A work of pure genius! - Brilliant "Revetec" Engine

[revetec]

Have you compared the X4v2 with a motor that your hoping to replace (light aircraft engine) ?

Yes. About 9% increase. Most people don't know enough about aircraft so benchmarking to automotive engines is more understandable to the average person.

The benchmarks we supplied is for an engine with mild aviation camshafts. We are in the process of modifying components to automotive type spec which will only increase the power/torque. Of course with modifications we can improve engine performance like any other engine. If you weren't so argumentative you would probably be chuffed with our initial figures using far less top end technology. It only gets better now.

Sorry , I assumed it was X4v2 in the Trike .


Quote #762
"Not much time was spent on the trike mods. Note that it is the X4v1 engine in the trike, not the X4v2."


Quote #728
"Firstly at the time we announced that we were putting it into a vehicle we hadn't received the trike and we were considering a car at the time of the announcement. So at the time the comment was accurate. GTM has now supplied us with a new trike, so we are now fitting it to that. Some things do change on a day to day basis and one of them is the vehicle we are first trialling the X4v2 in. This in no way means that we are not fitting it to a car after that."


It doesn't help potential Investors when your announcements are all over the place.

Cant you get your hands on a old (crop duster, plane) to trial X4v2 in.

For a start the plane wont fit in our workshop. A huge amount of trials on the road has to be completed to confidently put it in the air. Liability insurance will kill us at this stage. We cannot perform tests and mods as readily as on the dyno or trike. I'm not going on any further, it's just common sense.

The Rapptor seems to be familiar with parachutes , and by the look of what he does with a motor strapped to his body
he would probably test pilot it , (no adventure no gain) this would receive international media attention & possibly a small fine.
If this can be done , wouldn't it be worth it.

I don't think you need to strap a 2.4litre engine for powered parachutes, but he'll probably address this on further.

Quote #752
"It'll be about a month before I'm back on here, at which time I'll answer any questions then. Just go to our webpage until then.
cheers
Brad"

Are YOU still going to ?

Not if the questions are something you could logically answer yourself with a little bit of thought what is involved in developing an engine.

BTW: Didn't you notice by my posting on 11th that I came back, and also our testing report?

[revetec]

For the current X4 to be adapted to the above PPG buggy, their would need to be HUGE weight reduction as one of the best ultra light engines is a 313cc (@ cost AU$2`000.oo) & weighs in at about 19kg fully dressed.

2400cc/313cc = 7.667
Current weight (fully dressed) 130kg/7.667 = 16.9kg
Est Prod. weight (fully dressed) 105kg/7.667 = 13.7kg

Of course scaling down doesn't quite work that way, but it is probably achievable to get it close.
And the X4v2 is water cooled, air cooled is lighter by far.

Cheers

[The Raptor]

2400cc/313cc = 7.667
Current weight (fully dressed) 130kg/7.667 = 16.9kg
Est Prod. weight (fully dressed) 105kg/7.667 = 13.7kg

Of course scaling down doesn't quite work that way, but it is probably achievable to get it close.
And the X4v2 is water cooled, air cooled is lighter by far.

Cheers

Typically in the PPG industry, a water cooled engine would be a lot better, as the water jacket insulates the radiating noise from the engine.
Nothing worse that a loud whipi-snippier, so to speak !
Although, if the X4 is designed to run at low revs, this would also assist with noise pollution as the 313cc engine revs out to 6`250rpm.

Q: Is it possible for the X4 to be configured to run with a 2 stroke cycle ???
As I think I recall one of your early engines did this.

Thanks Brad, I think we are all awaiting the first contract to bitch slap the sceptics.

[The Raptor]

That is a totally unproven claim and it only refers to frictional losses (and sorry to burst your bubble but a normal bottom end is also 90% in the same sense). What is the translative efficiency? (how efficient is it at making torque from a given cylinder pressure). Also Revetec has not proved any increase in top end thermal efficiency.

If this is confidential, it should come out in good time ....be patient.

Another unproven claim and again current systems are only slightly worse (roughly 80% efficiency) The biggest advantage of CRSV system is the increased breathing efficiency.

I was ONLY trying to compare the INDIVIDUAL technologies by leaving out the worst part of each engine.
That is the "bottom end" of the CSRV engine & the "top end" of the Revetec engine! (I may have NOT made myself clear on that point ..sorry)

You forgot about a whole host of other things to consider. You obviously know absolutely nothing about ICE design so I suggest you stop contemplating it until after you take the neccessary classes.

No...

