI from the car, atmospheric issues, and more that throw can cause GPS problems. Remember the GPS satellite only transmits at 5 watts, so tree cover, bridges, and troposphere effects are real.

[DanW]

I believe an important point is missed in this discussion.

The accuracy of "commercially available" GPS with 10-12 satellites is advised at about 10 meters(33 feet), so about 2-1/2 car lengths. Unfortunately our beloved government won't share the "Good Stuff Signal" with us, which is thought to be about 1 meter(39"). Nobody's talking but it might even be better than that. That's why there is such variance between our GPS based in car dataloggers and the wire buried in the track triggered by our Xponders when you look at the clubs lap sheets and your GPS data.

The Mylaps Xponder emits a 90degree cone downward from 6-24" within the car. The baud rate is 1mbit/second. Timing resolution is 2ms. That's far more accurate and consistent than GPS.

This simple device is inherently more accurate than a handful of obsolete satellites orbiting at 22,000 above us.

Racehero is gone, but Race Monitor is still with us, via subscription.@ $6.99/year.

[BeerBudgetRacing]

Racehero is gone, but Race Monitor is still with us, via subscription.@ $6.99/year.

Has anyone discussed taking over Racehero ?

[Rick Kirchner]

IIRC the USG started giving everybody the "good stuff" years ago (no more SA code required), and it can be "tuned' to higher than advertised accuracy in a given geographic area. Your cell phone is better than 33 ft. It that was the accuracy you'd never be able to accurately see your lines. 10 years ago in the middle east we were getting under 3ft.

https://www.gps.gov/systems/gps/performance/accuracy/

Total system performance is more about base station receiver configuration... For instance - you have a good receiver on an aircraft, less so on a weapon, but in many cases you do a "transfer alignment" between the two which gives the weapon more accuracy than it could receive on its own.

There are also systems that use GPS, Galileo, and Glonass. More satellites, more accuracy - its just a chip and striplines on a circuit board for each group.

GPS is on the road to integrating with Galileo and the next gen of satellites - Block IIIF will start launching in 2026 - but Block IIIs are still being produced and launched at about 2/yr. That integration will likely not only increase accuracy but reliability.

You only need good signal at S/F for timing and scoring - but if things get totally automated, what level of accuracy is needed to determine a pass under yellow?

AMD hasn't done anything really new with the system since its inception (for our use at least). The. Flagtronics system is good for F&C. Seems like someone would have come up with a GPS/transponder based system that does the whole thing - might be pricey but does more, especially if it integrates with data systems (instead of the data systems doing the GPS work).

In the end, its all about standards (or lack thereof) and incentives for companies to work together (or lack thereof).

If you really want to blow your mind consider this visualization: Your GPS position is best thought of as an oblate spheroid stood on end (a football) - GPS is less accurate in the Z plane than in x and y. This football is constantly changing shape - but your actual position is a spot constantly moving about on the surface of the football. Now, the place you are going to is also an oblate spheroid standing on end, constantly changing shape, and its actual position is a spot constantly moving about on its surface. What your navigation algorithm is trying to do is drive your spot on the surface through the destination's spot on its surface....

[romoman]

If you really want to blow your mind consider this visualization: Your GPS position is best thought of as an oblate spheroid stood on end (a football) - GPS is less accurate in the Z plane than in x and y. This football is constantly changing shape - but your actual position is a spot constantly moving about on the surface of the football. Now, the place you are going to is also an oblate spheroid standing on end, constantly changing shape, and its actual position is a spot constantly moving about on its surface. What your navigation algorithm is trying to do is drive your spot on the surface through the destination's spot on its surface....

I'm not an expert but have been involved in this a little. It really depends on what map reference system is being used. Tracks like NOLA provide interesting results as they move a lot and vary from year to year.

[rtjonespe]

I don't know the answer other than to say no reason it couldn't. I guess it lacks the actual start/finish line pickup.

ChampCar has been the leader in Flagtronics use. I know they are talking about doing away with the transponder and only doing Flagtronics. I think ChampCar will likely work out the bugs

[DanW]

IIRC the USG started giving everybody the "good stuff" years ago (no more SA code required), and it can be "tuned' to higher than advertised accuracy in a given geographic area. Your cell phone is better than 33 ft. It that was the accuracy you'd never be able to accurately see your lines. 10 years ago in the middle east we were getting under 3ft.

https://www.gps.gov/systems/gps/performance/accuracy/

Total system performance is more about base station receiver configuration... For instance - you have a good receiver on an aircraft, less so on a weapon, but in many cases you do a "transfer alignment" between the two which gives the weapon more accuracy than it could receive on its own.

