[Rumours] Next Scandal...Engine Revs

[zac510]

Originally Posted by JamesH

Hmm. I think you have to be careful with using sounds transmitted from the car - the video is always compressed, and I thing the audio would be as well. Lossy compression (where information is thrown away to reduce size e.g. mp3) can lead to artifacting in the replay, which could make any measurements rather dubious. Depends on the amount of compression.

Here is some info that will interest you:
http://www.f1technical.net/forum/vie...t=5307&p=70721
http://www.f1technical.net/forum/viewtopic.php?t=2218

[Mr V]

Originally Posted by jab

The original source which said it was the Mercedes engines overrevving was Autosprint magazine

Can't have been Heikki, he didn't go fast enough in the race

[Rgms320]

Originally Posted by JamesH

Hmm. I think you have to be careful with using sounds transmitted from the car - the video is always compressed, and I thing the audio would be as well.

Not to mention Doppler shift. I think there have been attempts in the past to glean engine info from analysis of the exhaust note, but I don't think any have ever amounted to anything. I wouldn't expect this one to go anyplace, either.

When I found that article, and saw that Glen Crompton wrote it, I expected it to be a humorous piece. Granted, it wasn't very funny, but I didn't take it seriously, and won't do so until I see this reported by other sources.

To refresh your memories, here is a sample of the stuff Crompton used to write, back when he was writing in the "Rob and Crompo" partnership. It was published on the long-defunct Alternate Championship Headquarters site following the 2002 Silverstone race.

Unlike the Pitpass article, this one actually is funny.

Tom

[Rgms320]

Originally Posted by zac510

Here is some info that will interest you:
http://www.f1technical.net/forum/vie...t=5307&p=70721
http://www.f1technical.net/forum/viewtopic.php?t=2218

Well, the audio from the onboard camera would solve the Doppler problem, at least. If the director would stay on somebody's onboard for a whole lot of a lap, I suppose you might be able to figure something.

For results good enough to make precise measurements of engine RPM, though, it would probably take some tricky filtering, signal processing, and frequency measurement.

If you want to experiment with your own measurement, you might try an audio editor such as Audacity or Goldwave. Try bandpass filtering around the exhaust note in question. (Lessee... 8 cylinder four stroke gives four "pops" per revolution. At 18000 RPM, that's 300 revs per second, times four pops. So try a bandpass filter centered around 1200 Hz.)

See what you get. Is the result good enough to feed to a frequency counter? Can you get a good, stable frequency measurement that doesn't bounce all over the place due to noise? If not, what further processing do you need to do to clean it up?

Hmmm.... maybe there's something to the audio analysis method, after all.

Or maybe there isn't.

Tom

[Marbot]

Originally Posted by VIVA GT

If something is done to 'fool' the management system into thinking that the engine is running slower than it really is, all of these other actions will be thrown completely out too, so even if the engine still ran, the power output would be so severely compromised for this to be a total waste of time.

For a while now it has been possible for drivers of diesel cars fitted with 'common rail' systems (just like F1 cars) to buy a 'tuning box' that fools the high pressure pump into thinking that it isn't providing enough fuel pressure.Consequently the fuel pump provides more pressure which increases the engines torque and bhp by up to typically 25% or more.All of that from what is basicly just a 'resistor' in the wiring with no modifications to the engines ECU!

So by 'corrupting' one sensor input you can affect engine performance quite drastically.

[VIVA GT]

Originally Posted by Marbot

For a while now it has been possible for drivers of diesel cars fitted with 'common rail' systems (just like F1 cars) to buy a 'tuning box' that fools the high pressure pump into thinking that it isn't providing enough fuel pressure.Consequently the fuel pump provides more pressure which increases the engines torque and bhp by up to typically 25% or more.All of that from what is basicly just a 'resistor' in the wiring with no modifications to the engines ECU!

So by 'corrupting' one sensor input you can affect engine performance quite drastically.

I appreciate that it's possible to 'fool' the management system of a diesel road car to give you more power. I just don't see (for the reasons I explained) how you could fool tghe ECU on an F1 car to enasble you to run higher RPM.

[Marbot]

Originally Posted by VIVA GT

I appreciate that it's possible to 'fool' the management system of a diesel road car to give you more power. I just don't see (for the reasons I explained) how you could fool tghe ECU on an F1 car to enasble you to run higher RPM.

Well,if you know which bit of wiring to put the resistor in.......

If it were possible to mess with the rev-limiter then I'm sure that beyond that limit the engine would run has it has run previously beyond that limit.

