The author of these two articles essentially argues that, given enough octane it seems (I'm unclear on how exactly octane is going to affect all these relationships), you are both safer and making more power by retarding timing presumably during peak load/combustion pressure. The safety of the retarded timing should allow you to run a leaner mixture, which in turn yields more power. Running richer is safer not so much because it cools combustion, but because it slows the flame-front.

These two articles written by one of the main guys at Innovate Motorsports, which of course sells all sorts of tuning products including widebands:

Some quotes:

http://www.innovatemotorsports.com/resources/rich.php

Quote :

"Because air has a specific heat of about 1 kJ/kg*deg K, the air charge is only 3.8 C (or K) degrees cooler for the rich mixture compared to the optimum power mixture. This small difference has very little effect on knock or power output. "

Quote :

"Richening up the mixture results in a slower burn, moving the pressure peak later where there is more leverage, hence more torque. Also the pressure peak is lower at a later crank angle and the knock probability is reduced. The same effect can be achieved with an optimum power mixture and more ignition retard. "

http://www.innovatemotorsports.com/resources/myths.php

Quote :

"If the mixture is ignited to early, the piston is still moving up towards TDC as the pressure from the burning mixture builds. This has several effects: The pressure buildup before TDC tries to turn the engine backward, costing power. The point where the pressure in the cylinder peaks is much closer to TDC, with the result of less mechanical leverage on the crankshaft (less power) and also causes MUCH higher pressure peaks and temperatures, leading to knock. Many people with aftermarket turbos don't change the spark advance very much, believing that earlier spark creates more power. To combat knock they make the mixture richer. All that happens really then is that the mixture burns slower and therefore hits the peak pressure closer to the right point. This of course reaffirms the belief that the richer mixture creates more power. In reality the flame front speed was adjusted to get the right peak pressure point. The same result (with more power, less emissions and less fuel consumption) could be achieved by leaving the mixture at the leaner optimum and retarding the ignition more instead. "

anyone have thoughts on this? Also, how exactly would octane fit in? I know this isn't really a hardcore tuning forum but I'm trying to get some thoughts as (God help me) I attempt to tune my own engine on race gas for 400+ whp. Things might apply somewhat differently on a rotary but I imagine the principles are the same.

[Edited on December 19, 2007 at 5:28 PM. Reason : formatting]

12/19/2007 5:27:14 PM

Octane is a big determinant in flame speed. I pretty much agree with what was written, at least at the end. To me, though, there's a lot more to tuning than simply throwing more fuel at the situation, or taking a stab at less spark advance (or boost retard, whichever you prefer). Are you tuning a standalone controller, or will you be modifying the stock computer?

The commentary you quoted seems to be directed more at folks who are adding a turbo to a vehicle that was not originally turbocharged and are only adding fuel to combat knock. This is far from ideal. There's a fuel mixture that will produce maximum brake torque, and there is a spark advance that will also offer maximum brake torque. You want to be able to achieve each of those and not have to mask the inability to adjust one by compromising the other. Plus, the engine will perform better and run better if both settings are optimal for every engine operating condition.

If you were to start from scratch, or even start from a known good normally aspirated spark advance curve and fuel table, you could increase boost incrementally by a few psi at a time, then tune both either by a manifold pressure compensation or in the main fuel and spark tables, depending on the load calculation methods being used. If mass air (not those flapper air flow meters) is determining fueling, you should be able to expand the mass airflow axis of the table to cover the areas not reached normally aspirated. Or if you have capability to compensate the main table based on manifold pressure, you could do it that way.

As you adjust the fuel mixture (at least on an engine dyno, I'm not sure how this would compare on a chassis dyno, but I imagine it is similar), torque will follow a curve. Starting from a lean mixture, say 1.1 lambda and enriching the mixture, the torque should increase and eventually peak. If the mixture is further enriched, the torque output will begin to fall. For gasoline, peak torque occurs around .85-.90 lambda. Likewise with the spark advance: starting from a small amount of advance and gradually increasing, a peak torque will be observed (which often is after the point of knock, so monitor knock if possible) and then will decrease as more advance is applied. The engine will run the best and make the best power and torque when the fuel is in the optimal ratio and the spark is advanced as far as possible without knock. Any compromise of one to mask an incorrect setting of the other will not produce favorable results, and I think that's what the guy was getting at.

12/19/2007 6:13:30 PM

i read the first sentence or two of the original post and can say that drag racers have been knowing this shit for years.