There are also systems that use GPS, Galileo, and Glonass. More satellites, more accuracy - its just a chip and striplines on a circuit board for each group.

GPS is on the road to integrating with Galileo and the next gen of satellites - Block IIIF will start launching in 2026 - but Block IIIs are still being produced and launched at about 2/yr. That integration will likely not only increase accuracy but reliability.

You only need good signal at S/F for timing and scoring - but if things get totally automated, what level of accuracy is needed to determine a pass under yellow?

AMD hasn't done anything really new with the system since its inception (for our use at least). The. Flagtronics system is good for F&C. Seems like someone would have come up with a GPS/transponder based system that does the whole thing - might be pricey but does more, especially if it integrates with data systems (instead of the data systems doing the GPS work).

In the end, its all about standards (or lack thereof) and incentives for companies to work together (or lack thereof).

If you really want to blow your mind consider this visualization: Your GPS position is best thought of as an oblate spheroid stood on end (a football) - GPS is less accurate in the Z plane than in x and y. This football is constantly changing shape - but your actual position is a spot constantly moving about on the surface of the football. Now, the place you are going to is also an oblate spheroid standing on end, constantly changing shape, and its actual position is a spot constantly moving about on its surface. What your navigation algorithm is trying to do is drive your spot on the surface through the destination's spot on its surface....

Rick,

Thanks so much for the update. I had no idea the higher precision systems were released for commercial use. I always thought the GPS track positions were interpolated and averaged to give us the lines round the corners with the available technology. Very interesting explanation of how the GPS systems work now.

[Rick Kirchner]

I'm not an expert but have been involved in this a little. It really depends on what map reference system is being used. Tracks like NOLA provide interesting results as they move a lot and vary from year to year.

The standard is WGS-84: https://en.wikipedia.org/wiki/World_Geodetic_System

I'm not an expert either, but my targeting systems relied on this data and i got to work with people that were PhD cartographers and absorb just enough to be dangerous along the way.

[romoman]

The standard is WGS-84: https://en.wikipedia.org/wiki/World_Geodetic_System

I'm not an expert either, but my targeting systems relied on this data and i got to work with people that were PhD cartographers and absorb just enough to be dangerous along the way.

I'm in the same boat. In motorsports realm, there is a lot of misinformation and misunderstanding of how it all works. Most of my experience comes from dealing with the unexpected outcomes from it and why they happen. It's also interesting to hear what level of accuracy people think is needed vs. what they can obtain in their driving!

[Matt Clark]

This may be slightly deviating from the FlagTronics topic, but given that we already have AMB transponders (I know there is hate there too), what is really keeping SCCA from utilizing the AMB "on board marshalling lights" capabilities of what kinda sounds like the same function of FT?

Side note- the "on board results" bullet would be kinda nice to have the same official standings/results displayed on your dash (and RaceStudio later) as what MyLaps shows, instead of the GPS based ones we get now. Especially during qualifying.
Imagine if AIM made a small color dash version of the GS dash so we could use that instead of a 2nd flag display, and everything interfaced & functioned as one unit, instead of several standalone systems.

https://mylaps.my.salesforce.com/sfc...OepNycsz2E1vQk

[GregAmy]

...what is really keeping SCCA from utilizing the AMB "on board marshalling lights" capabilities of what kinda sounds like the same function of FT?

The first thing that comes to mind for me is that:

- It would immediately require replacement of all legacy AMB transponders (this program only works on the TR2 versions);
- It would immediately require everyone buying the subscription AMB transponders ($100/yr for five years, unless you gave them $800 for the "lifetime" subscription version...)

Nope. - GA

[Matt Clark]

The first thing that comes to mind for me is that:

- It would immediately require replacement of all legacy AMB transponders (this program only works on the TR2 versions);
- It would immediately require everyone buying the subscription AMB transponders ($100/yr for five years, unless you gave them $800 for the "lifetime" subscription version...)

Nope. - GA

Fair.
I know not everyone would want/like to get new transponders. I personally was already kicking it around, given the price of selling non-sub ones, if I could get proper timing on the dash. But if it was forced, then the non-sub market would not exist.
And the $800 lifetime sub would be tolerable to me (not happy, but tolerable) now that the formula car FT is $550.

