• Protip: Profile posts are public! Use Conversations to message other members privately. Everyone can see the content of a profile post.

Rebic's cube

Joined
19 May 2003
Messages
1,393
Location
Green, New Jersey
Hey, I wanted to start a new thread, rather then get blasted for wandering off topic.

Anyway, here's what I wrote earlier with some additions:

Damn, I was afraid that's what was meant by "extra injectors" That really stinks, I would rather have larger injectors instead of seventh in the throttle body:(:(

Are we positive about the extra injector, or are we speculating. In other words can the Rebic handle the more conventional larger injector route?

Also, does the Rebic handle timing? or is that still left to the OEM controller.

How does the CTSC get away with using the stock injectors and no extra one???

I'm afraid I'm quite new to fuel mappings, and the various ECMs, thanks to you guys I'm learning more everyday:):):)
 
Not to sound like a newbie (cause I am) but who is Gerry? I have been reading quite a bit about him.

Sorry for the dumb question.:(
 
Let me start by putting things in perspective. Once upon a time we had carburetors. In reality they were much better than people now give them credit for but they still didn’t compare with fuel injection for achieving power, efficiency, and drivability all from the same engine, not to mention doing it safely. But we managed quite well with carbs for many decades and my old Z-car running 17+ psi and many cars like it were proof enough that with care you can produce lots of power reliably without hi-tech.

So my prior comments about the Rebic IV were in the context of modern day electronics where we’ve come to expect a lot. With installation and tuning by a competent tech I’m sure it can be made to work well enough to achieve reasonably close to the same power as a standalone. The difference is in how finely you can adjust the maps. The “resolution” is so low that if you chart the A/F and find a low spot here and a high spot there, the best you can hope for is a nice compromise between them where you never get so rich or lean as to be a danger. The result may be a few less HP at those points but the difference is not dramatic. Likewise there are a lot of fine-tuning and controlling options that can be done with a standalone. But the good news is that the stock ECU ensures great drivability, which requires skill to achieve from a standalone.

But looked at another way, if Corky Bell had the Rebic when he introduced his TT kit there would be a lot more of them around today.

As for running the primary injectors, no, the Rebic IV won’t do that. It is a simple controller driven by boost and RPM so it can’t replace the stock injector controller (like the Tec3) and is not designed to work with it (like the SS box). So if the Rebic IV is the controller then auxiliary injectors are it. It also does not control timing, and I certainly do not recommend any NSX turbo system without timing control. The base CTSC gets away with it but just barely.

The sad, but hopefully temporary reality is that your more advanced vehicle has left you working with yesterday’s technology until the aftermarket sees enough profit potential to adapt their systems to work with OBDII.
 
Originally posted by sjs (snip)The difference is in how finely you can adjust the maps. The “resolution” is so low that if you chart the A/F and find a low spot here and a high spot there, the best you can hope for is a nice compromise between them where you never get so rich or lean as to be a danger.(snip)
What would you consider fine enough resolution for the fuel map when on boost? 16 points? 32 points? 256 points?
 
sjs,

Thanks, for the info. If you have a 4 wire O2 sensor, can you monitor air/fuel ratio while driving, or do you need to have it on a dyno. In other words, lets say you tuned your car perfectly on a dyno, but wanted to continue monitoring while driving throughout the years. I wouldn't want to bring it back to a dyno every year, although nothing should change drastically, it would be nice to see the F/A ratio. Example: lets say after a year the Rebic, or whatever device you have goes faulty and throws your mixture out a little, if you didn't know it I would assume major engine damage could occur.

In addition, you may have hinted this before. If you have an additional injector in the throttle body, isn't possible for the cylinders closer to the throttle body to be running more rich than the others?? I would think that if the cylinders furthest away ran lean this could lead to big problems?? can this happen, or no??

John
 
BryanZublin said:
What would you consider fine enough resolution for the fuel map when on boost? 16 points? 32 points? 256 points?

Although not a quantitative answer I would say whatever it takes to smooth out the A/F curve to an “acceptable” degree. My loose definition of “acceptable” is probably +/- about .15 A/F so if my target at a particular portion of the map is 11.9 I would want enough resolution and control to keep it between 11.75 and 12.05. Obviously I’d like even tighter control than that but we’re talking “acceptable” and I’m not sure what the practical limits really are in terms of things like measurement accuracy (both absolute and repeatability), trim for environmental condition changes, and even controller precision. Hopefully I learn more about these when I get back to tuning on my own cars.

Anyway, if a given fuel management solution can’t achieve that degree of tuning and consistency then the only safe approach is to move the A/F curve somewhat richer so that the lean areas never get dangerous. That’s where I landed with the Rebic IV, but I am not a master tuner of that device nor did I have unlimited dyno time to tinker with it. Only after giving up on it did I get my own wideband (I waited 10 months for one I paid for an never received, but that’s another story) so it could be said that I didn’t give it a fair assessment. None the less, the degree of adjustability is well short of anything like the SS box or a full ECU so tossing it aside was a no-brainer.. And at 50% the price of an AEM it is no longer bargain for those who don’t have an OBDII .

