FDNewbie

iTrader: (10)

Tookwik, that would be GREAT if you could do it. And for less than $200?? Wow.

A couple of things tho.

1) If you could find out exactly what it entails to do the conversion, maybe even a write-up?
2) What are the drawbacks (if any) of going w/ the MAF sensor? Are we going to be giving up any abilities/features of our cars, or is it all for the better?
3) Are you running the sequential twins w/ your AEM EMS? Because that's at least *my* major concern...since as previously mentioned, running MAF sensor on a single (and I'd assume parallels) isn't that difficult to do. It's doing it on the sequentials that's harder (due to limitations of the ECU's available to utilize MAF sensors)

Thanks
~Ramy

tookwik

Man if it goes through, I would love to give you guys a write-up. As for now, I don't have a computer in my car at all so until I get it in I'll be waiting. We'll see how things go, I'm actually in a wierd position... I ordered the apexi but its on back order so If i get the money back i will get the AEM and supposedly that will be easier to convert than the apexi. I am running sequential twins right now as well.

DaleClark

iTrader: (55)

On swapping the Cosmo AFM to a hotwire - VERY complicated. The sliding cone/flapper door type AFM has a VERY nonlinear response to airflow. Remember, it snaps back via a spring, and has a very weird output. MANY people have tried to find easy ways to run anything besides a flapper door AFM on RX-7's, MR2's, etc. with no luck.

I would be interested to know if Cam at Pettit is actually running true sequential on the 20b or just non-sequential. I think the only way you'll really run true sequential is with the stock ECU or a PowerFC - the funky sequence of events that needs to transpire for sequential to work properly isn't simple.

I know the AEM ECU can do a ton of stuff and can be programmed for a lot, but it really seemed to not take off that much in the RX-7 community. Dunno why, though - haven't looked into it much.

So, let's go over a few things -

- Trying to get an airflow meter to work with the stock ECU or PowerFC - possible, but VERY tricky. You'd need some sort of computer to translate the hotwire's output into input for the MAP sensor, intake air temp sensor, etc. and even then the ECU is gonna be treating it like a MAP input. Unless the PowerFC could have its code re-written for a hotwire MAF, which is possible, but I don't know anyone who could pull off a stunt like that.

- Getting an aftermarket computer to work. Much more possible, but you might not be able to control sequential operation.

- Using an ECU from another car - VERY bad idea. The rotary has too much weirdness - that would just be a bizarre hack job.

I've never really heard of the lack of response by the MAP sensor to be an issue as far as blowing engines up. Is this really that much of a problem?

Regardless, this is an interesting thread and thought experiment .

Dale

SPOautos

I was thinking of this to be more along the lines of a non seq or single turbo car and using a aftermarket standalone and probably a aftermarket hot wire sensor. If the AEM can control seq system and has maps already in it that are maf based then that might be the one to look into.

STEPHEN

su_maverick

not to be rude but lets stay away from single turbos for right now because it was stated before that singles are much easier to convert. Our main issues right now are
1.) Which ECU will give us the required flexibility
2.) Which type of MAF to use
3.) List of everything involved in the conversion
4.) Will it do its job correctly
5.) Who can we trust to do the conversion and tuning
If I missed anything, let me know

maxcooper

Throttle response is supposed to be better with no MAF, since the sensor is closer to the engine. I haven't seen many MAFs after the turbo(s) -- not sure if they can't take the heat mechanically or if perhaps they don't work well when the air temp varies so much (cold off boost, hot on boost). Also, I would think that very large MAFs aren't very accurate at low flow rates. However, it would be cool if you could get a low-restriction (hotwire) MAF to work well in the car.

-Max

FDNewbie

iTrader: (10)

Max,

I'm sorry, I didn't get your first sentence...did you mean to say throttle response is better WITH an MAF, since the sensor is closer to the engine?

I don't think putting the MAF after the turbo was something at least I was considering. It made a lot of sense to me to put it right before the turbos, the closer the better I believe, so it's reading the total air coming in, and minimize the issues that may arise regarding vaccum leaks, etc.

While a very large MAF may not be too accurate at low flow rates, here's my question...would that pose a problem, or affect the problem we're trying to solve? I think not...since at low flow rates, you're running low boost. You're not really running the risk of detonation, so I don't believe this would be much of a problem. The most important function of the MAF I saw was it's very accurate nature at high flow rates, and slight fluctuations in those rates when you're really getting on the throttle...cuz that's when we detonate, run lean, have improper AFRs, etc.

Lastly, I wouldn't think we want something too big or too small...rather, a size that's "just right" to give us optimum performace throughout. I think the MKIV Supra may very well be a good example to copy and adapt to the FD, because I beleive it's facing many of the same issues and load our cars do (lots of boost, sequential twins - more or less, lots of power, etc...yes?

