Did you forget that the 20B has 3 rotors?

It's been awhile since I've done any research, but options for engine management were fairly limited when I bought mine. Regardless, a system capable of running a 20B would be different than one capable of running a 13B, with or without a MAF sensor. I'd concentrate on systems that are known to work with 13Bs and MAF sensors (like the Wolf 3D) myself.

 

Yea, it would have been

Quote:

You can find answers to your questions about the 20B in the 20B forum... or find someone who can answer them for you.

Jim, I was brainstorming, not expecting you to sit and answer every question I have But since the 20B does NOT have an MAF based system, I'm not really interested in the ECUs they run.

 

I didn't forget, but I guess I oversimplified. In my oversimplified thinking, I assumed if you have an ECU that can be programmed to run a 4 cylinder, 6 cylinder, or 8 cylinder car, you can vary an ECU to run 2 or 3 rotors. But I'm sure that's not the case. Bad assumption.

But you definitely have a good point. Stick to something close and similar. Less issues to take care of. Good lookin out Jim. Thanks.

 

FDN- When you get back from RP, let us know what Bret says about the MAF conversion, I would like to know if he has any take on it.

 

I talked to Bret, and he said he has no experience w/ the Wolf ECU. I'm not saying anything about his worksmanship, but when I talked to him, he didn't really seem to have a good sense/grasp of things when it came to tuning...

He said he didn't really see much benefit to it, so I told him it would be very beneficial in preventing detonation from boost creep and spike. He was like yea but you just need to upgrade your injectors, and you can use a Supra fuel pump (denso pump), etc.

While all that is true, none of that will help you when it comes to boost spike, since you're not tuned for that specific boost level. And it can pretty much happen no matter what level you are tuned at...because by definition, boost spike is spiking to a boost level HIGHER than what you've tuned for (if I understand it correctly). So you may very well have the fuel system to handle it, but you're not tuned for that sudden spike, and your speed density-based system doesn't catch on w/ such sudden and quick changes, you run super lean, and boom! you have detonation. So I believe an MAF sensor would be GREATLY beneficial. So in short, I don't think he's the best person to ask.

BUT, he DID tell me all 2nd gens came w/ MAF sensors...

Afterward, I passed by PFS to see if I could talk to Peter, but PFS was closed.

 

Hmmm, nothing against Bret, but do you think he is the best person in the area to take my car to for work? You did seem suprised in the past that he did work on 3rd gens.

I can try to stop by PFS this week and ask around. Im always runnign around NoVA on work.

 

I just emailed KDR performance up on PA to see if thay have any ideas on the MAF convesion. We shall see what they say

 

*From what I know* he's the man when it comes to 1st and 2nd gens. I simply didn't know he worked on 3rd gens much.

Also, someone on the forum has in their sig something along the lines of their mods, then it says "and f---ed up turbos, thanks to Rotary Performance." So I'm a little weary.

Prob. is, I've heard shady stuff about PFS, the wait is forever to get your car in, they're expensive, and they really don't like workin on 3rd gens much anymore.

That leaves you w/ Demetrious of Reactive Racing, who I've heard nothing but rave reviews about. Only problem is, I have yet to get ahold of him, and it's been 2 months and counting...lol

For now, I would call or swing by PFS in the early morning and see if they can do a 1 or 2 hour diagnostic on your car, and figure out what the deal is w/ your 4 psi of boost.

I gotta run by Manassas to pick up my painted headlight cover anyways, so I'm going to pass by PFS and see if I can talk to Peter about the MAF.

I'm also gonna see if I can have a quick chat w/ Steve from Gotham about it...see what he thinks.

 

I emailed Demetrious and shall see how long it takes (if ever) for a response.
Basically the idea on the MAF is to talk to as many people as possible to see if someone has any ideas. I think its a great idea to help out people to keep engines out of the shop.

 

If Demetrious gets back to you, you better let me know, cuz I'm going to tag along haha

While it may be good to tell ppl about the MAF, nothing's gonna happen till someone tries it out, and it WORKS. So right now, I'm looking for constructive criticism more than anything else...

 

Honestly I dont think you'r going to get a hold of anyone that has tried to do this. I'm talking to some good piston engine tuners that have a lot of experience with a broad assortment of standalones. If I get this to work out I'll do a thread and let everyone know. I've already emailed Steve, he's VERY VERY VERY busy so you guys try not to bother him about this since I already have.

You guys rest assured if this can be done I'll prob know how to do it pretty soon and I'll let everyone know about it. I might even sell it as a pretuned kit or something...who knows, just depends on what all is involved and how much it would cost....which is going to hinge on the different good performing ecu's capable of doing this.

