From, Chad Dornsife,
To: Roger Hedgecock Show, AM 600 – San Diego, CA



In support of your commentary that strongly suggest there is something very wrong here with this whole RED LIGHT CAMERA program. WELL YOU’RE RIGHT!!! We have all heard the Carnage and large numbers quoted by camera supporters – A NATIONAL CRISIS! The reality, this number is in the hundreds for all causes of signalized intersection accidents (approx. 1.6 percent of the nations 42,000 plus total fatalities).


What we do know, it appears that a large portion these 700 plus intersection fatalities were caused by the negligence of the FHWA in not requiring local authorities to comply with federal safety standards – thereby, directly contributing to this epidemic of improperly timed traffic signals and poorly engineered intersections. Further, the leading motorist caused factors at signalized intersection is failure to yield on left turns and pedestrian accidents that have nothing to do with red light running.

How do we get ourselves in these situations? We have the Highway Safety Act of 1966 which MANDATED sound engineering practices to be the law of the land with one national standard – administered by FHWA and its MUTCD.

Our government is schizophrenic, its agencies are paralyzed because of unfounded political conjecture ruling over humble engineers who go to work each day trying to do their jobs, while maintaining them. Leaving us with regulatory agencies doing what most of their staff knows is wrong, but they do it anyway. As a Nation we crucify a company like Firestone for its tire failures, and look the other way when the oversight agency is itself (DOT) sponsoring the killing of many times more by its actions – read inaction.

The FHWA is no longer defending engineering standards and practices and has openly caved into political pressure (Millennium Edition of MUTCD). We have a national press that has little time in today’s world to check the veracity of public press releases in its quest to provide massive amounts of daily content. Then we have the Federal Agency of Urban Myth called the National Highway Traffic Safety Administration (NHTSA), whose entire existence is based on like misinformation campaigns. They have an ad agency that is contracted to invent a crisis, launch press campaigns to build public consensus that something must be done, and then the need to intervene. This is how government agencies support their existence. This agencies constituents are not the people, it’s the enforcement agencies, courts, cities and insurance companies whom profit from these enterprises.


When the US DOT launched this “RED LIGHT” campaign several years ago, NHTSA headlined 2,800 deaths a year related to this problem. Well as with all its campaigns, the suspect word is always RELATED. In this case “related” is intersection accidents, yet only 25–30 percent of those where at signalized intersections, and there is no data on how many were caused by intentional running of red lights as opposed to normal expectation of intersection accidents. Nor do DOT press relases ever mention the fact that as many as half of these fatalities were caused because basic engineering practices (required for federal funds) were not being followed in the first place.

For those in San Diego, your signal timing isn’t in compliance with mandatory federal regulations and sound engineering practices – California’s official traffic manual isn’t in compliance either.(see middle section of this of this email for detailed evaluation) In fact, most of California’s signals are timed 10-15 miles below the federal engineering requirements. Worse yet, almost all of the Red Light Cameras are located at intersection in left hand turn lanes with very high volumes and very low accident rates. This is called “low accident rate intersection” when you factor in traffic volumes being served.


You ask, why are the majority of these cameras in San Diego targeted at high volume low accident rate intersections and left turn lanes? The clue is "HIGH VOLUME" equals greater opportunity when the yellow is truncated. I wonder, in this crisis that led up to the installation of these cameras in San Diego, how many fatalities occurred at signalized left turn lanes. How many of those were caused by intentional red light runners in which the violator was making the left turn (target of enforcement) – AND I don't mean a failure to yield at a signalized intersection where left turns are permitted when it is safe to do so. Millions of vehicles, untold numbers of signalized intersections and a handful of accidents at most, very low rates for the high volumes. What crisis was the city addressing. Show me the numbers!

The more you look into the numbers it gets very scary??

The city of Mesa added one second to yellow light duration and the number of citations dropped 56%. This article tells an interesting story, each intersection according to federal standards requires its own engineering study to determine the prevailing approach speed of traffic. When looking at these stories in is clear that the previous 3 second yellow was a city wide standard and if you have driven there this is way too short for the prevailing speeds. Moreover, adding a second doesn't mean they are now in compliance either.

