I might have been missing some common knowledge all these years but I've never really heard of people discussing the way laps are timed in F1. I guess it might not be much different from most major racing series and their timing systems but I'd still to know what the specifics on it are.
I imagine either some sort of magnetic strip in the track that registers the pressure of the car as they pass over it. Either that... or lasers.
Enlighten me!
And I'll also take this chance to introduce myself. F1tech's been my favorite source for news and statistics as well as the brilliant community for discussion. I decided to register and plunge in 'cause this question was just bothering me more than usual today while I was watching classic clips of the old Interlagos circuit races. What a beast of a circuit it used to be, wow.
I don't know the specifics of F1, but they shouldn't deviate much from the general basic idea of motorsport timing loops ...
Each car has a transponder unit fitted to it, the transponder has a unique ID and it will transmit this ID when the unit is energized by passing over a timing loop buried in the track surface.
The timing loop is effectively two bits of wire that run across the track and back, (where the white line is painted). It supplies the radio signal that energizes the transponder, and detects the unique ID transmitted back. You use the "crosstrack" parallel lines formed by the two halves of the loop to determine when you first detect a transponder response on the first bit of the loop, to when you lose it on the other half of the loop ... the middle point between these two events occurring is the time the car was half-way between loop elements, and the transponder directly above the white painted line.
Not sure what the exact F1 regulations are for the fitting location of the transponder ... all the public version of the rules say is
"20.4 Transponders:
All cars must be fitted with a timing transponder supplied by the officially appointed timekeepers. This transponder must be fitted in strict accordance with the instructions detailed in the Appendix to these regulations."
As far as the location goes, the primary transponder is mounted at the bottom of the tub, below the front axle centerline. The cars also have backup transponders at the rear of the car just below the rain light.
Below is the official appendix to the 2010 f1 regs. It's actually quite an informative pdf as it even has a number of CAD drawings of camera housings etc. .
Transponders transmit their code constantly when energized. The signal is relatively weak and requires that the transponder be not higher than a fixed height above the surface, ususally not more than appr. 60 cm. Every loop goes to a receiver with a decoder which decodes the unique ID of the transponder and sends it to the timing software on a main computer.
As soon as the start is given, the software is activated and when each car passes over the S/F line loop for the first time starts counting the time till it crosses the sector loops and the S/F one, the latter giving the total lap time.
As far as I know there are auxiliary loops on the pit lane to determine when a car enters and leaves and maybe there is a loop now at the first SC line to track eventual rule transgression for overtaking when SC leaves the track and race is restarted. Also the SC has its own transponder and is identified by the system.
There are backup systems which vary on different levels of racing. Starting from a light beam which flags an error if a beam cross signal is detected without a transponder response and the operator must identify the car manually to second transponder in the car.
The starting list is entered into the software with the unique ID's of the transponders assigned to each car, so the system is able to identify every car on the track and in the pits.
Note: I am not a professional in the field and the above is based on my humble experience on local level and on observations of news and live timing during races.
F1PitRadio @F1PitRadio : MSC, "Sorry guys, there's not more in it"
Spa 2012
I don't exactly know how it is done in F1, but i know quite a bit about timing systems in other racing series and i think in F1 it is done in similar fashion. What Feynman and DragonFly described is usually called RFID. I really doubt anyone would use passive transponders (powered only by the magnetic field generated by the coil, or loop as named before), as suggested. Passive systems require very strong magnetic coupling and are not reliable at all . Probably they use semi-active RFID-like transponders operating on 433mHz band, as in other series. I've worked with available systems and done such system myself once and must say that it is not 100% reliable technology, as there is no collision avoidance at all and it is not very accurate either.
btw... I've made working (moreless ;p) prototype of timing system based on completely different concepts, but it took all my spare money in process and costs of various testing and certification procedures are waaaay out of my reach right now ;p but maybe i'll manage to finish that project one day ;]
The car carries what is called a transponder, as mentioned before has a unique code (car identity) that is sent into a timing loop built into the race track. There are several loops placed around the track which is why you see timing splits comparisons.