~Ramy

Mahjik

iTrader: (1)

Throttle response is suppose to be better with a MAP (from what I've heard, nothing to back it up other than hearsay). Max was saying the response is suppose to be better with NO MAF (so using a MAP), I believe.

clayne

Utilize both and augment differences by weighting MAP vs MAF or MAF vs MAP, etc.

It would require a custom ECU ala MoTeC.

FDNewbie

iTrader: (10)

Yea that's what I thought he said, but I thought the MAF could be placed right in front of the turbos, and thus would give a more accurate response... My bad.

su_maverick

how much restriction would that cause in the airway? and how pricy would the new ecu be that could read both sensors?

maxcooper

Yes, I meant that speed-density systems (that use a MAP sensor rather than a MAF sensor) generally give better throttle response, or at least that is the conventional wisdom. MAF sensors are usually quite far away from the engine (since they are usually before the turbo) so they are slow to respond when you squeeze the gas pedal.

Accuracy at low flow rates will affect your idle quality, which is important in my opinion for general happiness with the way the car runs. You have to find a size that works well with a 1.3L (=2.6L piston flow rate) engine at idle and doesn't end up being a big restriction at 8000 RPM and full boost. That's a wide range of flows to cover, and it might be hard to find one that works well at both ends of the spectrum.

I'm definitely not saying it can't be done or even that it isn't a good idea. I'm just trying to point out some of the challenges that you'll face in making this work.

-Max

FDNewbie

iTrader: (10)

Max, I totally understand, and that's why I value your input.

And you're right, that IS a large range to try and cover. Maybe I'm too carefree, but idle quality?? I have a PFC w/ the notorious idle bounce, and when I had a prob w/ my TB and it was idling at 2K solid lol....I realized how much I missed the idle bounce. So I guess I've just come to accept that FDs = bad idle hehe. As long as it works...it won't stall on you etc, at least for me, I'm more than willing to forgoe a little imperfection in idle and very low flow rates to maximize efficiency at very high flow rates, since again, the point of all this (in my mind) is to maximize reliability when you're at WOT.

Please feel free to point out flaws in my logic or other problems that we may come across, as I'm learning a lot from it.

Thanks
~Ramy

jimlab

I suppose I should jump in at this point and say that while this may and probably will work, it's more along the lines of something Mazda probably should have done in the first place, along with a more robust fuel system. I don't recommend anyone going through the contortions of trying to make this work with a relatively stock car or even with stock twins. It's just not worth the expense and effort.

In fact, one of the first things that higher horsepower Supras do is get rid of the MAF sensor (theirs is more restrictive than a traditional heated wire MAF) and convert to speed density with an HKS VPC (vein pressure controller) to eliminate the restriction in the intake path. This usually accompanies a single turbo conversion, and the VPC also adds primitive fuel tuning and support for larger injectors.

My engine (1995 LT1) would have come with a MAF sensor, and in fact, I originally planned to use and bought a Granatelli aftermarket MAF sensor (now on John Pham's car, I believe), but when I decided to scrap the stock GM PCM to exceed the 7,000 rpm limitation, I went to a FAST (Fuel Air Spark Technology, formerly Fel-Pro) SEFI-8LO speed density system (no MAF, obviously).

I apologize if I gave anyone the impression that a MAF-based system was a magic cure for detonation. It is one of the reasons that Supra owners can modify with impunity (no "three mod" rule there), but they have an over-engineered fuel system to go along with the MAF sensor that makes it possible. I do think that a MAF-based system would be more forgiving of changes in air flow and temperature, so it is worth looking into, but it would require a considerable investment in time and money to successfully implement.

DaleClark

iTrader: (55)

I'm with Jim. The concept of putting a MAF on an FD is an interesting one and good intellectual exercise, but at the end of the day I don't think it would be worth the CONSIDERABLE effort and cost.

If Mazda would have gone with a more capable fuel system and a hot wire in the first place, that would have been awesome. But they didn't, unfortunately.

Dale

FDNewbie

iTrader: (10)

I totally agree w/ what you're saying...Mazda shoulda done it right the first time around. But unfortunately that's not the case.

Jim, a couple of quick things:

- So when the Supra guys ditch the MAF, go single, and switch to SD, they're susceptible to detonation from boost spike just like us, right? NOTE: I'm saying susceptible to detonation specifically...not blowing an engine, because I know a) the 2JZ engine is nearly indestructible hehe, b) they have an incredible fuel system. But detonation would still be a concern for them, correct?