“Red-light tickets drop 56 percent in Mesa”
“In November, the city issued 1,639 left-turn arrow citations at the six intersections patrolled by cameras. In December, the month after the change, the number fell to 716. In October, the month prior to the change, Mesa issued 2,645 citations.”

My math fails me at times... December is 716 citations after the change and October was 2,645 citations. That is a 73% percent drop, and if the signals were timed to reflect the actual prevailing speeds of traffic the reduction would be greater than 90%. The good news, in other reports the city is complaining the RED LIGHT PROGRAM is no longer PROFITABLE? Maybe there is hope.

What is the promise do these numbers mean, 20 cameras in one area of New York alone issued 1.4 million citations using 3 second yellows – like lemmings to pay their fines for signals violations NOT SET IN ACCORDANCE WITH FEDERAL SAFETY STANDARDS. Let’s expand it nationwide – what bureaucrat wouldn’t jump at the chance to join in on this new found cash cow. If New York had California’s fine structure, since 1995 when they were installed, these 20 cameras would have generated 387 million in gross fines. Accident rates there are still extremely troubling and it will not change until they address the underlying engineering deficiencies.

Studies here and abroad have documented that by setting the timing in accordance with known engineering practices and allowing appropriate clearance intervals - accident rates have been reduced up to 50% without cameras and cameras alone have not reduced accident rates. AAA study from Michigan If you like I can provide with like studies from Australia where wide spread camera use had no effect and the test area where the timing was adjusted it did reduce accident rates.

As you can clearly see, NHTSA promotes these programs using less than half–truths and the “S” word to support the fleecing of America. Red light cameras are much more profitable than speed traps, because there are no personnel to pay that in time retire, thereby requiring the city to increase fines so they can pay for the new and the old officers, not to mention its staff and its courts. Why did I mention speed traps, because posted speed limits are required to be established with engineering studies.

Politicians don’t like engineering studies because the primary consideration is the 85th percentile or measured prevailing speed of traffic. Prevailing speed is the 85th rounded up to next 5 mph increment. This is the public’s consensus as to what speeds they have found to be safe for a particular section of roadway. By setting all traffic control devices to warn and guide traffic using the prevailing speed as found, safety improves. Most California limits are set 5, 10, 15 miles below this threshold and more – I have clear evidence where this is common and the studies have been altered or not done at all to support unjustified limits. The reason they don’t like it, if you set speed limits by an engineering study limits are raised and you reduce the number of citations issued.

Intersection Safety is compromised by lower than required speed limits because the measured prevailing speed in a study becomes the basis for the timing of the lights, size of turn lanes, distance of intersection pre–warning devices etc. Californiašs limits are not set as required by federal law, but rather with conjecture and safety has been compromised.

In addition, there is a truncated yellow non-standard being used for red light cameras that assure violations at predictable levels. Truncated because the timing is set to reflect the posted limit and a non–approved timing standard recommended and used with by red light camera vendors. This timing is done without the required MUTCD recommended timing and clearance intervals that are to be based on the engineering studies measured prevailing approach speeds – used as the basis of the lights timing.

What are the consequences of shorter duration yellow lights and the eliminated sufficient intersection clearance times? Tens of thousands of citations being issued to citizens who just a few years ago were not running red lights until the city shortened the yellows. And now, you have pedestrians and motorists watching the signal instead of the traffic see the light change and move into the intersection prior to it being cleared by cross traffic – resulting in increased catastrophic accidents and general disregard for traffic control devices.

Why, there are MILLIONS of reasons that have dead presidents on them! All this from only a handful intersections? If there are that many violators, then sound engineering practices require a safety review ­ not a program where the city profits by its inaction and unsafe practices.