When I did my apprenticeship, you weren't even conceived grasshopper & their were no computer controlled engin management systems around then.
However, you are partly right, I have forgotten what you probably know or are currently reading from your student books.

But then, I have been semi retired from this industry for nearly 30 years now, that is my excuse.

Also I would like to mention that the recent rapport on the testing of the X4V2 engine is a sham. Not even close to objective, or even technically accurate! Really I am not impressed at all, I don't really expect much from revetec but this was really well under what I was expecting.

You are a real bitch to please, ....then scepticism is a form of intelligent interest & I don't think that is all that bad, young lad.

Q: What were you "Expecting" ?

[hightower99]

If this is confidential, it should come out in good time ....be patient.

That is assuming that it actually has any advantage

I was ONLY trying to compare the INDIVIDUAL technologies by leaving out the worst part of each engine.
That is the "bottom end" of the CSRV engine & the "top end" of the Revetec engine! (I may have NOT made myself clear on that point ..sorry)

Why? that is an excercise that is so pointless that I would advise you not to even think it privately in your own head (but you posted it here?)

When I did my apprenticeship, you weren't even conceived grasshopper & their were no computer controlled engin management systems around then.
However, you are partly right, I have forgotten what you probably know or are currently reading from your student books.

what did you apprentice as? hopefully something that had at least alittle to do with ICEs (do you know what ICE stands for?)

But then, I have been semi retired from this industry for nearly 30 years now, that is my excuse.

The rules of physics haven't changed in that time and the Laws and Constants are still the same yet you still would like to use your age as a cover for you lack of understanding?

You are a real bitch to please, ....then scepticism is a form of intelligent interest & I don't think that is all that bad, young lad.

Actually it wouldn't take much from revetec to please me at all. and yes rational scepticism is a good thing.

Q: What were you "Expecting" ?

Well first I was expecting the article to be free of incorrect techincal information. But what I got in the article is really astounding.

First they continue to claim that because the new X4V2 only uses 2 valves per cylinder with no variable adjustment (as opposed to 4 valves per cylinder and variable valve timing and variable intake length) that it's peak torque is held back? This is incorrect. 4 valves per cylinder helps high rpm breathing (which they don't need if they are running at a true 3000rpm) they do not increase peak torque. Also variable valve timing and variable intake length systems only increases the width of the torque peak (it does not increase the actual peak value). They obviously want people to think that they are held back by the top end tech when they are not.

On top of that when descirbing variable length intake port systems they incorrectly state that:

"A short port provides better responsiveness and torque at lower RPM ranges, and thus a longer intake port provides greater torque in the higher RPM ranges."

This is the opposite of what is true. SHORT ports are used for HIGH RPM and LONG ports are used for LOW RPM.

The article is filled with simple "rule of thumb" guesses and large delta values, with no math at all. They also assume that the people who read the article aren't intelligent enough to put 2 and 2 together. The claim to estimate an imcrease in torque in the lower RPM ranges of anywhere up to 30% (can you get any more vague and meaningless than that?) but then they show a chart that compares the torque of their engine against a toyota 2.4L engine and a scaled representation of a GM 5.7L engine. The Toyota is only beat up to roughly 2000RPM by less 4 procent, and on top of that the toyota engine goes on to produce a peak torque of 220Nm (a whopping 9% increase!). They mention that the toyota engine has VVT, VLIP, and 4VPC, but again that didn't increase the peak torque it only gave the toyota a broader torque curve that is still excellent for it's size even at the low RPM shown.

They then go on to say how good they did against a "GM 5.7L engine" which is the LS1. Well the interesting thing is that the valve train system in the LS1 was optimised for higher RPM and because it isn't variable it suffers badly at low RPM. That explains why the revetec engine produces more than what 2.4L of LS1 can. So basically they have failed at producing any great increase in torque compared to convectional technology.

Also they mention that this was all done witha piston stroke of only 65mm but how does that matter in a revetec engine when the lever arm is based on the crank cam radius and not the piston stroke? they are again trying to attract the unitelligent.

They have a horrible section with a graph of the A/F ratios used during the dyno run, but they go on to say that the graph is ony a guide and that basically their engine is capable of fooling a high speed Lambda sensor (which isn't physically possible) because the graph conflicts with earlier (unpublished) testing.

They then go on to give a completely misunderstood idea on the concept of BMEP. BMEP has very little to do with thermal efficiency and design optimization. The true value they should be disclosing that allows you to compare it to any other engine is BSFC not BMEP. What they fail to mention is that BMEP is a fictional, theoretical value. BMEP is only good when talking about the same engine.

