The Lotus Cars Community banner
  • Hi there! Why not register as a user to enjoy all of the benefits of the site? You may register here. When you register, please pick a username that is non-commercial. If you use a name that appears on any search engine commercially, you must pick another name, whether it applies to you or not. Commercial usernames are for supporting vendor use only. If you want to become a supporting vendor and grow your business, please follow this link. Thanks!

Got my car back, and found something very strange

6K views 71 replies 34 participants last post by  globecom 
#1 ·
They ran my computer printout again for some reason, and something jumped out at me big time: Maximum Engine Speeds.

My top five are as follows:

Max RPM 1 = 8802
Max RPM 2 = 8745
Max RPM 3 = 8737
Max RPM 4 = 8729
Max RPM 5 = 8716

Now, isn't redline and the rev limiter set at 8600? If so, I either have a malfunction in my telemetry, or my rev limiter was set much higher. Any ideas?

I can post that section of the tape if anyone is interested. Oh, yeah-still no standing starts. :D
 
#53 ·
And the ECU is *fast*

Back of the envelope time:

Spark timing wants to be accurate to one degree or less. Power gains and losses are on the table at 2 degrees error, possibly lots of gains. In my bike tuning days, 2 degrees of error could cause your two-stroke to melt.

So the ECU needs to be able to know crank position something like 360 times the redline. Converting to time, we get:

8,000 RPM * 360 = 2,880,000 degrees per second

The time per degree is .3472 micro seconds at that speed.

I don't know how many degrees the ECU averages to get claimed RPM. Crank speed varies between power pulses as energy goes into the system from the pistons (power stroke) and out of it (compression stroke). So the instantaneous RPM jitters way too much.

Averaging over one full rev (I just pulled this out of a hat) is 125 microseconds. THe ECU is CLEARLY on top of things at this rate, because spark and injection timing have to happen far faster than this.

So the cutoff time can be very, very fast. Assume that the rev limiter averages in the RPM of 8 revs, that is still 1 millisecond between onset of overrev and cut out.

If the revs are climbing at 8,000 rev a second (full tach sweep in one second, wich is pretty quick) then using the asumed rev limiter averaging above, the engine gains 8 RPM between first rev above limit and rev limiter smack-down. That is to say, it goes from 8,000 RPM to 8,008 RPM before the rev limiter declares over rev.

Fully digital ECU are *fast*.

If the ECU decides "no fuel, no spark" the engine becomes an air compressor in one millisecond. The ECU might say, "less fuel, retard timing" to be gentle, but the guillitine [sp] option means you feel like you are going through thick mud. I've done it on my motorbike once, and it feels like the machine just died underneath you.
 
#54 ·
Yeah, I think you make some good points, but I don't think its as fast as you claim.

Typically crank/cam position is determined with a wheel with some number of teeth (~30ish) and a missing tooth. The computer measures the rate of the pulses and interpolates to determine the current position. It doesn't actually sample 360 times per revolution (which is the sampling frequency you'd need to measure 2 degree increments directly).

Keep in mind also that the code that does the rev limit isn't necessarily the same code that does the spark/fuel, so it could be running slower, and based on the measurements we have, we I think have to conclude that the computer isn't checking the limit this often.
 
#55 · (Edited)
And if a cylinder has fuel/air mixture and spark is present when ECU decides it is time to cut? Too late... overrev. In a 4-cylinder 4-stroke engine, one cylinder is always in the power stroke, so cutting fuel/spark will not affect one power stroke. If you're at 8000 RPM or so and the engine is under load, there's going to be power to reach a higher rev.

It will overshoot the rev limit every time when accelerating hard, I say, unless the ECU cuts fuel/spark preemptively (at a lower threshold).

BUT not by a lot. The real culprit has to be something else, like how often the ECU checks for rev limit. Overshoot is common in 1ZZ and 2ZZ with stock Toyota ECUs and replacement ECUs like Power FC.
 
#56 ·
Since the discussion about acceleration and force seems to be continuing I will add my 2 cents in the hope that it helps solidify the things already said about your acceleration stopping the instant the force goes away.

I find that when I'm teaching physics pictures are far more effective at explaining a subject than math and words are...so here's an attempt at ASCII graphics pictures to explain what is happening. The pictures are going to have arrows representing forces on a car. The arrows will show the direction of the force and the length of the arrow indicates the amount of force in that direction.

============================

Situation I: Foot on the floor accelerating.

Force from engine -------> CAR <-- Force from air drag etc

The two arrows are in opposite directions so they partially cancel...when you add them together you get:

Net force on car: -----> CAR

As has already been said, force causes acceleration in inverse proportion to the mass of the thing being accelerated. Or, in math: F = M*a or a = F/M. In this case the force points to the right so the car is accelerating to the right.