[zac510]

Originally Posted by Rgms320

Hmmm.... maybe there's something to the audio analysis method, after all.

Or maybe there isn't.

Tom

Well it proves results can be had, but whether they'd hold up in an FIA kangaroo court or not is another thing.

You would have to think that the FIA's own datalogging 'black box' is a more reliable source of the information.

[Rgms320]

Originally Posted by zac510

Well it proves results can be had, but whether they'd hold up in an FIA kangaroo court or not is another thing.

You would have to think that the FIA's own datalogging 'black box' is a more reliable source of the information.

Yes, exactly.

Putting on my geek hat and thinking it through, I can see lots of potential for error in the audio analysis method, particularly from a sound track recorded over a commercial broadcast link.

If I was going to try an analysis as described (...I won't...), even I wouldn't trust my results.

Considering the accuracy you'd require, I'd have to say that the audio analysis method is a non-starter.

Tom

[bravo]

I thought we worked out that the sound on the on-board shots was not true, with what sounded like strange short up/downshifts on straights?


http://tentenths.com/forum/showthread.php?t=114537

[jab]

But as we all know, whether there's any concrete evidence or not makes no difference to whether Max will punish them or not

[Marbot]

Originally Posted by jab

But as we all know, whether there's any concrete evidence or not makes no difference to whether Max will punish them or not

Finding evidence has never been a problem.Finding the right punishment has.

[duke_toaster]

They always need more of ze punishment!

Surely the best way of dealing with this rubbish would be to do what the SEAT Cupra Championship did (IIRC, I had the rulebook on a lappy that broke). The ECUs were handed back at the end of each race meeting, and next time round the driver would pick a number out and that's their new ECU. Failing that the ECU supplier could be given free reign to poke around to check no sensors have been "enhanced".

[Pingguest]

Originally Posted by JeremySmith

What is this now? Here we go again gate! I am fed up with this rubbish now...

Maybe its just a way to compensate for the lack of racing?

[VIVA GT]

Originally Posted by Marbot

Well,if you know which bit of wiring to put the resistor in.......

If it were possible to mess with the rev-limiter then I'm sure that beyond that limit the engine would run has it has run previously beyond that limit.

Unfortunately though dear Marbot there is no seperate rev limiter, it is all part of the (standard) Engine Control Unit. The sensor that tells this how fast the engine is turning is also used to calculate when the signal is sent for ignition and fuel injection. So, if you fooled the ECU into thinking the engine was only doing (say) 18,000 RPM, the ECU would only send ignition and fuel injector signals for this engine speed. Therefore it could not be running at (say) 20,000 RPM...

[Marbot]

Originally Posted by VIVA GT

Unfortunately though dear Marbot there is no seperate rev limiter, it is all part of the (standard) Engine Control Unit. The sensor that tells this how fast the engine is turning is also used to calculate when the signal is sent for ignition and fuel injection. So, if you fooled the ECU into thinking the engine was only doing (say) 18,000 RPM, the ECU would only send ignition and fuel injector signals for this engine speed. Therefore it could not be running at (say) 20,000 RPM...

So what's happening then! (if anything)

[RF_Racer]

quiet easy to fiddle around with the display multiplier settings and get it to record lower revs.. as most people do in club racing to pass through noise testing!!

[Road Train]

Originally Posted by Marbot

For a while now it has been possible for drivers of diesel cars fitted with 'common rail' systems (just like F1 cars) to buy a 'tuning box' that fools the high pressure pump into thinking that it isn't providing enough fuel pressure.Consequently the fuel pump provides more pressure which increases the engines torque and bhp by up to typically 25% or more.All of that from what is basicly just a 'resistor' in the wiring with no modifications to the engines ECU!

So by 'corrupting' one sensor input you can affect engine performance quite drastically.

The same 'tuning box' can also be used on diesel engines in trucks with great results (So I hear, as I would never use one myself ) and virtually undetectable during a download at your local service agent.
I doubt this story has any credibilty although I said that about the Piquet/Renault crashgate affair.
Maybe we're on the wrong track pointing the finger at Mercedes.
Maybe we should be looking at Renault (Just kidding)

[VIVA GT]

Originally Posted by Marbot

So what's happening then! (if anything)

Exactly Marbot. I just don't think it's possible (without tampering with the software), so I don't think anything is happening. (It's just a smoke screen...)

[Adam43]

Originally Posted by RF_Racer

quiet easy to fiddle around with the display multiplier settings and get it to record lower revs.. as most people do in club racing to pass through noise testing!!

I suspect that this is 'some' rather than 'most'. Unless I am in the minority...