[RussMcB]

Flagtronics and Regional Racing

I'm not sure if this has been discussed (I'm too lazy to search).

I may never do a Super Tour, so I asked if anyone knew about plans to use Flagtronics at regional races.

No time soon is the impression I got. An SCCA region (or division?) would need to buy their own system, and that will not happen for quite a while. And no race tracks are likely to take on that cost.

That was a response I got (in different words) when I asked on Facebook about it here in SEDiv.

[GregAmy]

I'm not sure if this has been discussed (I'm too lazy to search).

12 or so posts up, Russ. #844

I would be shocked - SHOCKED! - if we see this mandated* at any Regional race within the next five years. I'd almost take the over with some good odds for 7-10.

But that will depend a lot on how on-board the tracks are for the implementation (Road America already has it, I'm aware of others giving it serious consideration) coupled to how many problems racing regions are having getting butts on the corner (once you start paying for track flaggers then the economics of buying or renting FT gets better).

But my money is on "not very soon". - GA

*That does not mean it couldn't be "available" for those that have the receivers. However, many of the advantages of FT are lost unless everyone has it; without that, it just becomes another flag indication, which we should be looking outside for anyway...

[Hardingfv32]

At this time who responsible for the track related part of the FT system... SCCA HST team, Region or track?

Brian

[GregAmy]

At this time who responsible for the track related part of the FT system... SCCA HST team, Region or track?

"Whoever".

Super Tours: SCCA Topeka owns the FT system and carriages it to all events via the Tech trailer. Staff - I believe primarily The Scotts plus whoever else is available to help - sets it up at each race track the day before. Geofencing is updated via an antenna on the Grid. Topeka has a designated admin to manage the functions and interface with Operations during all races. The Scotts break it down and pack it up into the trailer after each weekend (chip in if you have the time).

Runoffs: Road America has the system already permanently installed, no setup/breakdown needed. I believe I recall seeing both SCCA staff and a track person managing it during the '24 Runoffs (but I didn't know everyone that was in Control).

Next Runoffs, wherever that is: Probably going to be one of the two above, depending on if the track has it installed or not.

An SCCA-sanctioned track in New England I know considering it: "If" they do it - and it's a big financial nut to swallow, something like "starting at" $15k - the track will install it permanently and possibly provide the management of the system -- for a per-event fee, of course.

Individual Racing Regions: Sure, they could buy the system and carriage it and set it up and tear it down. But that is a lot of work and a lot of money.

Hybrid Version: Combo of the above two. Regions and their local track invest in it together and leave the system in place for other orgs to use (for a rental fee of course, to recover those costs).

Everywhere else? All other orgs? Dunno. "Whoever".

It's all gotta get fingered out.

[romoman]

"Whoever".

Super Tours: SCCA Topeka owns the FT system and carriages it to all events via the Tech trailer. Staff - I believe primarily The Scotts plus whoever else is available to help - sets it up at each race track the day before. Geofencing is updated via an antenna on the Grid. Topeka has a designated admin to manage the functions and interface with Operations during all races. The Scotts break it down and pack it up into the trailer after each weekend (chip in if you have the time).

Runoffs: Road America has the system already permanently installed, no setup/breakdown needed. I believe I recall seeing both SCCA staff and a track person managing it during the '24 Runoffs (but I didn't know everyone that was in Control).

Next Runoffs, wherever that is: Probably going to be one of the two above, depending on if the track has it installed or not.

An SCCA-sanctioned track in New England I know considering it: "If" they do it - and it's a big financial nut to swallow, something like "starting at" $15k - the track will install it permanently and possibly provide the management of the system -- for a per-event fee, of course.

Individual Racing Regions: Sure, they could buy the system and carriage it and set it up and tear it down. But that is a lot of work and a lot of money.

Hybrid Version: Combo of the above two. Regions and their local track invest in it together and leave the system in place for other orgs to use (for a rental fee of course, to recover those costs).

Everywhere else? All other orgs? Dunno. "Whoever".

It's all gotta get fingered out.

Do you know how the MyLaps system works? I thought it went off multiple timing loops, which means very little install by the region/final users, no? How expensive is the admin equipment?

[GregAmy]

Do you know how the MyLaps system works? I thought it went off multiple timing loops, which means very little install by the region/final users, no? How expensive is the admin equipment?

In New England Region - and, I imagine, most others - NER owns the computers and timing equipment that is used for AMB lap timing. That is carriaged to each event and placed in the T&S area, where it is setup and managed by NER members.