12/19/2007 8:18:40 PM

^^very good post.

running something rich to combat knock is just a bandaid, and a poor tune. i agree fully with what's said there, you're better off running a leaner mixture and retarding timing. very simply put, timing is essentially what causes knock, so i've never understood why some people just choose to dump in more fuel instead of dialing timing curves in. i guess it's just easier to do the former.

i'm not sure if you're just trying to get the car tuned enough to run or you're trying to really tune it yourself, but i would strongly reccomend having it dyno tuned. you've obviously spent a fair amount of time and money, don't ruin it by running on a half ass tune. not only are you taking chances with the safety of the motor, you're never going to get the power out of it that you could with a good dyno tune. if there's one place to spend some money, it's on a tune.

12/19/2007 8:55:16 PM

One of the main points of my project is to tune the car myself (keeping in mind advice from knowledgeable people) because I'm really interested in this stuff and just so I can say I did it (not because I'm a foolish cheapass who does not know the value of a tune). And if I blow the motor it will be a learning experience I guess, but that's why I'm trying to understand tuning more and more each day and hopefully avoid that.

I am using a standalone speed density system (Apexi Power FC) with a basemap designed for a mildly modded turbo vehicle and trying to dial it in for max "safe" power (whatever the hell that is) on a much bigger turbo than the map was designed for, with a lot more boost. Right now I don't even have the car idling because my 20 year old TPS went bad, but I am replacing that once the part gets here. Then after I get driveability worked out under vacuum (acceleration enrichments, cold start, staged injector transition, light load AFR's, decel fuel cut) I will get a couple hours of dyno time and tune it WOT, and then I will eventually have to take it out to VIR to tune it on race track. I will be running open loop only, as the closed loop control system on this ECU is designed only for cruising AFR's on a vehicle with a cat and smog pump.

Given that this car basically won't see pump gas at all, my tentative strategy will be to first richen up the boost map across the board (to make sure I don't have a random cell that's dangerously lean) and datalog the AFR's (according to my calibrated Innovate wideband) and then overlay those on my fuel map, which I can do by using the 0-5v analog input on my engine management system. Then I will lean out the AFR's to whatever the motor seems to like (likely the .8 lambda range) and then play with timing around the peak torque area. Boost will be about 20psi, if this turbo can even hold that to redline. Given that it's a rotary there is no equivalent to adjustable cam gears to worry about, as "valve timing" is set by the shape of the engine ports.

Thanks for the input guys and I'd appreciate any more opinions you may have.

[Edited on December 20, 2007 at 3:35 AM. Reason : we all know what happens when a rotary detonates]

12/20/2007 3:29:49 AM

As you are setting the timing, dial out more advance than you think is necessary to start with, then gradually increase the advance. The initial settings will probably be fine for the moderate boost levels, but once you really start increasing boost above the basic map, take out more advance than you think you should to start with (for the boost levels above the table--the existing values should be pretty close for the moderate boost levels it was tuned for). It's not going to be as easy as if you were tuning on an engine dyno, especially trying to see when you reach peak torque. Definitely datalog as much as you can, and if you are unsure, error on the side of a rich mixture and a lower amount of spark advance.

[Edited on December 20, 2007 at 5:50 AM. Reason : .]

12/20/2007 5:48:42 AM

i didnt read any of it.

but ive been advancin and retardin for years now.

street side tuners nah'mean

12/20/2007 8:38:36 AM

^haha :-x

sounds like you guys are saying 12:1 @ WOT is cheating? and to just go 14:1 and retard? no thanks

12/20/2007 8:51:08 AM

whats all this cheating? you tune it the best you can without knocking/detonating

12/20/2007 9:36:19 AM

^^^^ That's basically what I planned to do. Start conservative for safety, and very carefully dial in more timing. Now if anyone is at all interested...

Another thing I have to deal with is the fact that Rx-7's have two rotors but two injectors per rotor and two spark plugs per rotor. Thus you have to set the split between the firing of the leading and trailing plugs in degrees.



Here is my leading map in terms of degrees... BTDC I think? I always get it mixed up. RPM is on the X axis, MAP is on the Y axis. 10,000 is 1 bar (well, technically Kg/cm^2, I think they're really close but not exactly the same, haven't checked the conversion) of MAP (atmospheric pressure), 20,000 is 1 bar of boost, etc. I will eventually have to rescale this map so that the Y axis is more evenly spaced.



Now here is my split map. This is the number of degrees between the leading and trailing plugs firing (trailing fires second and is physically positoned after the rotor reaches top dead center)



Now here are the injector settings I have to deal with. Basically on the left there you scale the injectors as a percent of the stock ones. Mine says 76.5 because I have 720's (versus stock 550's) in the primary position which makes it lean out. Then you have to set injector latency (to the right of 76.5) compared to the stock ones because bigger injectors take longer to open. Finally you have to set at what duty cycle of the primary injectors will the secondary's start to open up. These settings can all be tricky to figure out, and can cause lean or rich spots and hesitations as the secondary injectors open up. It's a lot to figure out, but most of the numbers I put in there are based on common settings most people use.