STEPHEN

 

HE'S ALIVE!!! LOL

Stephen,

Thanks for the info. I'm PMing you BTW.

~Ramy

EDIT: PS, I wasn't lookin for someone who's done it...just someone who understands the concept, and to get their ideas about it...

 

not bugging him about this.
Im bugging him about finding out why my car is running 4psi and a possible rebuild.

 

Thats ok....I wasnt meaning dont contact him if you have other things going on. I just wanted to let you guys know I'm already contacting him about it so there is no need....I dont want to take away from his business with a bunch of people asking him the same quesitons about this.

 

Update-
Talked with the boys at KDR. They havent heard about it being done but I have got some info. The wolf ecu will allow for a MAF FD. Aside from that, it comes down to which to use and finding someone that can tune the ecu properly without going through 4 engines to do it.

 

According to the japanese greddy e-manage manual even the 20b used a AFM.

FL - stands for flap type AFM
HW - Hot wire
PR - obviously pressure sensor or MAP based

http://www.mohdparts.com/emanage/man...oftware/54.JPG

 

so what do you expect to gain by switching to a maf? I'd like to see some kind of discussion of the tradeoffs. all of the "hot rodders" i know prefer speed density to maf.

 

I find it interesting that you guys actually WANT go to go a MAF based system. I'm accustomed to working with these, since I specialize in FC's.

I can tell you that you can blow an FC engine up just as quickly as an FD when improperly modded...sometimes easier. Some of this can be contributed to the older, slower, less sophisticated ecu's the FC's came with (as compared to the FD).

I can also tell you that any FC owner would KILL to have a speed density system like the FD...the intake restriction of the flapper type AFM on the FC is large. People take stock turbo II's and convert over to haltechs and gain 40-50hp...partially by retuned fuel and ignition maps, partially by removing restriction of the afm. IF you actually tried this, you'd want to go with one of the newer, large, hot wire type air meters found on newer cars. Flapper and cone style meters suck for air flow.

Another thing to keep in mind for you guys. AFM based systems require that there be ABSOLUTELY no vacuum leaks anywhere after the meter. ON most stock cars, the meter is placed before the turbo. Therefore, anywhere from the air flow meter back must be airtight or the car won't run right...getting an AFM based car to idle with a vacuum leak is a bitch. You guys have enough issues keeping your hoses straight enough to get boost...do you really want to make the already complex system even more so?

I know that mazda did some dumb **** with all these cars, but I have to give them enough credit to say that they probably knew what they were doing when they went away from AFM based systems in fc's and into SD based systems in FD's. Each has its good and bad sides, but speed density offers a freer flowing intake path and no need to worry about pre-throttlebody intake/vacuum leaks.

Still, I'll be interested to see if anyone comes up with anything in this area.

 

http://ausrotary.dntinternet.com/for...c.php?p=333460:


RXWASP
Get a Life


Joined: 25 Jun 2002
Posts: 2152
Location: Victoria

Posted: Sun May 02, 2004 10:14 pm Post subject:
Just to add some more info.

At the end of the day, what we're trying to do is meter the amount of air being consumed by the motor. If we know how much air is being consumed, we can then, relatively easily, work out the amount of fuel to be delivered.

Obviously, air flow meters, as their name implies, perform this task naturally. Whether they're a cone-displacement type, moving-vane type or hot-wire type, they provive an output proportional to the volume of air flowing past the sensor. Since the metering point is of a known and fixed cross-sectional area, it is relatively simple to translate that flow into a volume of air being consumed.

However, most aftermarket engine management systems utilise (as has been said above) either a MAP or TPS sensor to approximate the volume of air being consumed by the engine.

MAP-based systems rely on an air temperature sensor to determine the density of the air. This can be then, in turn, used to approximate the volume of air in the inlet system - a calculation that is made possible by having several fixed variables This point is highly relavent on turbocharged vehicles, as their inlet temperature range can vary immensely.

TPS-based systems attempt to approximate the volume of air being consumed by the engine, by assuming the amount of air being consumed by the engine is proportional to the throttle opening. Generally, this metering method is quite crude, but is the only choice for highly-ported enignes.

A MAP/TPS hybrid can solve some of the problems, but is still not perfect.