What most don't know is that AMB (who makes the timing system and equipment) has a "special" transponder unique to F1 (as far as I know) that can receive info from the timing loop. Mostly for commands for the onboard camera (like scrolling the protective film when blotches appear), it also has the capability to receive info, for instance regarding the flags on that section of track, and that could trigger lights/indications on the dash for red flag, safety car, lift trucks and corner crew on track.
"Driving a car as fast as possible (in a race) is all about maintaining the highest possible acceleration level in the appropriate direction." Peter Wright,Techical Director, Team Lotus
I realize this thread is a bit old, but it's still searchable on the web so if anyone is interested I thought I'd put in a few cents worth from someone that works in Motorsports Timekeeping as a professional. Forgive my very detailed explanation of this, but as drivers take their driving seriously, we timekeepers take our timing seriously!
There are several assumptions I've read here that are correct, and some that are bit off, I'll try to clear them up.
(As F1 is proprietary and there is much secrecy about scoring, I'll not elaborate here, but suffice it to say the basic principles are similar to what I will attempt to discuss here...)
F1 timing, a function of F1 Timekeeping using a variety of proprietary software, custom developed Tx (transponders) and other sensor specific onboard equipment.
The differences in F1 really stop there compared to other sanctioned events, such as FIAWEC, FormulaE, IMSA, Indycar, NASCAR, TransAM, SCCA, etc. (Additionally, MotoGP, Cycling, 10k's and even RC cars are also timed with transponders as well)
Amateur events and professional events are timed very similarly, with the major difference being with the type of transponder and track equipment.
For those interested I will try to break it down a bit here:
F1, Like FIAWEC and IMSA currently utilize a more advanced transponder, such as an AMB X2 data transponder. AMB is part of MyLaps, an equipment manufacturer as well as a software manufacturer.
What's a Tx?
The current X2 transponder (Tx) is an onboard vehicle powered transponder that emits a constant digital signal (this was previously referred to correctly, and then incorrectly assumed it was RFID. RFID it is not. RFID will work, but the SNR is woefully inadequate for racing purposes). The space the Tx occupies on each car, the height to track, the location from the front axel are all sanction specific, but due to ACO and FIA similarities/crossover, mostly the same in most professional motorsport for sports cars, and in F1, Indy and NASCAR are proprietary to those rule sets.
Loop?
The digital signal is received via a timing loop (also in almost all cases made by AMB/MyLaps, for reasons described in a bit) that is cut into the track. This loop, a specific size and depth, (typically 24"xtrack width) has proprietary resisters at each end, and culminates in a BNC fitting that is connected to a trackside box (TSB) that contains a decoder (AMB), a network controller, a battery controller and a DFO (fiber optic) TX/RX modulator (S/F and S/F pit are different as I will explain later).
The decoder also has an input for a GPS puck. The GPS signal is synchronized among all decoders on the track so that the exact time signal accurate to .001 is shared by all TSB's.
The signal then travels via either a fiber connector, a CAT5 cable, or a CAT5 cable to a NanoStation (a RF transmitter via WiFI/Nettalk to a tower receiver).
To the tower...
So, the basic system starts with a Tx, say on car 10, whose Tx # is a 7-8 digit number. That number is received at a particular loop on the track. Literally, the Tx signal is measured but the decoder in signal strength as it passes over as "hits" and the highest strength "hit" is recorded as the cars "passing". Each car Tx has many "hits" and this is recorded in the MyLaps X2 managers as a selectable graph, and looks just like a bell curve...low signal hits at the left of the graph and as the car passes over the loops, the signal strength is greater and the "hit" strength goes up, and as it passes the absolute center of the loop, the highest hit is recored as the passing...
In amateur, AmPro racing (SCCA, TransAm, HSR, etc.) there is typically only two loops of interest to the timekeeper for scoring laps, and that's the S/F and the S/F Pit. Basically one line. For professional racing, there are as many as 24-40 loops in a track, all with TSB's, decoders seeking the data. The TSB's are located around the track, at specific places of necessary interest to score the race. The S/F and S/F Pit are almost always hard wired straight to the tower due to their location, and therefore those decoders are rack mounted in the T&S room.