===========================

Situation II: No force from the engine...this could be because clutch peddle is on the floor, you have gone to neutral throttle, spark has been cut, etc.

Now the force diagram looks like

CAR <-- Force from air drag, etc

Now the net force is to the left, so the car is accelerating to the left (most drivers will call this deceleration).

Note that there is no mention of what was happening BEFORE you had no force from the engine. It doesn't matter...the current forces have no memory of the previous forces.

==========================

Of course, the real issue when it comes to rev limiters, as has already been discussed, is how quickly the force generated by the motor can be turned off. If it is turned off slowly, the acceleration of the car will decrease slowly...if it is turned off instantaneously, the acceleration of the car stops instantaneously as well.

Anyway, sorry for being so teachy...I just got done sending out my latest round of applications to open professor positions in university physics departments so I'm kind of in that mode. :)
 
#57 ·
Just because I'm in the mood, I think it's important that we all remember that when accelerating in the car, all of the force causing the acceleration of the car is located at the tires, "where the rubber meets the road" so to speak, and is due to friction. Hence stickier tires and tires with more contact to the ground are better for cars with enough power to get them moving, and why spinning your tires is bad (at least from a maximum acceleration point of view, since the coefficient of static friction is always greater than the coefficient of kinetic friction). I think I will draw a free body diagram...

Now, in this FBD (Which is not to scale), we have some lovely sources of friction, mainly air friction and rolling friction, represented in yellow. The lovely green color represents the normal forces of the road on the tire in the y direction (which, assuming the car is balanced 50:50, are equal). The nice magnetic blueish color is the weight of the car (mass time acceleration of gravity). And finally, we have the force of road on the tires in the x-direction, which is due to the engine/drivetrain/etc. applying a force on the wheel, which puts a force on the road in the negative x-direction (I have defined, if you haven't noticed, postive x to be in the direction of motion of the car, and negative x to be the opposite) (Also, on that note, engineers are at the center of the universe, because we pick the coordinate system). Now, a FBD of a body in motion is not complete without a kinetic diagram as well, but those are rather boring as they only show the accelerations, which in this case would be an acceleration of the body at the center of mass and a moment due to the fact that the force of the road on the tire is not completely offset by the forces of friction.

I'm sure if I made a mistake in here one of you will catch me, but I would appreciate it if you did it quickly, as my Dynamics II exam is on Friday morning. And if you wanted to change this from acceleration to deceleration, all you have to do is change the magnitude of that lovely force of the road on the tires in the positive x direction, since the rest of the forces are either based on velocity (friction) or weight (normal forces, etc.). You can change it by braking or taking your darn foot off the floor of the car and letting it come down from 8500 rpms and whatnot. Also, this FBD was for a car going in a straight line... a car that is turning is obviously completely different.
 

Attachments

#59 ·
I can confirm that in the Miata, the crankshaft angle sensor is a disc on the front of the crankshaft with about 4 teeth sticking out, which move past a magnetic pickup. So the crankshaft angle has to be determined by interpolation, if 2 degree accuracy is required for the spark timing. That implies that if there are significant variations in the speed of the crankshaft through a single revolution, they must be regular and predictable.

Evl does have a point that the routine in the ECU that knows the crankshaft angle accurately enough to time the spark, may not be talking to the routine that handles the fuel cut on overrev. Speed can be calculated from position, but a calculation takes processing time. The programmers may have chosen to calculate engine speed less frequently (or compare the speed to the rev limit less frequently) in order to conserve processing time for higher-priority calculations. The spark timing might need to be updated as fast as possible to maintain a 2 degree accuracy, but the rev limiter can kick in 100 RPM before the nominal value and not damage anything.
 
#60 ·
dfa2100 said:
Just because I'm in the mood...
Sounds good in paper ;)

1st the balance is not 50:50, there is more weight in the back of the car.

2nd, spining tires does help you go faster, what you want to acheive is maximum friction without getting the tires loose. This is more aparent on bikes and high power dragsters, if you look at the telemetry you'll see that they are constantly spinning tires and the closer you are to the edge of getting loose the better it's for you. Of course at your own risk. of course you can sping them too much but that's getting the tires loose and you loose all the friction.

3rd, acceleration happens at the rear wheels, but deceleration happens mostly in the front wheels. Just pointing it out.

4th there is a downforce coeficient that increases with speed and increases your contact patch making your tires more efficient.

5th Tires flex, hard to see on our cars, but it happens, not that it matters to your write up but just pointing it out, also tires will get taller and skinier and the contact patch changes depending on the speed and temperature, more contact towards the sides of the tires and less on the center of the tire. On our tires and at the temperatures they operate you can't tell visually but it does happen.

6th the forces of the road on the tire are null, (action reaction) pressing harder with your hand on a table causes the table to press harder against your hand until a breaking point if the table or the hand cant take it. In quantum physics they say that mater doesn't like to compress.