Many moons ago, the tracks agreed to bury timing loops in their tracks; you can see the slots they dug in which to lay those timing loops. Each loop is routed to T&S where they are physically connected to a "decoder". Each decoder is then connected with an Ethernet network to the "server". The "server" is then accessed via one or more laptops that parse the information and create thos charts (as well as RaceMonitor data, if the track has Internet access).

The tracks are responsible for purchase/maintenance/repair of the timing loops, NER is responsible for purchasing everything else.

As you can probably guess, the physical setup for AMB is significantly less demanding, since all it's doing is watching for a transponder to pass nearby. DING! DING!

Some other differences:
- AMB timing loops are permanently in place, so there's no setup/breakdown of them. We just connect/disconnect the decoders to/from those loops and setup everything else.
- FT, on the other hand, usually requires setting up repeaters and antennas around the track for proper coverage and communications (depends on the topography)
- AMB decoders are plugged into wall sockets for power. The timing loops are powered through those.
- FT tower equipment is plugged into the wall, too, but the repeaters around the track require generators/solar panels to keep batteries up to snuff (assuming no nearby power).
- Timing loops are physical and do not move. The decoders just sense when a transponder goes over them.
- FT requires setting up "geofencing" zones for each section of the track (and pit lanes). These may require adjustment over time (e.g., when a corner is not "manned") and must be maintained for each individual track.
- AMB has no requirements for the corner stations.
- FT requires controllers at each corner, along with trained persons to monitor and manipulate the local FT "flag" conditions. This also requries someone to distribute and recover the corner controllers, plus maintain and charge them.

I'm sure there's other factors, but that's just a few nuggets..

AMB is a simple system because it does one thing: DING! But FT allows us to collect and project addtional information and notifications.

I do not know the costs of the AMB decoders, plus laptops/network equipment. My guess is that it's substantially less than the FT systems (and it was paid for long ago). It's certainly much eaiser to setup/breakdown.

My two bits. Jump in with any other thoughts.

GA

[John Nesbitt]

...

AMB is a simple system because it does one thing: DING! But FT allows us to collect and project addtional information and notifications.

...

In addition to the hardware (decoder, server, laptops etc.), the T&S system has also software (Orbits) running on top of that hardware, and doing much (not all) of the work necessary to produce times, grids, and results. There is a fair bit of work/skill involved, especially if the Region is supplementing/duplexing the AMB/Orbits system with an optical timing line and/or manual scoring.

But, as Greg points out, the infrastructure for T&S is pretty much install/setup once.

[romoman]

I do not know the costs of the AMB decoders, plus laptops/network equipment. My guess is that it's substantially less than the FT systems (and it was paid for long ago). It's certainly much eaiser to setup/breakdown.

My two bits. Jump in with any other thoughts.

GA

I asked my question poorly. Do you have any experience with the MyLaps Race Control package that does the same as Flagtronics? I know the third player is EM systems which is what Indycar (and I think IMSA) uses. I don't know the cost of those in relation to the Flagtronics.

[John Nesbitt]

I asked my question poorly. Do you have any experience with the MyLaps Race Control package that does the same as Flagtronics? I know the third player is EM systems which is what Indycar (and I think IMSA) uses. I don't know the cost of those in relation to the Flagtronics.

No experience with MyLaps Race Control, but, as a practical matter, I very much doubt that there is any chance at all of SCCA's adopting it.

Considering all the sturm und drang around the process, and the fact that a large number of members have bought and installed FT, no matter how reluctantly, it would be a very brave SCCA superstructure that floated the idea of a competing system.

[GregAmy]

Ditto with John: I got nuthin' on the other systems. Nor any significant interest in learning more...

SCCA done chosen its dance partner.

[BeerBudgetRacing]

Are there tracks with permanent MyLaps Race Control installs?

Because there are some tracks with permanent Flagtronics installs.

It's really going to depend on the insistence of the different race series and the tracks desire to host them.

And the SCCA doesn't seem to have the influence anymore. Couple weeks ago SFR lost 1 of the 2 days they had scheduled for Laguna Seca to some other organization. I think not many years ago they ran there 4 times a year. Now it's 1 !
SFR supplies corner workers for Indycar and other events. I wonder if they going to continue or get replaced?
Just think about the visual impact of an illuminated flag board! I can see the neighbors complaining now.

Depending on where we want to race, we may end up with 2 systems in our cars....