In additional to S/F, S/F Pit, typically there is Pit In, Pit Out, Pit Speed, additionally P1-P7 at most tracks (loops along pit lane), S1,S2 (Sector loops), data loops for X2 data acq. and also S/F Backup, S/F Pit Back Up, and various track limits, back door entrances and testing loops.
Tracks?
Tracks such as Daytona and COTA for instance have fiber rings, so there is a limited use of nano stations to xmit data to the tower. The decoders are centrally located at some tracks with fiber, so access to the rack to a switch to connect all the decoders to the x2 manager is handled versus a central bullet receiver on the tower to pick up the signal of a myriad of decoders using RF WiFi. Street circuits such as Monaco, Mexico, Detroit and the like require a temporary fiber ring and in some cases nano stations for RF WiFi. Road Atlanta has partial fiber, so we still have to set up nano's on most of the track (some with repeaters to get around the hills and trees!).
Drivers?
So the equipment side is a complicated maze of a very simple system. Loops > decoder > tower. The info each car sends isn't limited to just a car specific "phone number"...Tx 1234567 for instance, also xmits alternate side bits designating things such as driver (pilot #). Each "team" driven car (Sebring, LeMans, Daytona for sports cars) has driver changes which are handled via a rotary dial or a multi switch or a specific pin in a driver comm plug that tells the Tx to sent 0-5 as a side code to the decoder as well. In this way as the Tx passes a loop (say Pit Out) it alerts the system as to which drive is plugged in.
NASCAR for instance uses a proprietary Tx that sends the actual car number as the unique number to the decoder, as has been the case in F1 in the past. In FIAWEC as well as F1, the Tx is formed in the car tub, and is not removable. In the event of failure a back up Tx takes over the function, and in a worse case scenario, the car returns to the pits to receive a battery operated Tx as a loaner.
Both Tx's are transmitting constantly, but in X2 Admin Manager, the timekeeper is made aware of which Tx is to be used as primary, and the alternate Tx # is blocked or hidden from the timekeeping software via the data stream.
2-way?
The most recent X2 Tx's used in F1, FIAWEC and IMSA have two-way communication capability. Typically bits of data can be sent to the Tx as a question, and bits of data can be received from the car. In essence - Q:"Are you over boosting?", A: "No" (0) or "Yes" (1), etc. This can be done via the timing loop, or in the case of F1 and FIAWEC, via alternate data loops specifically set a slow portions of the track to be able to accurately communicate on one loop to another. X2 Tx settings are known as CAN settings and are set prior to the event and are set to specific codes for Tx and Rx. This is done during the safety check/scrutineering, along with all other checks (Yes, in FIAWEC, Burt's car, IMSA and Indy, all safety cars, EV's and even pace cars have transponders so they can be hidden from the system as competitors).
F1, like IMSA and FIAWEC, uses a proprietary system to xmit safety information and other track conditions to the driver, FIAWEC uses a separate system from the Tx. Delphi, Magneti Marelli, and Bosch to name a few used in Pro. In addition, in INDYcar and IMSA, the Position lights system (Pos. digit) started out as an RF system and is evolving to be sent via Tx, but in FIAWEC (three lights) is still being sent via Marelli RF.
How does it know?
Ultimately, even the best timekeeping software (search Orbits, Alkamel, TimeGear or SMT) still relies on this digital phone number passing the start finish loop to score a lap for the competitor. When a Tx is received, it's logged at the decoder with it's info, a timestamp from the GPS and the decoder it passed. This is info is matched with a database record - e.g., Car 10 is driven by driver 1,2,3 etc, engine type, constructor, sponsor, owner, team, etc. Each passing is then sent via the data stream to another piece of software (the actual scoring/timing part) where an operator is watching. The passing shows as a line item, along with a lap time, (calculated with math: this time stamp - last time stamp at this loop = lap time), which car has the longest distance with the shortest time (position) and how many pit stops, fastest sectors etc.
It quickly becomes a wealth of information with each passing, and live reports can be requested in separate windows for a particular car, driver, lap of the race, etc. This is often used by Race Control to determine issues...
What if?