Another concern with regards to A/F consistency is the use of a variable fuel pressure regulator. Although a very clever device and adequate for modest applications, I suspect them of being less consistent than I’d like and thereby adding to the problem. I’m curious to know whether that has been done away with in the revamped Cartech design.
 
jadkar said:
...If you have a 4 wire O2 sensor, can you monitor air/fuel ratio while driving, or do you need to have it on a dyno. In other words, lets say you tuned your car perfectly on a dyno, but wanted to continue monitoring while driving throughout the years...

If you have a wide band sensor and meter (as opposed to the OEM narrow band) then it is certainly possible to monitor the mixture at all times and hopefully catch problems before they happen. And there are multiple reasons why your A/F might change, such as a dirty fuel filter or injectors, a failed injector or MAP sensor for the auxiliary controller, weak fuel pump, etc. Wide band sensors (typically 5-wire) and controllers are still a bit expensive but well worth having. (I now own three of them) The sensors seem more sensitive to damage than their lesser cousins (especially from unburned fuel, oil blow-by, and even condensation if mounted wrong) and there is some debate as to how long a sensor will last even under good conditions. They can be recalibrated but most of the popular controllers don’t offer that feature. One other thing is that they must always be connected (for the internal heater) when the car is running or they will quickly clog and can’t be cleaned, so you can’t just switch them on when you want to check the mixture.

Before someone asks... A follow-up note on wide bands being more sensitive to damage from contaminates etc., it may depend on how you look at it. A typical narrow band sensor is very nearly a two position switch, heavily biased at 1 lambda and indicating either rich or lean with very little resolution either side of the14.7:1 magic number. Even if it drifts with age it remains reasonably close to that point so everything is happy in closed loop mode. Who really cares whether it’s actually 14.2 or 15.2 except perhaps the CARB sniffers and their brethren when you go for a checkup. But if a wide band drifts by that much at 12:1 you may be betting an expensive 500hp engine on it. Do the narrows drift that much with age? I don’t really know, but the wide bands will.
 
Follow-up

BryanZublin said:
What would you consider fine enough resolution for the fuel map when on boost? 16 points? 32 points? 256 points?
sjs said:
...I would say whatever it takes to smooth out the A/F curve to an “acceptable” degree. My loose definition of “acceptable” is probably +/- about .15 A/F...
Although I refer to Tec3, this might relate to other stand-alones as far as resolution is concerned. In the A/F calibration section of their software manual, they state "The VE table is then linearized between these points and extrapolated beyond the defined RPM and MAP/kPa table ranges." So, the question is: If the A/F ratio is linearized between points, is it always a straight line from point to point? How does one know how much the A/f can deviate from resolute point to point?

BTW, software and manual can be downloaded at the Electromotive Web.
 
Black&Tan97Tnewportbeach said:
How come its not interpolated like as in oversampling like CD players even did in the 80's?

Just guessing here, but I suspect it is processor limited. In your CD example the output must be delivered at a predetermined fixed rate, so as read and processor speeds increased they were able to put it to good use. With the Tec3 and similar devices they split the processor between many inputs/outputs and are probably sampling at capacity to achieve the base resolution. The Tec3 is already somewhat behind with an 8x8 grid for the main fuel map compared to 17x21 for the AEM and even more for the Motec. Does the Motec do it differently? I don’t know, but I’ve heard the processor is 2x that of the AEM. Of course that’s only ½ the story, it still needs software/firmware that puts it to good and efficient use.
 
Black&Tan97Tnewportbeach said:
I thought the TEC3 was 16x16, and the TEC2 was the 8x8 one?
Ah yes, you're right. I've ben sititng here looking at WinTec2, having thought I downloaded the Tec3 software. At any rate, even though twice the resolution, my original question is still relevant.
 
KGP said:
Ah yes, you're right. I've ben sititng here looking at WinTec2, having thought I downloaded the Tec3 software. At any rate, even though twice the resolution, my original question is still relevant.

Indeed! I'm behind the curve again, and the guys in Vegas aided in the deception. (you know who you are. :) )

OK, well I feel much better about the Tec3 now. But did I answer your real question about over sampling?
 
"How come its not interpolated like as in oversampling like CD players even did in the 80's?"

Interpolation methods are used when sampling rates cannot achieve the desired data capture. So if you sample between two points in time, but you actually need data in between those two points you interpolate the values in between. An oversimplified example:

You sample a signal that has a dynamic range of 1-10. Your first sample is 5 and your second sample is 10. The interval in time between the samples is 10 sec.

What is the value at the 5 sec. point (which is data you did not/Could not capture). If it is a linear interpolation it is 7.5 (half way between 5-10)

Oversampling would indicate sampling at a greater rate then required and interpolation is not required. This relates to "Nyquist Rates" and sampling theory. A primer is here: http://www2.egr.uh.edu/~glover/applets/Sampling/Sampling.html if interested.