Another problem to consider with the MAP-based system, is the dual-purpose of the air temperature correction map (for fuelling). The first use is explained earlier: approximation of the volume of air being consumed. The second use is to account for the changes in the air's ability to carry a fuel mixture. Generally, both of these follow the same trend (i.e. an increase in fuel delivery as temperatures drop), but it's something to keep in mind.
_________________
FC3S Turbo
Slowly being reassembled!
"Only two things are infinite, the universe and human stupidity, and I'm not sure about the former." - Einstein
-------



bill shurvinton
Weekend Cruiser


Joined: 12 Dec 2003
Posts: 57


Posted: Mon May 03, 2004 7:39 am Post subject:
If I can just introduce some terms here to make life easier

A TPS based system is generally referred to as Alpha-N, or AN.

A MAP based system is generally referred to as speed density or SD.

Right that makes it all easier to type. Now SD systems actually solve the ideal gas law to estimate airflow (PV=mRT). AN systems rely on a loolup table for fuel. Both require air temperature to compensate for air density changes.

Now back to MAPdot lagging TPSdot, this is true. However it does get us into an interesting region, as in some cases TPSdot is actually leading the event (if the throttle has opened but pressure hasn't changed, then there is no required change to the fuelling requirements). I have seen logs where there are MAPdot events that do not register as significant in TPSdot. Now there are only a handful of people around who tune their engines to that level, but it exists.

The other thing to bear in mind is that MAP varies over a cycle. In fact this property is used by researchers to measure pulse tuning effects. There are 2 ways around this.

1. Put a low pass filter in the map line to remove them and take an average

2. Sample at a known point in the cycle. Motorbike EFI systems use this, so if anyone says that MAP doesn't work with big overlaps....

What this all comesdown to is that it is complex to optimise any system for a highly tuned engine. You have to look at the compromises involved and chose the tuning method that best suits you. Often this is the one you can get your head around.

And a brief note on mass air flow systems. They are great on stock engines, but there are a couple of reasons that the aftermarket do not use them. Firstly they have their own problems with dynamic range. Secondly they need to be calibrated to the inlet system the work on, as you have to be able to tell the difference between air going in and air going out. Given the levels of pulse tuning on a high performance engine, you can see why this gets to be beyond the amateur.

I hope this doesn't put anyone off

 

Another good one:

http://www.carcraft.com/techarticles/24549/

 

Ok, but say you were planning to stay below 350hp and trying to just upgrade reliability. This is just mentioning highly tuned cars. Would a MAF be better suited to saving the car from detonaion from boost? I appologize if you covered this and I am just not seeing it.

 

Some interesting ideas here. Here's my 2 cents -

I do believe the 20b used a larger version of the FC cone-style airflow meter. The Cosmo 13b may have also used an airflow meter. Problem is the Cosmo ECU's are useless to us, since they have all the automatic transmission controls in them, as well as some of the other luxury features in the car, and won't run properly without seeing the auto tranny and whatnot.

Not to mention the Cosmo airflow meter is a sliding cone, which would be a big step backwards.

The RX-8 uses a hotwire airflow meter, but it's ECU is VERY tightly integrated with the car - no way in hell you'd get that to work on an FD.

I think you may need to outline the goals a bit better. If you want to control the stock twins, you need a PowerFC or the stock ECU, or a chipped stock ECU - nothing else will do the trick, I don't believe. With a single turbo, you can use most anything.

If the sole reason of having an airflow meter is to compensate for boost spikes, that may be the wrong way to tackle the problem. Even airflow meters have lookup tables that have to be programmed - x amount of airflow means y amount of injector pulsewidth. It would have to be mapped appropriately to the limit of the airflow meter to be fully capable of the task. If you're doing that, you could simply take a MAP based system, go into the maps for higher than stock boost ranges, and put in a TON of fuel so it runs super rich if it overboosts.

Another solution is a pop-off valve. If you're mainly worried about accidental overboosting, a pop-off valve does the trick. You can use an air compressor to pressurize the intake of the engine to whatever the "overboost" amount is, then adjust the valve to blow off at that pressure. If something does happen, the valve pops open, instantly dumping boost pressure.

Pop off valves aren't good for routine boost control, only emergency boost control. When it opens, it isn't maintaining boost pressure at a level, it dumps all the pressure out, kind of like having an intercooler hose pop off. It's enough to have you stop and realize that you've got a boost control problem that needs fixin'.

In short, it's a case of it would have been nice if Mazda gave the FD a fast-reacting hotwire AFM or Karmann AFM instead of the MAP based system. The biggest advantage there is the ECU would likely have handled moderate modifications MUCH more easily without having to do something to the ECU so early in the game. Heck, there are DSM's out there running 11's or less with the stock ECU. But, airflow meters don't magically read airflow and provide the correct injection amount - they still look up fuel values in a table, just like a MAP sensor does. The only real advantage is an airflow meter can more easily compensate for a change in the volumetric efficiency of the system (ie, opening the intake/exhaust up) than a MAP system does, since the VE of the system is a constant in the ECU.