They also show a raw dyno chart which isn't corrected to SAE regulations, which means it looses some of it's value. The graph should be considered a single, unique, performance with very little value. It does show that the engine is capable when the right conditions are present though.

The conclusion at the end is therefore of little real value as it is not the case.

[revetec]

On top of that when descirbing variable length intake port systems they incorrectly state that: This is the opposite of what is true. SHORT ports are used for HIGH RPM and LONG ports are used for LOW RPM.

I made an error in my report and have amended it to remove the reference to port length effects. Thank you for your input.

They then go on to say how good they did against a "GM 5.7L engine" which is the LS1. Well the interesting thing is that the valve train system in the LS1 was optimised for higher RPM and because it isn't variable it suffers badly at low RPM. That explains why the revetec engine produces more than what 2.4L of LS1 can. So basically they have failed at producing any great increase in torque compared to convectional technology.

The top end of the LS1 is similar to the X4v2 and the gains are consistent through the tested range. If you are correct, the gap should decrease significantly as the revs increase. So your point is?

Also they mention that this was all done witha piston stroke of only 65mm but how does that matter in a revetec engine when the lever arm is based on the crank cam radius and not the piston stroke? they are again trying to attract the unitelligent.

It does matter when it comes to thermal loss. A shorter stroke engine dissipates less heat into the cylinder wall. But a conventional engine with a short stroke produces less torque than is practical in automotive use.

They have a horrible section with a graph of the A/F ratios used during the dyno run, but they go on to say that the graph is ony a guide and that basically their engine is capable of fooling a high speed Lambda sensor (which isn't physically possible) because the graph conflicts with earlier (unpublished) testing.

The Lambda sensor measures Oxygen not fuel. If the desired combustion is achieved there is a trace of Oxygen in the exhaust gas. If the combustion is on the rich side, oxygen is reduced, and if the mixture is too lean a greater amount of oxygen is measured. The Lambda readings are a guide because we have proven independently that running a leaner mixture in the 2,000-4,000rpm range provided a normal lambda reading. This means that the desired combustion was achieved using a leaner mixture. Either that or we had a 50% restriction in the intake at 4,700rpm. Torque was good so I think there was little restriction.

They then go on to give a completely misunderstood idea on the concept of BMEP. BMEP has very little to do with thermal efficiency and design optimization. The true value they should be disclosing that allows you to compare it to any other engine is BSFC not BMEP. What they fail to mention is that BMEP is a fictional, theoretical value. BMEP is only good when talking about the same engine.

Please go to Autospeed where they explain BMEP. BTW. We have used a Kesler BMEP plotter that has a sensor in the combustion chamber measuring and plotting pressure, so BMEP is not fictional. It is the average cylinder pressure over 4 complete cycles.
You can work out a theoretical BMEP pressure with the following formula which is close to the real thing:

BMEP =

kW x 1200
----------
litres x rpm

They also show a raw dyno chart which isn't corrected to SAE regulations, which means it looses some of it's value. The graph should be considered a single, unique, performance with very little value. It does show that the engine is capable when the right conditions are present though.

The conclusion at the end is therefore of little real value as it is not the case.

In the right conditions? The dyno we have corrects to standards using air temp and barometric pressure adjustments.
For your Info the test conditions are as follows: Room temp (26.7 degC) Humidity (60%) Barometric Pressure (1010 mbar) Intake Temp (26.5 degC)

With all your arguments in mind, go to TopSpeed.com and read the article on our engine.

Look if I published everything and all data and calculations, it would take me a year to tabulate it. This report was aimed as an overview of recent testing. If we used the same technology as current engines and had a budget to optimise all the systems, then the result would be even better. We are currently making further upgrades and will better these figures soon.

It is quite easy to sit and criticise anyone, it quite different to do the job. Have you done any actual published development on engines?

[hightower99]

I made an error in my report and have amended it to remove the reference to port length effects. Thank you for your input.

Well that makes me happy. I thought you might have disagreed.

The top end of the LS1 is similar to the X4v2 and the gains are consistent through the tested range. If you are correct, the gap should decrease significantly as the revs increase. So your point is?

The only shared things between the heads of your engine and a LS1 is the fact that they are both pushrod operated 2VPC heads. Other than that I would wager that unless you based the total head design on the LS1 (which would be a bad move on your part) that the designs are not overly similar. First the tested range is only from 1250RPM to 3000RPM (Only a 1750RPM spread). The LS1 valvetrain is optimised at something around 4500-5000RPM and therefore is actually causing a loss in performance everywhere else in the RPM range. Furthermore I don't even know how you got torque ratings for such low RPM from a LS1 the majority of dyno charts I have seen start above 2200RPM. The gains are not consistent at all. As revs increase the gain is decreasing until about 2600RPM where your engine closes in on it's optimized RPM while an LS1 is still more than 2000RPM away from it's optimized RPM. It is plain to see that the torque curve of the LS1 is steadily increasing as revs rise.