HTH,
LarryB
 
Great information guys. Larry, the wheel nut removal was easy as per your instructions. Now I can use both sets of wheels with the longer studs. Steve, Gerry did add a RRFPR on my car. This suprised me. I thought taking away a constant fuel pressure would make mapping the fuel a nightmare but it runs great so far. Still a lot to learn here. Thanks All. Dan
 
Originally posted by tunapie (snip)Gerry did add a RRFPR on my car. This suprised me. I thought taking away a constant fuel pressure would make mapping the fuel a nightmare but it runs great so far.(anip)
Dan, please say it ain't so. :) I thought one of the major advantages with a standalone ECU (TEC3 in your case) is that it does not require a RRFPR. Did Jerry say why it was needed? What is the flow rate of your injectors?
 
I'm no expert on engine tuning, but in relation to the discussions on sampling.

When a CD player is sampling it is doing so at a fixed rate. The oversampling means it samples at a higher rate than necessary, but fills in the samples with zeros. This has the effect of resetting the D/A converters for a better sound, but I digress.

With a car ECU though, doesn't the sample rate change with engine speed? I would have thought that readings of throttle position, air mass etc, would be done based on crank angle? So sampling frequency would vary with engine speed?
 
Suprised me also, when I started this project I did the calcs and came up with 430cc, MB was using the 440s so that is what I used. I will ask Gerry for his reasoning with the RRFPR. Dan
 
tunapie said:
Suprised me also, when I started this project I did the calcs and came up with 430cc, MB was using the 440s so that is what I used. I will ask Gerry for his reasoning with the RRFPR. Dan

Me too. Besides making tuning tricky the RRFPRs don't seem all that consistent, at least compared to the tight control available today. Very odd.
 
I personaly think to ME that using the rrfpr makes tuning simple.
Here how it usualy works for ME, install Tech, start car and get the injector perimeters set, drive car around under different conditions so that all is drivable, this usualy takes around 30 minutes to an hour, go to the dyno and make a moderate run, do some adjusting, and usualy I can have a car off the dyno finished in less than 5 dyno runs. I think that if you can tune a car in that amount of time and get good safe results than it must work great.
Aren't some of the FI cars still after months of tuning still trying to get their cars tuned?
Anyone that has been in my car for a ride or even driven it can tell you that it drives like a stock NSX, only when you put it under load it feels like a monster V-8.
I think that we are talking about 2 different things, when I say rrfpr, it is a one to one not a FMU, for every pound of boost it increases the rail pressure one pound, almost like stock but you can adjust the rail pressure at idle.
 
Last edited:
Hey, whatever works. No doubt like many things, if you’ve got enough experience with it you can make it do what you want in a hurry. It effectively allows you to create the base map via pressure then use the Tec3 for finer tuning, which as you say can speed the tuning process and possibly yield better results. My only concerns are the consistency of an RRFPR (which may be fine in most cases) and the fact that I’d prefer to run modest fuel pressures if possible. And of course the fewer failure points the better.

As for other people having problems getting full standalones to work, I’m not sure that’s a fair comparison. Certainly there are some very well sorted Motecs and at least a couple AEM’s now, but for the most part NSX folks don’t yet have the level of experience with them that you have with the Tec. And of course they are more complex because they run everything, including things that cost many hundreds of bucks to add with other systems. I think you’ll see plenty of happy owners with those in the near future as well.
 
Originally posted by Gerry Johnson (snip)I think that we are talking about 2 different things, when I say rrfpr, it is a one to one not a FMU, for every pound of boost it increases the rail pressure one pound, almost like stock but you can adjust the rail pressure at idle.
Gerry, thanks for the clarification. I believe that most people say "RRFPR" when they are referring to a device that raises the fuel pressure by MORE than 1:1 relative to boost pressure. This type of regulator is used on the Comptech SC for fuel management.

As you correctly mention, all stock fuel pressure regulators are referenced 1:1 to manifold vacuum/boost. So I presume that you have replaced the stock fuel pressure regulator with an after market regulator for the sole purpose of adjusting the static pressure differential? This is like a "global" fuel adjustment that will richen/lean the mixture with a simple turn of the adjustment screw. Do you typically increase or decrease the pressure relative to stock? The stock NSX pressure is around 50 psi at WOT on the NA engines.
 
Last edited:
Originally posted by sjs (snip)My only concerns are the consistency of an RRFPR (which may be fine in most cases) and the fact that I’d prefer to run modest fuel pressures if possible. And of course the fewer failure points the better.(snip)
Please read Gerry's clarification above, he is not using a RRFPR but a 1:1 regulator.

I just realized that lowering the fuel pressure may be a good idea for engines using significantly larger injectors compared to stock to improve idle quality. Using larger injectors results in shorter injector pulse widths and less control because of fixed open & close times of the injector. So lowering the fuel pressure a bit will increase the pulse width and may improve idle quality. The lower fuel pressure will also require less demand on the stock fuel pump during boost.
 
Back
Top