So even the best equipment breaks, fails or trips up. In the past, an operator of the laptop with the software timing the race had to be vigilant by watching each car as it passed the S/F and compare it to the passings appearing on the S/F loop. People called "tapers" also worked in race control, T&S to watch the S/F, and get this, right down each car number as it passed in a list, and then turn the page when the leader started another lap. Often 2-10 tapers were required for enduro's, and typically 2-3 at a time so you could collect the "tapes" in case something was amiss with timing software...like a Tx not getting a hit, but the car completing the lap none the less. In this case, a high speed light sensor (photocell) is employed (Tag Heuer) and each cars passing is recorded (sometimes as a group of cars as one) and then compared in the software electronically to the passing made by the Tx timestamp. If the number is reasonably close (like .002 e.g., , a setting that is adjustable for rain) it blanks out the photocell and accepts the Tx. If there is a photocell and a Tx and there is no relationship, the timekeeper can look at the car, see the number, modify the photocell record to the correct car and accept it as the official passing. If you miss the car, you get the "tapes" and see what car showed up in the fray among the Tx's you got hits on!
Currently, tapers have gone the way of the dodo, and we use high speed camera systems located on S/F, Pit In Pit Out and other Track Limits areas, as well as rely on TV broadcast feed (12-30 cameras around the track) and update the information. Software has become very adept at using the input from all of these networked devices and comparing the information received and weeding out the bogey's.
Some aspects of X2 Timing Manager and other software also display a sequencer, which allows the operator to see a specific car travel around the track as it hits each loop. This info is fed into an algorithm (last lap time, predicted lap time) and is often used to display a track map with the cars circulating on it. The more loops used to build the sequencer, the more accurate the info as to position, lap traffic and passes on track.
Teams?
In addition to this information being sent via a data stream (believe it or not, an "RMon" feed, using $ lingo and just text) it is sent to the announcers, TV and the teams. For the teams, they can also see their specific cars individually as well as the whole race. And BTW, since this "stream" is captured live, using the software, you can actually "playback" the entire race, second by second, lap by lap, over and over.
Costs?
Typical X2 packages for teams with wiring and looms can run as much as $12,000, while AMB Tx's for Amateur racing is under $800. A fully equipped decoder TSB can run upwards of $10,000, not including the battery system needed. The average # needed? About 32, so $32,000 just to hook up to the track. Each venue typically is responsible for maintaining the loops, but in some cases, we need to repair them, cut in new ones, or move them. This is done in the week leading up to the event. Timing at FIA F1 tracks is wonderful, because the tracks by requirement have the entire system from loops to decoders already up and running and we only need connect our computers to the system via their network. Other tracks, street circuits and older venues require days in advance to set up, install systems, cameras and decoders before we can go live.
Most timekeeping crews travel long hours, work long hours, live all over the world 26+ weeks a year, and often don't get a chance to shine a little light on the magic that happens in the tower. We are part of the circus, rub elbows with drivers, teams, fans and bosses, and are just as smitten with racing as anyone. A crew of 6-10 people are responsible for bring the races to the public via timing and another 10-15 for race control, and we all take our work very seriously when the lights go out. The data is always analyzed and the correct information is put forth. There's nothing else I'd rather do.
Thank you for the chance to be so verbose on this subject, and if I have over indulged your patience, I apologize!
⏱ Timekeeper by day, F1 Driver by night (whilst sleeping of course...)
Thanks for sharing a really interesting post. Things are always more complex, and interesting, when they are viewed from the inside.
Fortune favours the prepared; she has no favourites and takes no sides.
Truth is confirmed by inspection and delay; falsehood by haste and uncertainty : Tacitus
The accuracy of the system is 1/1000th (by FIA req, and as a default of the GPS technology). That said, the inaccuracies that potentially could eek in are variables outside of the T&S system, some of them are, but not limited to:
-Improper placement of the Transponder (this would have to occur post scrutineering)
-Improper voltage or operation of the Transponder resulting in low hits (hits = +10 to over 900, with a signal strength of +/- -30 )
-Failure of the Transponder
Since the modern X2's have not only vehicle power input via the wiring loom, they also have internal battery backup (in the event of vehicle disablement, the Tx even sends passings while on the crane in tow on track over the loops!)