It does matter when it comes to thermal loss. A shorter stroke engine dissipates less heat into the cylinder wall. But a conventional engine with a short stroke produces less torque than is practical in automotive use.

Yes but that aspect was not mentioned or even implied it is plain to see that in the context of the article you are referring to the fact that you are producing almost as much torque with a shorter piston stroke as if a shorter piston stroke should be considered a disadvantage (like it would be in a common crank system). This is mainly used as an excuse for the fact that the Toyota engine produces more torque for more than half of the shown range even though you claim an increase "anywhere up to 30%". You are making excuses for the apparent lack of any advantage.

The Lambda sensor measures Oxygen not fuel. If the desired combustion is achieved there is a trace of Oxygen in the exhaust gas. If the combustion is on the rich side, oxygen is reduced, and if the mixture is too lean a greater amount of oxygen is measured. The Lambda readings are a guide because we have proven independently that running a leaner mixture in the 2,000-4,000rpm range provided a normal lambda reading. This means that the desired combustion was achieved using a leaner mixture. Either that or we had a 50% restriction in the intake at 4,700rpm. Torque was good so I think there was little restriction.

Firstly I know exactly how a Lambda sensor works so lets assume it is a known for both of us. You cannot cheat or fool a Lambda sensor without doing so on purpose (by injecting excess air into the exhaust of by burning excess fuel in the exhaust). You apparently are forgetting that you fuel metering sensor system could also be defect. Explain to me how your engine can burn less fuel but still fool/trick the Lambda sensor into thinking that you are running 14.7:1? If you are burning less fuel then more oxygen will be present in the exhaust and the sensor will detect that and tell you that you are running lean. How are you fooling it?

Please go to Autospeed where they explain BMEP. BTW. We have used a Kesler BMEP plotter that has a sensor in the combustion chamber measuring and plotting pressure, so BMEP is not fictional. It is the average cylinder pressure over 4 complete cycles.

Unfortunately Autospeed doesn't do a good job of describing what BMEP actually is. They do however make one simple point: BMEP is a better value to compare different engines than Specifc Power. However BMEP has very little to do with efficiency. Think about it some of the most inefficient engines in the world produce extremely high BMEP figures while some of the most efficienct engines produce relatively low BMEP figures. However BMEP is even better suited for talking about the same engine. Because for the same displacement engine more torque requires a higher BMEP.

I suggest you read this: MEP

Notice where it says "It’s important to remember that the values produced by the formula are for theoretical analysis only, and do not reflect the actual pressures inside an individual combustion chamber."

In the right conditions? The dyno we have corrects to standards using air temp and barometric pressure adjustments.
For your Info the test conditions are as follows: Room temp (26.7 degC) Humidity (60%) Barometric Pressure (1010 mbar) Intake Temp (26.5 degC)

That is all important information that should have been in the article! The article says that the Dyno chart is raw output with no SAE smoothing which leads me to think that it isn't corrected to SAE standards for ambient conditions.

Look if I published everything and all data and calculations, it would take me a year to tabulate it. This report was aimed as an overview of recent testing. If we used the same technology as current engines and had a budget to optimise all the systems, then the result would be even better. We are currently making further upgrades and will better these figures soon.

the article is referred to on your site as a "Comprehensive testing report" it comes no where near comprehensive. I am not asking for a full on totally complete technical analysis of all systems and aspects, but I am certainly asking for actual data as opposed to your own interpretations of the data.

[revetec]

Firstly I know exactly how a Lambda sensor works so lets assume it is a known for both of us. You cannot cheat or fool a Lambda sensor without doing so on purpose (by injecting excess air into the exhaust of by burning excess fuel in the exhaust). You apparently are forgetting that you fuel metering sensor system could also be defect. Explain to me how your engine can burn less fuel but still fool/trick the Lambda sensor into thinking that you are running 14.7:1? If you are burning less fuel then more oxygen will be present in the exhaust and the sensor will detect that and tell you that you are running lean. How are you fooling it?