- I'm sure you know exactly what you're talking about in reference to this being an endeavor on a highly modded car, and requiring a serious fuel system as well. But the point I'm at is my car has all the bolt ons, original engine & twins. I'm waiting to get the car in the shop, getting a new engine, ported, upgraded twins, new fuel system, etc. There's going to be considerable work done to the car, and the fuel system is going to be majorly upgraded, considering the amount of power I'm looking to get. So I think it would work very well. But again, the biggest issue remaining would be the ECU...although it seems that the AEM that Supras use to control their twins may work...or possibly whatever ECU Pettit is using to run the cosmo twins sequentially (which he rebuilts and improves) on the 20B.

So hopefully, I think it can be done the right way, the way Mazda should have done it, and I think this may very well be the prime time to try it out, since my car's going into the shop anyways.

jimlab

Detonation is a concern for everyone, but in this case, we're talking about a single turbo setup with an aftermarket wastegate. Boost spikes are highly unlikely.

For that matter, I've never heard anyone complain about boost spikes with the stock twins, either. I certainly never had any problem, probably because the stock wastegate is much larger than the one on the FD, but I did have an automatic.

ZoomZoom

iTrader: (2)

I Also thought about this back in the late 90's. I asked some Engineers in the Automotive Field. They said the MAF isnt as accurate as the MAP. I think they said it worked well with Lower boost and N/A applications but it either had too much lag time with higher boost or the Resolution or Accuracy was too limited. Its been nearly 5 years so I dont remember the exact explaination but I do remember it not being Viable after hearing it. Also MAF might have Improved in this area but You can probably make this call to Motec or Electromotive to get a good answer.

jimlab

I know for a fact that a MAF sensor works just fine up to 21-22 psi on a MKIV Supra, so I think "not being viable" is inaccurate at best. I question the advice of your "Engineers in the Automotive Field".

As far as accuracy... "In a perfect world, virtually all street-performance engines would use Mass Air, due to its superior accuracy and greater tolerance for engine changes." You can read the rest here...
http://www.carcraft.com/techarticles/24549/

FDNewbie

iTrader: (10)

Yea ZoomZoom, either you mixed it up, or the engineers that told you that got it backwards. All the info I've read states the exact opposite...that a speed density-based system (utilizing a MAP sensor) is NOT accurate. It essentially "guesstimates" how much air you're seeing, but it doesn't actually measure it precisely. And it reacts to this guesstimate via the fuel map it has programmed for that supposed amount of air.

Definitely give that article that Jim linked a good read. It spelled it out for me perfectly...MAF is much more accurate, very precise, and I'd assume much quicker to react (since it's accurate) to slight variations in air flow...something of major concern.


Hmmm...Jim, this question prob. has a REAL obvious answer, but I'm going to ask it anyways, simply because I know you guys know a LOT more than I do. Can porting the wastegate, or getting an aftermarket larger wastegate (whether internal or external) essentially eliminate your chances of detonation from boost spike, given that the wastegate is large enough to handle whatever spike you have? Eg, you're tuned to 15 psi, you usually can spike to 17 psi. Now that you put a new wastegate on that can handle 20 psi...would you run lean at all? Is detonation possible? I'd think you wouldn't even spike at all, since the wastegate would never allow you to build that much boost (17psi), because it's large enough to vent all the excess boost??

Cuz if that makes sense...wouldn't a larger wastegate be a no-brainer when upping the boost? (if it already is, sorry, I'm kinda out of the loop haha)

jimlab

We (Trev Dagley and I) considered trying to integrate an external wastegate with the stock twins back in '97, but cast iron is a pain in the *** to work with and weld, and it just wasn't worth it. We ended up opening up the stock wastegate orifice as far as we could and now that's a pretty common modification for people who have their turbos apart. In most cases, it will help control creep and spiking, but some cars seem to have trouble where others don't. Creep is also more common in non-sequential cars, I believe, especially in "properly converted" cars where the gates have been completely removed and the exhaust manifold has been ported.

The stock wastegate actuator opens at 7 psi. Reaching higher boost levels is simply a matter of restricting or bleeding off pressure to the actuator so that it takes longer to open. A boost controller simply "holds back" pressure and releases it to the actuator when a programmed level is reached. This is intended to promote quicker boost response by keeping the wastegate closed until needed. However, keep in mind that the wastegate isn't venting boost. It's venting exhaust energy, or more accurately, bypassing the turbo(s) by channeling exhaust flow downstream (or to atmosphere, in some cases). Boost is limited by how much exhaust flow bypasses the turbo(s). You simply have to be able to vent enough exhaust energy to limit boost to the level you desire, and that varies with different configurations.

RotorMotor

iTrader: (5)

.

RotorMotor

iTrader: (5)

is this what youre looking for?

FDNewbie

iTrader: (10)

Rotormotor, NIIICEEE But didn't someone say something about the Cosmo MAF being restrictive (since, after all, it IS a sports luxury car, not a high performance car)

jimlab

You can tell that it's restrictive just by looking at it. Here's Granatelli's LT1 MAF sensor for comparison...