Even with Transponder failure covered in my last post, photocell and high speed camera images use synchronized GPS data as well, so correlating the image to the passing can be down frame by frame.
At the professional level, the timing operators are fully trained and autonomous to fix issues with scoring, and as proper race officials, they are usually the last word on the system data. There is a hierarchy established as to the chain of command at events, and typically it falls as:
#1) Race Director - The end all decision both live and post race
#2) Safety Director - The decision to stop a race for track conditions, usually imparts this to the RD
#3) Starter - When the checkered flag is dropped, the race is over. The RD has to over rule this
#4) Chief of Timing & Scoring (Timekeeping) - Rarely questioned, the expert on the equipment, data and results.
and so on..
In rare occasions, there may be a delay in reviewing/correcting the information, and this is usually due to equipment failure or some anomaly (Why results are "Provisional" until made "Official") . Even time penalties are made as entries during the event, and subject to being transacted during or post race.
Since the system is redundant (there are two timekeeping servers running at all times, and redundant backups of each, all running on conditioned power, so data loss isn't a factor either, LOL)
So all in all, the amount of data we receive, the ability to cross reference the data with multiple devices, and the ability to review and correct on the fly, the accuracy of the results viewed by the TV audience is pretty much spot on.
If I get some time, I can share some images of the equipment, setup and screen caps if anyone is interested...
Cheers!
⏱ Timekeeper by day, F1 Driver by night (whilst sleeping of course...)
Very good question...something that is on the minds of many over the years.
Without being too boring, I'll try to address this without being overly technical-for the purposes of this post, I will presume that there is a general understanding of managed networks and their security, firewalls, and e.g., white listing of mac addresses for systems connected to them. I cannot discuss F1 security and specifics here for obvious reasons due to confidentiality obligation. I can try to relay my perspective on general knowledge of professional timing systems.
So...not matter how bullet proof the electronic timing systems are, one fact remains:
Ultimately, winning a race requires being the first competitor to go the greatest distance in the shortest amount of time once the lights go out to the checkered flag.
That said, regardless of how many electronic devices or how the data is collected, the results of that data still have to bear out what happens on track.
Let's look for a moment on how the existing systems are set up and go from there:
-Track loops are fixtures of the track, so any "result" of crossing a loop for one car is a result for all cars crossing a loop.
-Decoders are usually not "secured" other than from the weather, and are placed in locations where there is limited foot traffic (Pit Lane, Corner Marshals)
-Team equipment such as Transponders are factory sealed, usually integral to the body work, and operate with checksum systems.
-If needed, decoder RFWifi uses hidden SSID and exists on it's own private network.
-Fiber Optic systems are extremely secure, and require a TX/RX loop for communication.
-Each segment of the system occupies it's own private network within the system, so teams for instance, can't ping timing equipment, and RFWifi doesn't see other systems.
Let's accept this first: Basic racing weeds out almost all manner of cheating the timing or scoring system...by that I mean you can't make your car go "faster" on track via the timing system, because in order for Car A to finish ahead of Car B, it has to do that in front of a live audience of participants and fans.
So a few words on what wouldn't work-
-Physically cheating the track: High Speed cameras unveil any monkey business on the track, whether from onboards, TV, fixed vision or track limits cameras, regardless of what the cars electronic data indicate. In some cases, a chicane will have a loop to catch short cuts, as well as a motion camera, so placing a car where it shouldn't be doesn't happen without notice. Corrections are usually made via the penalty system rather than the timing system.
-Cheating the Tx location: We validate the location electronically during scrutineering by slowly passing the car over a test loop, so moving the location of the transponder on the car to gain a 1000th here and there would be detected. Even an elaborate system to have it move while racing wouldn't yield anything other than a fast lap because of the next item:
-The only portion of the track that is official "scoring" is the S/F line. It it observed via a high speed GPS synchronized cameras with a time stamp. So if a car was able to somehow speed it's "data" time up electronically, the time stamp of it crossing the finish line would be unaffected, so all for not, the car is still timed correctly.