To explain: At the normal ignition point our piston is twice as high in the cylinder than a conventional engine. This means that the fuel molecules are closer together. We learn't on our first injected engine that if we use a normal air fuel ratio at the lower RPM ranges then provide a fuel map that is consistent with a conventional engine that our engine runs very rich, especially in the higher RPM ranges when the ignition timing is higher and the effect on changed piston position is at its greatest. We then leaned off the mixture (in some tests to 45% less at around 4,000rpm). At this point the engine fired cleanly, the plugs were at the right colour, and the Lambda sensor was reading a normal type reading. We then took the 1.35litre engine to India for testing in their facility. One of the tests was fuel consumption, and it was found that the engine in that test was using roughly the same fuel quantity per hour at 4,000rpm (full throttle) as it was at 2,000rpm (full throttle) This related directly to our fuel map. Throughout the test the Lambda sensor was reading normally. We have found that when the mixture is compressed to a denser state at the point of ignition, that the fuel mixture requires to be leaner to provide the same level of combustion. It actually wasn't expected that the Lambda sensor would read in the way it does.

I'm not trying to run a theory by you. We didn't expect it and we experienced this trait during testing.

I have noticed that Toyota have a development engine with two injectors, one in the port that injects a lean mixture and one in the head to provide a richer mixture for ignition. From our testing, I would assume that they experience the same trait to some degree from their Lambda readings. Is there anyone who knows about this? I'm very interested.

Notice where it says "It’s important to remember that the values produced by the formula are for theoretical analysis only, and do not reflect the actual pressures inside an individual combustion chamber."

Did you note that I said we have a kesler BMEP tester with a sensor in the cylinder head. The figures work out about the same, equipment or theory.

That is all important information that should have been in the article! The article says that the Dyno chart is raw output with no SAE smoothing which leads me to think that it isn't corrected to SAE standards for ambient conditions.

The release was a cut down version of the one sent to the automotive companies we are talking to, with a little added for those who don't understand as well as you do. Well I posted the data for you, all you had to do was ask instead of getting all uptight about it. The test environment was a little higher in temperature than a normal test but our dyno compensates for it.

the article is referred to on your site as a "Comprehensive testing report" it comes no where near comprehensive. I am not asking for a full on totally complete technical analysis of all systems and aspects, but I am certainly asking for actual data as opposed to your own interpretations of the data.

Sorry it was not comprehensive enough for you, but it is more information than normally released by any manufacturer. It's a hard balance to reach.

[hightower99]

To explain: At the normal ignition point our piston is twice as high in the cylinder than a conventional engine. This means that the fuel molecules are closer together.

What exactly does that mean? I assume it means that when you fire the spark plug that the piston is closer to TDC, right? But what exactly does "twice as high" mean?

We learn't on our first injected engine that if we use a normal air fuel ratio at the lower RPM ranges then provide a fuel map that is consistent with a conventional engine that our engine runs very rich, especially in the higher RPM ranges when the ignition timing is higher and the effect on changed piston position is at its greatest.

To help me understand you are going to have to be alittle more specific. Does "normal air fuel ratio" mean 14.7:1? (i.e Lambda 1.00). Also I am confused again if you are feeding the engine so that a 14.7:1 mixture is made then it can't run rich... Rich means that you put in more fuel then the can be burned with the air that is ingested. There is a difference between homogenous rich mixtures and local rich formation. I assume you are referring to the later.

We then leaned off the mixture (in some tests to 45% less at around 4,000rpm). At this point the engine fired cleanly, the plugs were at the right colour, and the Lambda sensor was reading a normal type reading.

Again be more specific. is Lambda 1.00 a "normal type" reading? I think you are being confused by local rich formation.

We then took the 1.35litre engine to India for testing in their facility. One of the tests was fuel consumption, and it was found that the engine in that test was using roughly the same fuel quantity per hour at 4,000rpm (full throttle) as it was at 2,000rpm (full throttle) This related directly to our fuel map. Throughout the test the Lambda sensor was reading normally.

Ok lets think about this for a moment. If you were running at 14.7:1 (Lambda 1.00) at 2000RPM at full throttle, then in order to only ingest that same amount of fuel per hour at 4000RPM you have to run 29.4:1 (Lambda 2.00) under full throttle conditions. You cannot fool a Lambda sensor to read Lambda 1.00 when you are feeding it with Lambda 2.00, that is just not physically possible. Also if that related directly to your fuel map then you are saying that the fuel map shows that it decreases the amount of fuel injected (The fuel map in the article did not show that). Also what about power at 2000RPM and 4000RPM in that test?

We have found that when the mixture is compressed to a denser state at the point of ignition, that the fuel mixture requires to be leaner to provide the same level of combustion. It actually wasn't expected that the Lambda sensor would read in the way it does.