At your request, I did an initial cursory review of parts of Caltrans Chapter 9 as noted. My review is cursory and therefore my comments should not be construed as final in nature or 100% complete. My observations and comments based on what I see in Chapter 9 are:

1. Suggested yellow duration signal settings are shown in 9–04.5 entitled “Yellow Change Intervals”. I compared the suggested values for each speed against the comparable values that result from using the contemporary ITE “reduced yellow” calculation method as described in ITE Traffic Engineering Handbook (5th edition). The results from each source are the same for all practicable purposes (see attached chart).

2. The findings of #1 above make for a reasonable conclusion that the underlying parameters used in the ITE calculation method are used in the Caltrans calculation method as well. Accordingly, the figures shown in the Caltrans traffic manual for yellow time duration would be based on approaches to the intersection being level, a reaction time of one second, and a deceleration rate of 10 feet per second per second.

3. Speed – The manual states “The length of the yellow change interval is dependent upon the speed of approaching traffic”. In the chart itself it states “Approach Speed”. Other sections of the manual (for example, 9–02.2, item #6), identify the 85th percentile speed as the critical speed of approaching vehicles. Nowhere that I could find, does it state or imply that using the speed limit to establish yellow time durations is acceptable. From this, the manual does not provide for using simply the speed limit in determining yellow time duration. To establish the yellow time duration, one must know the “speed of approaching traffic” and the only way to know that is from an engineering study. Because many speed limits today are underposted, using the speed limit instead of the actual approach speeds to establish yellow time duration can have terrible consequences, for they shorten even more the yellow time provided by contemporary “reduced” yellow time calculation methods.

4. “The purpose of the yellow signal indication is to warn traffic approaching the signal that the related green movement is ending....” The MUTCD (1988) provides that “a steady circular yellow indication shall be given following a circular green indication in the same signal face”, hence, once the yellow indication appears, the green has already terminated. Giving warning of an event that has already occurred (ending of the green) is pointless, hence this element of the definition is pointless from a contemporary standpoint. I say contemporary as it may be that some jurisdictions have signals that display both green and yellow simultaneously (likely a holdover from older days). In all likelihood, such a condition was more likely years ago as opposed to now as years ago different color indication combinations were evident.

5. “The purpose of the yellow indication is to warn........that a red indication will be exhibited immediately thereafter........”. The word “immediately” is not a technically correct word. A more accurate word might be “soon”. Or better wording might be “is to warn that the onset of a red indication is imminent”.

6. “The purpose of the yellow indication is to warn that......traffic will be required to stop when the red signal is exhibited”. The connection between the yellow and the red indications as stated is irrelevant, especially when the duration of the yellow indication is inadequate. If someone were to insist that the yellow and red indications be linked in the sentence, a better wording would be “The purpose of the yellow indication is to warn that.....traffic is required to be stopped in advance of the intersection when the red signal is exhibited.” Such a wording would tend to place a less ambiguous obligation on the signal operator to provide sufficient yellow time duration to allow a driver to comply.

7. Number 6 above brings us to the “probability of stopping” issue and the related matter of the exclusionary nature of contemporary yellow time duration settings. Contemporary yellow time durations provide for “entries on red” because of the exclusionary reaction time used in the calculations and the unreasonable provision for “brake” time. Close scrutiny of the probability of stopping characteristics of drivers (see TRAFFIC CONTROL DEVICES HANDBOOK 1983 PAGE 4–29 and other research provided to you before) shows that, using the contemporary “reduced yellow time” calculation methods, there is not just a possibility of not stopping, there is a probability of not stopping built into the contemporary yellow time duration settings and, for large numbers of drivers that continue, it is impossible for those drivers to reach the intersection (without accelerating) before the light turns red. You may have seen over the last few years a tendency of drivers to accelerate toward and through intersections. My view is that “continuing” drivers are adapting to the impossibilities presented to them by the only means available, accelerating.