-Pit stops or pit lane activity is closely watched with visual systems, and typically not subject to timing errors. The only exception is driver ID, and in F1 this isn't an issue, since there is only one driver per car. In multi-driver series, as I mentioned there is a plug/ID switch system, and that reports the current driver to the system at the pit out loop. This is the one factor of timing that is often an issue, since stint times are calculated to the second, and are subject to the sanctioning body's rules. FIA backs up the electronic detection of a driver with pit lane officials who identify the driver by site in addition to any timing and scoring method. We also employ a team instant message system that can query the team as to the identity of the driver, and this is cross checked with the officials on the pit lane if needed.
So let's cheat-
Let's say that somehow one would want to either hack the system or corrupt it to gain an advantage. They'd have to push through the network security and their presence would immediately be detected anyway. But for fun, let's say they do that. What would they attempt? Overwriting the data stream with information to give them an advantage would essentially corrupt all the other data, and it would be easily detected. They'd have to delete their car's data and replace it with something else, and do this of course all the while trying to make it match what it "looks" like is happening on track.
Even qualifying laps are still synchronized with a motion camera, so data from the electronic side has to reconcile with the time stamps electronically sent from the camera within the software...any discrepancies alert the operator and are dealt with by researching the offending lap or passing. So you'd have to hack the decoder (which is located within our eye sight in the timekeeping room), hack the camera, and do all of this unnoticed prior to the event and oh by the way, have this hacking endure all the free practices, officials testing, qualifying sessions, support series runs, and the race, unnoticed. Highly unlikely.
So even if they had open use of the system, trying to "insert" something into the stream we use it would be useless. Think changing 1+1=2 to something else and still getting 2 without changing any of the numbers or adding variables or funny commutative properties. Remember the timing is 1/1000 accurate, so their super computer would have to do this within those parameters as well. Money spent elsewhere would yield a better result. Even if you wanted to bribe everyone (let's not discuss conspiracy theory or how many people can keep a secret!), you'd have to bribe audience as well!
Additionally, the timekeeping room is restricted from observation by competitors and their assigns, so being there would throw up a red flag (no pun intended).
All in all there are a few places that *are* vulnerable:
-Collapsing the entire system. We'd see that, reboot it, fix it, or just cancel the race! Problem solved.
-Interfering with another competitors electronics. This is often done unintentionally since modern transponders can actually "spam" a loop with too much signal, and interfere with the operation of the decoder. The decoder still records the information internally, but pushing the data through the pipe gets delayed a bit...like 2/10ths of a second delayed...remember, with each hit or passing, the decoder is only sending simple text line by line, something like ("1234567","2","5","-35","0","06012017 12:00:00:001"), so even spamming doesn't slow the system down enough to worry about. In any event, even blotting out someone else signal would still be noticed and compensated for by photocell or camera, and the problem would be investigated during the race.
-Tampering with decoders or removing equipment. The health and status of every device is constantly monitored with software, QOS signal health, voltage condition and other data. All equipment is on battery back up, including RFID Tire systems and even Pit lane exit lights. We'd see any anomaly and send a tech out immediately to investigate.
I hope I've addressed any concerns in the general area of corruptibility, and shed some light on why it's really not feasible to attempt to collude to alter any timing system. As "unseen" as it is to the general public, internally the system is extremely transparent . In addition to our timekeeping, keep in mind that individual teams are using beacons and other devices to determine if our timing is accurately representing what they are logging for their car as well as their competitors.
Since there can be only one "official" timing, the end result is an open system where everyone has to agree on the result. Discrepancies are dealt with within the body of the rules, and protests are lodged and resolved if there is any anomaly.
Cheers!
⏱ Timekeeper by day, F1 Driver by night (whilst sleeping of course...)
Great reply, I've always wondered wether qualifying results may have been rigged and wether the teams could discover this if a fast one was pulled on them.
.
. Probably they use semi-active RFID-like transponders operating on 433mHz band, as in other series. I've worked with available systems and done such system myself once and must say that it is not 100% reliable technology, as there is no collision avoidance at all and it is not very accurate either.