I think what is happening is that for some reason your engine seems to be good at concentrating the fuel in a small area of the combustion chamber (focussing it around the spark plug I would assume) meaning you are running Heterogenous mixtures. This allows you to be able to ignite much leaner (homogenous rating) mixtures because you are effectively making the mixture directly around the spark plug rich. However this will still not fool a Lambda sensor. You can't get the same amount of fuel to use more oxygen then it chemically can.

I'm not trying to run a theory by you. We didn't expect it and we experienced this trait during testing.

Well until you find the cause and confirm a theory then you really are just guessing and assuming that your equipment hasn't been fooling you.

I have noticed that Toyota have a development engine with two injectors, one in the port that injects a lean mixture and one in the head to provide a richer mixture for ignition. From our testing, I would assume that they experience the same trait to some degree from their Lambda readings.

I assure you that Toyota engineers would not except faulty readings in their Lambda readouts. If the lambda readings don't match what you are feeding the engine with then something is very wrong.

Did you note that I said we have a kesler BMEP tester with a sensor in the cylinder head. The figures work out about the same, equipment or theory.

You mean you have a cylinder pressure sensor. MEP is basically meant to mean the pressure that would have to constantly be exerted on the piston (during down strokes) to make the same torque as the engine actually makes. It is not the average cylinder pressure. It isn't even a good indicater of peak pressure.

The release was a cut down version of the one sent to the automotive companies we are talking to, with a little added for those who don't understand as well as you do. Well I posted the data for you, all you had to do was ask instead of getting all uptight about it. The test environment was a little higher in temperature than a normal test but our dyno compensates for it.

I wish I could get a copy of the full report then? glad to hear that your dyno compensated (that should have been mentioned in the article instead of referring to it as raw info).

Any chance I could read the full report?

Sorry it was not comprehensive enough for you, but it is more information than normally released by any manufacturer. It's a hard balance to reach.

My point is you shouldn't dumb things down at all. Everyone you want to attract will be able to understand the majority of information you can put out.

[Matra et Alpine]

ht, WHY are you asking for details that requires Brad to release info that would clearly be sensitive.

Also, the disparity between theory and practice doesn't mean the practice is bad. It CAN mean the theory is flawed CCEs take engines into areas few have ventured ! If you can take advantage of the practical knowledge someone else can do the theory after the success MANY inventions are built and not fully understood till later. Engineering history study will help explain that for you and give insight into wht TOyota's experience as reported may be totally valid.

You also are mis-reading "fooling Lambda" imho.
Sensors use a component to measure and produce a value which in "normal" combustion engines aligns closely with actual air/fuel burn. If there is something different happening then it's method of measrement may be invalid ! It's like using a colour film in your camera instead of b&w and then realising that you can't determien contrast the same way you did before !!

Brad, continue with your business and please don't go roudn the path previously trod of trying to appease analyticals who can't accept the gaps or fill them in with their own experience and intellect.

[hightower99]

ht, WHY are you asking for details that requires Brad to release info that would clearly be sensitive.

Because apparently if i ask I shall recieve... and actually I didn't ask for anything that would be sensitive. I asked what his definition is of the general terms he used.

Also, the disparity between theory and practice doesn't mean the practice is bad. It CAN mean the theory is flawed CCEs take engines into areas few have ventured ! If you can take advantage of the practical knowledge someone else can do the theory after the success MANY inventions are built and not fully understood till later. Engineering history study will help explain that for you and give insight into wht TOyota's experience as reported may be totally valid.

Relax matra it is still just a piston moving up and down inside a cylinder. And usually when you find disparity between theory and a single practice you normally look for flaws or faults in the practice before you throw the theory out. So far revetec are the only people who don't think something is wrong when there lambda readings don't match what they are feeding the engine with.

You also are mis-reading "fooling Lambda" imho.
Sensors use a component to measure and produce a value which in "normal" combustion engines aligns closely with actual air/fuel burn. If there is something different happening then it's method of measrement may be invalid !

Lambda sensors measure the oxygen content in the exhaust. Since we know exactly how much oxygen it takes to burn any given amount of fuel completely and since we can measure pretty accurately how much oxygen enters the engine and because we control how much fuel goes in then the Lambda readings can not be fooled. The only way the revetec engine could fool the lambda sensor is if they are somehow getting the same amount of fuel to burn more than completely and use more oxygen or the engine somehow absorbs oxygen itself? (maybe due to the use of unobtanium??? ) those are the only ways that you can get a lambda 1.00 reading even though you are feeding the engine at lambda 2.00 (without overly obvious things like faulty sensors ect.)

[Matra et Alpine]

Because apparently if i ask I shall recieve... and actually I didn't ask for anything that would be sensitive. I asked what his definition is of the general terms he used.