7th, the faster you go, time goes slower and space compresses ... nevermind ;)
 
#61 ·
Re: And the ECU is *fast*

nak said:
Back of the envelope time:

Spark timing wants to be accurate to one degree or less. Power gains and losses are on the table at 2 degrees error, possibly lots of gains. In my bike tuning days, 2 degrees of error could cause your two-stroke to melt.

So the ECU needs to be able to know crank position something like 360 times the redline. Converting to time, we get:

8,000 RPM * 360 = 2,880,000 degrees per second

RPM is Revolutions Per Minute, not second. So divide your numbers by 60. Shouldn't be any trouble at all for a ECU.
 
#62 ·
Miguel said:
3rd, acceleration happens at the rear wheels, but deceleration happens mostly in the front wheels. Just pointing it out.
Depends on whether you are braking or lifting...this is why we like rear-wheel drive. :)

4th there is a downforce coeficient that increases with speed and increases your contact patch making your tires more efficient.
Also the air-drag increases with speed.

6th the forces of the road on the tire are null, (action reaction) pressing harder with your hand on a table causes the table to press harder against your hand until a breaking point if the table or the hand cant take it. In quantum physics they say that mater doesn't like to compress.
He has that in there....the normal force (up arrows at the tires) is cancling out the force of gravity from the mass of the car (down arrow in the middle of the car). As you pointed out, the CM of the car is not in the middle of the car...when the blue arrow is moved toward the rear of the car the green arrows will have to change in order to make sure all the torques cancel...we don't want to have a cartwheeling Elise!
 
#63 ·
Thanks for your post, Ara. I'm trying to forget all my college physics, but these guys keep wanting to refresh my memory! :D

I think it likely comes down to a) my rev limiter being set slightly higher than its supposed to, or b) bouncing off the limiter repeatedly at autoxs tends to spike the limit in between terminations.

I'm not too worried either way about it; I believe it highly unlikely I'd have problems with Lotus/Rothrock if there were mechanical problems, and I doubt I'll have any problems of substance in the first 36,000 miles anyway (as I knock on wood). BTW-stellar service by their bodyshop! My car looks brand new, and they also painted my engine covers with the leftover paint, taping off the logo I'd painted. The covers look amazing with the pearl. WOW! Pete Padley get my kudos once again.

:bow:
 
#65 ·
One thing that is not being considered is that the stopping of acceleration when the rev limiter kicks in is not instantaneous.

As pointed out Jerk is the rate of change of acceleration. When the engine hits the limiter or the driver lifts off, the acceleration change is not instantaneous (large jerk) - it gradually (relatively) tapers off from what ever acceleration rate the car was experiencing to zero acceleration (and then negative acceleration). Depending on how sudden this occurs, the car will still be accelerating a bit past the point when things are "told to stop" accelerating - the engine could easily gain a few more revs before they stop increasing. Unless, of course, there is a large jerk when you hit the rev limiter and the driver and passenger are thrown forward against their seat belts. But since that doesn't happen, there is a much smaller jerk, and the rate of acceleration is (relatively) gradual...
 
#66 ·
My mechanic and body man spoke about the revs, and they said it would NOT be an issue for warranty. The mechanic mentioned to Pete that not all cars will be exact in where the limiter is set.

I have no concerns.
 
#69 ·
TimMullen said:
One thing that is not being considered is that the stopping of acceleration when the rev limiter kicks in is not instantaneous.

...
Correct, imo! Remember that not all the energy put out by the engine prior to rev limiter goes into forward momentum. The front of the car lifts, the tires deform to provide forward acceleration, oil is being pressurized through the engine. The oil will continue to flow and pull the components previously pushing it with it after rev limiter hits. A minor thing, but likely to happen to some degree, imo. The tires will unwind from deformity and release its energy after rev limiter. The front of the car will come down releasing its energy. Etc.
 
#70 ·
The oil is an interesting theory. Take a look at smokey yunik's "power secrets" for an amusing section on that. I don't think you're going to see much power from that tho, esp. if the engine has any kind of oil scraper for the crank. Does anyone know?

As for the energy in the tire's deformation, that will be pulling the other way, and I don't think there is enough energy in the weight transfer to be measureable.
 
#71 ·
Evl said:
As for the energy in the tire's deformation, that will be pulling the other way, and I don't think there is enough energy in the weight transfer to be measureable.
I think you are quite correct. My bad! :bow:
 
#72 ·
The answer is so obvious about the rev limiter and higher RPM's

It is the Bristish electrical system sending the impluses to Japan instructing the Japanese engine to limit the revs. Simply a delay in signal.



















I hope no one really takes this comment seriously, but after having read some real doozies here, I have to be carefull and full disclose! ( ha ha ).
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top