8. I see no indication in the manual of the need to increase yellow time durations for downhill conditions.

9. I see no indication in the manual that the “suggested” yellow times shown are minimums.

10. I see no indication/requirement in section 9–04.5 of the manual for measure of effectiveness tests for the yellow time duration settings after they are established. However, section 9–04.2 “Review of Traffic Signal operations” states that “All State highway traffic signals should be periodically reviewed for proper operation”. The point of a measure of effectiveness test for the yellow time duration is to confirm that it is sufficient to allow all drivers a reasonable chance to negotiate the intersection safely. My own view is that sufficient would mean, especially in the context of using red light cameras, enough yellow time to stop before the red indication shows, enough yellow time for continuing vehicles to “reach” before the red indication shows and enough yellow time for continuing vehicles to clear before the red indication shows. In my opinion, the yellow time duration should be sufficient to accommodate all three conditions.

11. Clearance Intervals – Section 9–04.6 says that red clearance intervals are not required. On the assumption that it also means that yellow clearance intervals are similarly not required as such are not indicated by section 9–04.5, the duration of the yellow indication becomes critical to avoiding dangerous conflicts. Not making provisions for continuing vehicles to clear the intersection before releasing conflicting vehicle and/or pedestrian movements (by displaying prematurely a green or “go” indication for them) can be disastrous for obvious reasons. In a sense, inadequate yellow time tends to be a form of intersection roulette whereby yellow time is established based on a “let’s hope for the best” mindset.

12. PHILOSOPHICAL - With the duration of the yellow solely determining how fast the lights turn red coupled with the yellow time durations being “suggested” amounts, how does one justify installing and using a red light camera? The point is the word “suggested” may very well indicate there is no legally binding standard for how fast the lights turn red at all, in which case it would seem impossible to accuse someone of “running a red light”.... no legal standard against which to make such an accusation or conversely, to mount a defense.

In the context of the contemporary “reduced yellow” signal timing methods, for the most part red light cameras are simply measuring the driver failures that are being designed nowadays into the signal timing settings.

In your discussions with others, ask the following questions:

1. Why was it necessary (for law enforcement purposes....i.e. red light camera use) to change the national engineering codes to make the lights turn red faster?

2. Why couldn’t a jurisdiction simply install and use a camera and leave the national engineering codes (and many state and local codes by extension) alone?

It would be interesting to see the historical elements of Caltrans’ Traffic Manual regarding this topic, particularly to identify changes in methods to establish how fast the lights turn red, because how fast the lights turn red (determined by the duration of the yellow indication) is the key to understanding this whole “red light camera enterprise”.

Red light camera enthusiasts and peddlers rely heavily on people incorrectly assuming that drivers nowadays have lost their collective minds and have decided to “run red lights” with reckless abandon. Contrary to that assumption, we know that what is different is that the lights are turning red as much as 30–40% faster than before, at least where the contemporary “red light camera” methods of establishing yellow time are being used or have been implemented. It is inconceivable that one would make the lights turn red faster given the increasing age of the driving population, the increasing complexity of today’s intersections and the increasingly diverse driving population.

I am sending a copy of this message to NMA as well as to others who have expressed an interest in this important safety matter.

Gene Quinn
Vienna, VA

PS When the yellow time is reduced per contemporary calculation methods, large numbers of statistically probable “continuing” vehicles cannot reach the intersection without accelerating before the light turns red. If cameras are measuring “entries on red”, the cameras are measuring the “red light light running” created by the change in yellow time duration calculation methods.

Note that the older yellow time calculation methods, in almost all cases/circumstances, would allow statistically probable “continuing” vehicles to reach the intersection without accelerating before the light turns red. Under the circumstance, I am not surprised about my own observations of drivers accelerating toward and through intersections on a fairly regular basis. You'll often hear complaints about this phenomena from the very same people who promote reduced yellow time along with red light cameras, some of whom are now promoting the idea of issuing speeding tickets using the red light cameras.

Australian Signal Timing Standards



In regards to your question.

“3. Notice in the Australian change interval methods sheet I sent yesterday that it shows V 85. I take it to mean the 85th percentile velocity (speed) but I have not confirmed that with certainty, although it would seem a reasonable interpretation.”