Your words ..."Any chance I could read the full report?" From that I thought you meant all th details, all the facts, nothing sanitised.

Relax matra it is still just a piston moving up and down inside a cylinder. And usually when you find disparity between theory and a single practice you normally look for flaws or faults in the practice before you throw the theory out. So far revetec are the only people who don't think something is wrong when there lambda readings don't match what they are feeding the engine with.

erm, Brad pointed out Toyota have been seeing somthing similar ?
The theories built up over the years are based on "normal" piston movement.
THis has differences.
Even small ones can make a difference in our theories -- butterfly effect.
Nobody suggested THROING THE THEORY out, very seldom does a theorem get discarded, usually it is enhanced as chamber compbustion HAS done for decades. Small changes, especially in the area of flame-front burn.

Lambda sensors measure the oxygen content in the exhaust

That's OK, that's a classic measurement mistake
WHat Lambda sensors ACTUALLY measure is the voltage at the platiunum electrodes with zirconium to react with the exhaust gas.
THe output is a voltage. How a voltage is decided to be a certain O2 content is based on PRACTICAL measurements of the design during the development process. The o2 is not actually being measured. A chemical reaction involving oxygen is. There can be a difference
How often that voltage is sampled to determine current value can give significantly different answers. How it's "smoothed" in the closed loop control system can then be "confused".
Lambda sensors are also sensitive to temperature and so if the discharged exhaust differs in temperature, swirl, density or a hundred other things it can lead to the sensor getting a "different answer".
In producing more torque at lower revs and lower gas flow then it's entirely feasible that the Lmbda rule for which violtage equates to how much oxygen is incalid in those cases.
There are many areas where the model in the monitoring system has multipolle variables any one of which may be the significant difference causing the difference in readings.
Brad continuing his tests and confirming results longer term and on multiple test benches is much more useful than assuming it's wrong and digging too deep AND assuming the theory is "wrong" and not jsut operating in an area it's not developed for.
Measurement systems often encounter issues like this - trust me I've patented and delivered enough

Since we know exactly how much oxygen it takes to burn any given amount of fuel completely and since we can measure pretty accurately how much oxygen enters the engine and because we control how much fuel goes in then the Lambda readings can not be fooled. The only way the revetec engine could fool the lambda sensor

What Brad said was the Lambda sensor was giving amounts not consistent with normal practise and THEN the "fooled" the ECU maps. Where did he say HE was fooling the sensor ?

those are the only ways that you can get a lambda 1.00 reading even though you are feeding the engine at lambda 2.00 (without overly obvious things like faulty sensors ect.)

or the monitor/control loop the Lambda sensor is in doesn't fit the burn model of the engine.(sic)

[hightower99]

Your words ..."Any chance I could read the full report?" From that I thought you meant all th details, all the facts, nothing sanitised.

apparently what is posted on the revetec website is a dumbed down version of a paper sent to other companies. That is the report I want to see. It shouldn't be sensitive because it was sent to companies that haven't made any agreement or deal with revetec. I believe I mentioned that I don't expect a full technical report covering all details.

erm, Brad pointed out Toyota have been seeing somthing similar ?

No he didn't he guessed that toyota might have possibly experienced the same thing based on his understanding of a new engine model.

The theories built up over the years are based on "normal" piston movement.
THis has differences.

only very subtle differences that will not have such a great effect as too cause a need to redefine current theory.

Even small ones can make a difference in our theories -- butterfly effect.

Only if those seemingly small differences are fundamentally different.

Nobody suggested THROING THE THEORY out, very seldom does a theorem get discarded, usually it is enhanced as chamber compbustion HAS done for decades. Small changes, especially in the area of flame-front burn.

Yes and those changes have not been caused by single incidents. The majority of changes have been refinement. Basically nothing radically new has been added for years. What revetec is describing is radically different and if it is truly the case then you are looking at totally redefining current theory.

