Speed Cameras - technical query

[RacerMDR]

My profession is testing. As such, I assume there are defects present in all machines and systems.

What occurs to me is that tax cameras may not be accurate, and if they are, they aren't accurate all of the time.

My question is really, who is responsible for signing off the accuracy of such equipment? Do we trust them?

I think fixed tax machines are probably more stable (Gatsos, Truvelo etc)

However, I can't believe that every time someone is 'done' for speeding with a camera that its accurate.

Average cameras for example, at what point do they read the numberplate? As you approach the M3 from the M25 (anticlock) there is a camera looking up the road and yet the 50mph sign is under the camera. Therefore if I approach that camera at 70mph, then get down to 50mph at the sign (exactly with a rapid decrease of 20mph) surely then my average speed would be reading high even though I wasn't breaking the limit at the time the camera read the plate. I know its read over two points - but surely the distance between those two points would be relevant in this scenario.

I'm sure I could think of many other scenarios where these things could be inaccurate.

If drivers are blindly accepting their punishment, they may in many cases be unfairly taxed.

However, the nature of the law assuming you are guilty and therefore giving you a discount if you pay up quickly is outrageous as I would want to see documented evidence of the testing before I accepted guilt. What would happen if a person requested this?

In my day to day life, the amount of evidence I need to produce is staggering.......why not for these tax cameras?

They should be tested for each site, and every time they are moved they need to be tested again. I want to see the signoffs!

and no, I haven't been done for speeding!! (well not this week)

any thoughts?

[captainzep]

There's a programme of calibration I'd guess?

Also, there should be a visual check of the evidence in each case to ensure the prosecution is safe?

Clearly both of these checks don't happen all the time though...

[jimmyb]

This is a good point. I have always thought that even fixed camera readings will never give an accurate measure as there is road vibration and weather aka wind involved menaing lots of very minute adjustments of position so that it will rarely/never give a prefect reading.

[RacerMDR]

agreed.

If i was to design a test plan for these devices, to enable me to professionally sign them off, I'm pretty sure that we would have noticed the tests as roads would need to be closed to allow them to be tested properly.

This is why, i'm certain they are not being tested properly. (on each and every occasion that they are installed or moved)

Experience tells me that this means there will be defects.

I think most people have heard that the changing lanes thing may confuse them. I think this is probably more than that, I think changing lanes means that you are covering a different distance, further.......which for average speed cameras will affect the results.

The tax camera partnership people will probably argue that they have tolerances built in, so that they don't issue fines for 51 in a 50 - but they set them to 55 or maybe 58 therefore covering off a number of the scenarios. I don't believe this will cover off all scenarios.

Edited by RacerMDR on Friday 28th November 16:38

[RacerMDR]

PS - should this be in SP and the LAW?

I originally put it in P^P as didn't think it was specifically law related?

[GFO875]

On the Specs average speed cameras, there is a small yellow line on the road (when they are first put in ) which is the point where the reading takes place. I know that parallax error will come into play where the plates are at different heights but that is a miniscule percentage of the total distance of the reading. Speeding offences are committed by driving in excess of the posted limit rather than precisely being for what your speed actually is. e.g. you are not convicted for doing 96mph, the offence is exceeding 70mph. People have been aquitted when errors are shown in the measuring device but I think folks are getting hung up about using possible errors of 1 mph to evade prosecution when there speed is well in excess of the limit; and they know it.
The speed reading is given as evidence of the offence.
Gets bullet-proof vest from rack and waits for a flaming......

Best wishes
Dave

Edited by GFO875 on Friday 28th November 16:51

[RacerMDR]

i actually wasn't coming from the perspective of getting off a speeding ticket. Although i admit i despise the bloody machines.

I was thinking more from a professional viewpoint that they shouldn't be live without appropriate testing.

you are correct though - they aren't going to be 10mph out (probably) - so it probably doesn't require that level of testing.

[jimmyb]

I fail to see how a bullet proof vest is going to protect you from a flame thrower

Edited for p1ss poor spelling....

Edited by jimmyb on Friday 28th November 18:36

[bryan35]

what gets me is 'what is an acceptable level of false prosecution?'

When you're prosecuting millions of people a year for speeding there will be some that are on false evidence, even if it isn't proveable.

Blind faith in machines is a real problem these days, or rather blind faith in the salemen who market the things as being 'digital' 'computerised' etc.

is 1000 prosecuted innocent motorists a year acceptable? Maybe it is.

[milessimpson]

RacerMDR said:

My profession is testing. As such, I assume there are defects present in all machines and systems.

You may be correct in what you say however the 'defects' can be moderated, when they are known, to have no adverse effect. Let's say that the accuracy is +/-2mph; the camera would not be used in a situation where it is likely that it would need to work within that tolerance.

RacerMDR said:

What occurs to me is that tax cameras may not be accurate, and if they are, they aren't accurate all of the time.

Moderation is again applied because they do not work on one calculation or measurement to produce their output; there is always 2 independent measurements so one is not used in error.

RacerMDR said:

My question is really, who is responsible for signing off the accuracy of such equipment? Do we trust them?

The Home Secretary has appointed ministers and a team of civil servants, lawyers and scientists and engineers to do this. The way to assure integrity is to have a whole host of individuals rather than one as it would be very difficult to corrupt a team even if it is as described.

RacerMDR said:

I think fixed tax machines are probably more stable (Gatsos, Truvelo etc)

However, I can't believe that every time someone is 'done' for speeding with a camera that its accurate.

Well you haven't displayed any reason why your view is of any value as you have no idea how the assurance is attempted.

RacerMDR said:

Average cameras for example, at what point do they read the numberplate? As you approach the M3 from the M25 (anticlock) there is a camera looking up the road and yet the 50mph sign is under the camera. Therefore if I approach that camera at 70mph, then get down to 50mph at the sign (exactly with a rapid decrease of 20mph) surely then my average speed would be reading high even though I wasn't breaking the limit at the time the camera read the plate. I know its read over two points - but surely the distance between those two points would be relevant in this scenario.

If you get down to 50 mph at the sign and maintain a speed that is 50 mph or less, or even for short bursts more than 50 mph you will not be detected as mainataining a speed that is averaging above 50 mph. The average camera systems adjust for position on the detected image and always have the minimum distance a vehicle will cover between the cameras so there is no chance of them over reading your average speed. Pretty simple concept and one that is not too demanding on the imagination especially if you have any form of engineering training.

RacerMDR said:

I'm sure I could think of many other scenarios where these things could be inaccurate.

...but would they be of any value?

RacerMDR said:

If drivers are blindly accepting their punishment, they may in many cases be unfairly taxed.

However, the nature of the law assuming you are guilty and therefore giving you a discount if you pay up quickly is outrageous as I would want to see documented evidence of the testing before I accepted guilt. What would happen if a person requested this?

I don't think your viewing the documented testing would be of any benefit

RacerMDR said:

In my day to day life, the amount of evidence I need to produce is staggering.......why not for these tax cameras?

Already done

RacerMDR said:

They should be tested for each site, and every time they are moved they need to be tested again. I want to see the signoffs!

They are but I don't see what you would be able to do with the sign-offs with your limited understanding.

RacerMDR said:

and no, I haven't been done for speeding!! (well not this week)

Well done.

RacerMDR said:

any thoughts?

Another technician with limited ability willing to comment upon what he knows little about.

[Mill Wheel]

RacerMDR - I Told You So!