Engineering tenets are regardless of the posted limit, all traffic control devices are suppose to be set to warn and guide traffic traveling at the prevailing speed. The 85th percentile is the accepted international value for prevailing speed. That being said, the true prevailing speed is the 85th percentile of free flowing traffic raised to the 5 mph increment (new FHWA recommended posted limit). Once determined, the engineering study process is suppose to include actual site reviews to make sure there are no anomalies that may need additional adjustments. The object of the engineering study is to make sure the traffic control devices are doing their job to give adequate notice to safely guide traffic or warn motorists of unforeseen hazards; thereby, reducing accidents by reducing flow conflicts. Truth, rarely done, most traffic control in the nation set by unfounded political decree – in my opinion illegal under federal laws that are totally ignored and unknown to politicians and the judiciary alike..

The timing of the yellow and all red clearance is suppose to be set on the basis of an engineering study of the approach speeds. This definition is also fraught with problems because these findings (engineering studies) have also become heavily politicized and the methods used to obtain the documented prevailing speeds have been altered to obtain lower numbers. In my opinion the measured prevailing approach speed is just that, any lowering of the number is a political act and if the devices are set to reflect the lower number then on its face, the devices are no longer being set to guide and warn traffic traveling at the prevailing speed – a violation of a core engineering tenet that has shown to reduce accident rates.

In the last few weeks your emails have included reference to a 3 second yellow for all intersections in many of these jurisdictions. A clear indication to me that signal timing in these areas are not being set on the basis of engineering studies, but rather by local city wide design standard handbooks that treat all streets the same regardless – without engineering studies. This is the case in both Reno and Las Vegas – not legal under federal law.

In California there is a wrong assumption that because engineering studies are required by law, that posted limits are reasonable and likewise the timing of the lights. This simply isn’t the case and the posted limits rarely, read never, reflect the actual prevailing approach speeds of traffic.

California as most of you know has a speed trap statute that requires an engineering study to enforce a speed limit with radar. It requires periodic reviews, and that if a limit is posted more than 5 mph below the 85th percentile that the reason be documented and germane, and those hazards already apparent are not justification for a limit to lowered. In no case can it be set below the next 5 mph increment.

In the cases I have helped others on I have uncovered an unbelievably high incident rate of intentional deceit. I have seen limits set as much as 20 mph below the prevailing speed being radar enforced, with actual reversed engineered documents on file with the court by Caltrans showing the 85th only 1 mph over the posted limit. These same studies are what the timing of the signals where based on. A short list of just some of what I have found;

1. State DOT traffic engineers being told not to report their findings, thereby not creating a public record that could be challenged when speed where found to be high. (at the request of a State Senator to the Local Caltrans district head)

2. Studies being done on the heaviest traffic days of the year when the speeds are known to be there lowest

3. All traffic reports and not the free-flowing that is required by law. Very very common.

4. Many cities and the state contract out their engineering studies to small contractors. In the contractors’ instructions, not in the actual engineering studies on file, they were instructed not to count too many fast vehicles and before they file their reports check with local officials, and if necessary to redo the studies if they appear to be too high.

5. The state altering its Internet traffic reporting to show traffic speeds up 15 mph and more below the speeds that were being recorded when they were posting the raw data.

6. Speeds being measured at flow choke points and curves, conspicuous by its absences not a single strait section of the 15 mile section was measured. Again in violation of the state traffic manual and of course it was certified by the district engineer. A posted limit 20 mph below the prevailing speed being enforced by the CHP and local Sheriff's department.

7. The state manual virtually requires measurements to be done with radar when it has been known for over 25 years that this will bias down, by several miles per hour, measured speeds. In most cases this reduces the posted limit by one 5 mph increment.

The moral of this story, even if there is an engineering study don't accept it on its face, in particular if the resulting posted limit numeric is the basis of the timing of the light.

On its face a posted limit set in accordance with California law, done correctly, can be as much as 10 mph below the FHWA 85th percentile raised to the next 5 mph increment and 15 mph below the FHWA safe speed risk chart. My experience tells me that it is more like 15–20 below and more if a study was done at all. Which would make a significant difference in the duration of the yellow timing.

Chad Dornsife