Dale

 

The AEM ecu is another possible path, that would also allow sequential turbos and knock control. I think they have both MAP and MAF input sheets in all boxes. Use the supra MAF and AEM supra base fuel and timing maps to start with ( 3L-T vs 2.6L-T) . Maybe WBO2 could help trim the maps.

Key thing is to use an ecu with MAF base maps for a similar sized application, and that oem MAF sensor.

 

WOW...hehe. All of a sudden, we got a whole rack of some serious technical info. That's good...cuz I'm still a noob, so I was talking more of concepts than mechanics... So thanks to everyone who contributed.

RotaryResurrection, first, I was considering using an MAF setup off a very programmable ECU...so that eliminates one concern. Also, from the little that I have read about MAF systems, there are quite a few diff types (such as the hot wire air meters you mentioned) which I believe work very well, and they come in all diff sizes. I take it the bigger the better (so it poses less restriction)? So of course if we were to actually go w/ an MAF system, we would be sure to pick the most efficient and appropriate sensor.

I didn't know the bit about vaccum leaks and what not... I believe Jim had stated earlier that the Supra's MAF is right after the intake, since they don't have their BOV venting back into it. So I do think ideal placement can be achieved on FDs...

So in short, while our speed density system offers a freer flowing intake path, I believe picking the right MAF sensor, the correct size, and correct location should provide optimal (and better) results than the FD's current speed density system.

Clayne, I think it's a no-brainer that if this venture were undertaken, it would only be professionally tuned (Demetrious, Steve Khan, etc). And that's some good info you posted. Rhode Dog had posted this link up o the previous page, (http://www.dejontool.com/Fuel.htm) and it hit on a lot of the same points. I didn't know/read about TPSdot being the leading event tho. Definitley some interesting stuff...

Dale...you hit the problem on the nose: What ECU? As is clear, running a single w/ an MAF isn't a problem. It's keeping sequential twins and using an MAF that's challenging. While I understand the stock Cosmo ECU won't work, my immediate question is, what ECU is Pettit running on his Banzai, that runs the 20B AND his upgraded sequential twins?? That may very well be a possible option.

Oh and when you say the Cosmo airflow meter is a sliding cone, can't you change/upgrade the MAF sensor in a vehicle? Because that link I posted above (originally posted by Rhode Dog) has all sorts of upgraded MAF sensors for some domestics...

As for my goals (to answer your and RotaryResurrection's questions), I'm not simply trying to control the stock twins. Like you said, we have PFCs, which seem to work great in doing that. Compensating for boost spikes is definitely one of the strong points in my opinion, since the way I understand it, the MAF sensor is much more accurate and quick in response, that it should be able to detect sudden changes in airflow, and thus adequately change fuel delivery. Yes, that's still off of a program, but when you are programmed for only 13 psi on a SD system, and you all of a sudden see 16psi, even if you have provisions in your ECU for such a case, if your SD sensor never sees or senses that change quickly enough, nothing will happen. While with an MAF sensor, if you program your ECU accordingly for such a spike, the MAF sensor - being dynamic and very accurate (not just an approximation) would be able to immediately sense the change in airflow, and thus respond accordingly, in time to actually correct the problem, or so I believe/understand.

Also, no one wants to run super rich. That's basically like saying run untuned lol. I wouldn't have this whole thread going if getting max hp wasn't an ultimate goal, so I don't think running super rich would be a viable option.

This isn't the first time I've heard about the pop-off valve solution. I believe it was DCrosby who mentioned it a while back in another thread I had started regarding whether or not you need to port your wastegate to eliminate boost spike (there's another solution! porting the wastegate! lol) While I'd hate to lose all my boost suddenly, esp. if I was trying to take another car, I'd much rather lose the race than lose the engine. I just might run a pop-off valve either way, just to be safe.

Yep...that pretty much sums up the rest of the positive points of running an MAF.

Lastly, KevinK2 jumped on my idea...which I'm actively exploring. Any input would be greatly appreciated.

The one thing I did hear is that the Supra twins aren't running a truly sequential setup...they work a little differently. I don't know anything about them...so this part is all on you guys. Jim, you'd know, right? You used to have a MKIV, right?

 

This is wierd that this came up at this time but I was looking to get an AEM EMS tuned at my local shop (which is AEM certified to work on the EMS) and he recommended me changing to a MAF based system. All for less than $200. I really want to let him try it out. As long as he pays for it if he messes up, I got nothing to loose right?