That's OK, that's a classic measurement mistake
WHat Lambda sensors ACTUALLY measure is the voltage at the platiunum electrodes with zirconium to react with the exhaust gas.
THe output is a voltage. How a voltage is decided to be a certain O2 content is based on PRACTICAL measurements of the design during the development process. The o2 is not actually being measured. A chemical reaction involving oxygen is. There can be a difference

Actually I think you will find that Lambda oxygen sensors have a solid-state electrolyte ZrO Ceramic sensor. When it is hot it becomes conductive and generates a characteristic galvanic change at the sensor connections. This voltage is an index of the of the gas oxygen content, the maximum variation occurs at Lambda 1.00. But that is for narrow range sensors. Broadband Lambda oxygen sensors (which is what he needs) are normally wafer sensors of multilayer ceramic design. Basically you have you conventional concentration sensor which acts as a galvanic cell (Nernst sensor) and a limit-current or pump cell. A voltage is applied to the pump cell which is high enough that a limit current sets in (which is proportional to the difference in oxygen concentration at both ends of the sensor). The pump cell then trys to pump oxygen either into or away from the concentration cell so that it is maintained at Lambda 1.00. THe output signal for this type of Lambda oxygen sensor is the relative pump current (i.e 0mA is Lambda 1.00 while any positive current means leaner mixtures and any negative current is rich mixtures). So oxygen is measured but the output is (once translated from a voltage or a current) a lambda value.

How often that voltage is sampled to determine current value can give significantly different answers. How it's "smoothed" in the closed loop control system can then be "confused".
Lambda sensors are also sensitive to temperature and so if the discharged exhaust differs in temperature, swirl, density or a hundred other things it can lead to the sensor getting a "different answer".
In producing more torque at lower revs and lower gas flow then it's entirely feasible that the Lmbda rule for which violtage equates to how much oxygen is incalid in those cases.

First you don't have to be running a lambda sensor in a closed loop control. You can use it solely as a monitoring device. Generally Lambda sensors are relatively easy to diagnose if there is any problem and this is the first thing that should be done when an apparently incorrect reading is seen. Lambda sensors work at very low RPMs and even during idle conditions... why do you think that the slow moving exhuast of the revetec is going to effect the readings?

There are many areas where the model in the monitoring system has multipolle variables any one of which may be the significant difference causing the difference in readings.

except the interpretation of the output isn't based on a multivariable model. there is a simple direct relation in both types of sensors.

Brad continuing his tests and confirming results longer term and on multiple test benches is much more useful than assuming it's wrong and digging too deep AND assuming the theory is "wrong" and not jsut operating in an area it's not developed for.

I certainly agree with continued testing and confirming results. I however disagree with him using a single measured incident as proof of concept. He hasn't proved that the results from the test are correct and repeatable (notice that the current article shows an engine that runs rich (not overly so) and that uses a relatively common style fueling map? also notice that there is no mention of this anomally being applied to the current motor at all?). All I am saying is that something is up and before he starts redefining current theories he should know alittle more about what really is going on.

Measurement systems often encounter issues like this - trust me I've patented and delivered enough

No doubt at all. But current broadband Lambda sensors tend to be very very hard to fool. And Narrow band is hard to fool because it uses ambient atmosphere as it's reference for oxygen concentration.

What Brad said was the Lambda sensor was giving amounts not consistent with normal practise and THEN the "fooled" the ECU maps. Where did he say HE was fooling the sensor ?

actually what he said was that the sensor was still reading normal (I am assuming Lambda 1.00) even though according to other measuring equipment was telling him that the engine was running leaner (i.e using the same amount of fuel per unit time at 4000RPM as it did at 2000RPM and I am assuming an increase in airflow was measured aswell).

or the monitor/control loop the Lambda sensor is in doesn't fit the burn model of the engine.(sic)

Again the sensor can be used in a open ended system and they do not rely on complex burn models infact the whole idea of Lambda (ie exces air value) is a theoretically based idea. Lambda 1.00 means you are feeding the engine with the theoretical amount of oxygen (in the form of a gas in air) needed to completely burn the injected fuel.

[Matra et Alpine]

ht, I can't be bothered. Been over this type of thing with you in the past and we are generally NOT impressed with google'd text
Your sweeping generalisations, unwillingness to grasp what is being said and basic "abuse" of control theory make it a 10 page exercise to awaken the points. You've kind of lost the right to that level of effort from others based on the attitude.

[hightower99]

ht, I can't be bothered. Been over this type of thing with you in the past and we are generally NOT impressed with google'd text

I didn't google it, that is what I remember from my Bosch automotive handbook

Your sweeping generalisations, unwillingness to grasp what is being said and basic "abuse" of control theory make it a 10 page exercise to awaken the points. You've kind of lost the right to that level of effort from others based on the attitude.

I am not the one who is making "sweeping" generalisations! You wanted to be high and mighty and point out that the output of narrow band lambda sensor is a voltage... and tried to imply that they don't measure oxygen concentration directly but through the use of a complex multi-variable model. I simply pointed out how a broadband lambda sensor works (since they would probably be using a broadband instead of a narrowband) and that there is a well known direct relation between the voltage or current output and the gas oxygen content.