[Edited on December 20, 2007 at 10:56 AM. Reason : .]

12/20/2007 10:53:49 AM

comic sans really inspires confidence

12/20/2007 11:26:05 AM

Be sure to fill in that table of injector dead time vs. battery voltage. If you can get a manufacturer's table, that would be even better than guessing. I've never messed with timing a rotary, and without a true TDC to go by, I'd be lost if I had to. I have seen the rotary timing split function in Motec, but I've never messed with it at all.

Quote :

"sounds like you guys are saying 12:1 @ WOT is cheating? and to just go 14:1 and retard? no thanks"

Not saying that at all. .85 lambda is approximately 12:1 a/f running on gasoline. That's right on the mixture range for maximum power/torque. I got the impression from the original article that people were going richer than that to quell detonation without fixing the timing, but by going richer, power is going to be reduced. It's best to get the timing and the fueling correct. I'm like you, though, especially under high load/boost, I wouldn't go leaner than 12 or 12.5:1. The only time to run stoich ratio is at part throttle highway cruise speed where there is not much load on the engine.

[Edited on December 20, 2007 at 5:59 PM. Reason : .]

12/20/2007 5:57:26 PM

^ I would love to mess with a motec. I've heard they're sick, but I don't know much more about them, and they're like $5000+ right after you buy everything (sensors etc)?

another thing I have to deal with is the INSANELY high EGT's of a rotary engine, at least under track conditions. about 1800 degrees during hard driving... at lambda cruising I remember it ran about 1400-1450 on my gauge, and under boost maybe 1200, obviously depending on the mixture etc. I actually sold my autometer EGT gauge to zxappeal because I regularly pegged it at its 1600 degree max during back roads driving.

12/20/2007 8:04:37 PM

Yeah, about 5000 for an M400 with sensors and wiring or around 2500-3000 for an M4 with sensors and wiring. The new software is amazing and the flexibility is much better than the old MS dos based M4 software. I haven't really tuned with anything else, but the Motec did everything I've ever tried or wanted to do. I know that there are other stand along controllers and software that can do a lot of the same things as Motec, but we always found the Motec to be worth it in terms of help files and being able to phone or email someone for very knowledgeable and helpful tech support. Not to mention the ability to be configured for just about any type of crank/cam trigger system and just about any other sensor imaginable.

I don't think I'd be able to let myself spend that much on one for use in my personal vehicle, though, unless it was making me money.

12/20/2007 10:31:11 PM

Quote :

"comic sans really inspires confidence"

hahahahaha that was good.

12/20/2007 10:47:27 PM

mm&ff had and articale about this a while back and you can squeeze more power out of 87 than 89. my unit is tuned for E85 idk what is better. hell i drive it about 3 tmes a month now. but good luck on timing issues

12/21/2007 2:59:12 AM

^ are you fuckin serious? you have no idea whether gas or e85 is better for power? if you can dump about 60% more fuel in, you can make similar numbers with e85

Quote :

"my unit is tuned for E85 pussying out and not showing up for a race"

[Edited on December 21, 2007 at 7:03 AM. Reason : somehow i knew your dumbass would reply to this thread]

12/21/2007 7:02:25 AM

Quote :

"if you can dump about 60% more fuel in, you can make similar numbers with e85"

Only about 35% more E85, and we actually made more power.

12/24/2007 12:01:17 AM

If yall have stations there (my senior year all they had in the area was govt e85) i would do that


things to keep in mind

1 To get an accurate AFR reading you need to switch the meter to Lambda or equivalence ratio setting rather than AFR. Most O2 sensors assume you are running gasoline and will report a stoichimetric mixture as 14.7:1 which is the proper value for gasoline. E85 has a Stoichemetric mixture of between 9.7 - 10:1 and a max power mixture of about 6.98-8.5:1 or so, where with gasoline it is 12.5:1-to 13.1.

2. i think E85 goes to e70 ( i read that somewhere so its worth very little )

3. 105 octane that shouldn't cost you much more per mile than premium.... excellent.

in charlotte e85 is about 15% cheaper but you need 30% more so... not really too bad for 105 octane.


[Edited on December 24, 2007 at 12:20 AM. Reason : .]

12/24/2007 12:16:17 AM

You can still tune using the gasoline a/f ratios, because behind the scenes, that O2 sensor is using lambda. If you shoot for the gasoline max power a/f ratio, it's gonna be close for E85 (12:1 on the gasoline scale when running E85 is still .85 lambda, even though the a/f ratio displayed is not the actual a/f ratio of E85). Makes it confusing, but it's do-able.

12/24/2007 8:41:16 AM

Quote :

"on a much bigger turbo than the map was designed for, with a lot more boost."

Sounds like you need different software

12/24/2007 9:24:31 AM