I know from working on ford 2.3t's that a weak link in many early EFI systems is the "vain air meter," or the trap door/flapper meter. Obviously Mazda switched to regular MAF for the 89 and up cars, but how much are we S3 and S4 guys losing out because of the 'antiquated' design. Also, how feasible is it to convert to mass air? Has anyone converted an S3/S4 to MAF?

Seems like the 2nd gen section would have a wider audience to receive this question than the carb guys in the 1st gen forum.

 

I've said it before and I'll say it again, the chief source of restriction for the rotary engine is in the exhaust, not the intake, at least for moderate setups. I mean eventually you get to the point where the AFM will hold you back. But that limit is higher than most people think.

The answer my friend is about 0 horsepower.



Alright, I don't necessarily mean that literally. But on that dyno, there is my old nonturbo car with stock ECU + SAFC (red), and then in blue my car months later (summer time) running no AFM with a Megasquirt. Smoothing was disabled in the graph and SAE correction was used.

Now you could argue that maybe my tune wasn't ideal, or that the two runs weren't back-to-back or something. It was so long ago now that I can't give you many details about the tune (from a local shop) as I didn't know that much about tuning at the time. But how many horsepower could I have possibly picked up on a nonturbo car when literally nothing else changed? Whatever performance differences there may have been from removing the AFM were clearly within the horsepower tolerances of any engine. And it's not like you can really feel some small change like 3 horsepower anyway. Remember that any system has an operational range of output.

Mod list:

street ported s4 motor with no s5 parts
T2 rotor housings
RB header, RB catback, straight midpipe
Pineapple racing auxiliary port sleeves

and here's a dyno sheet from this thread: https://www.rx7club.com/time-slips-dyno-128/dyno-348-horse-279-torque-%40-14-psi-rtek-2-1-a-746141/

stock s4 turbo AFM, Rtek 2.1

 

If you were getting less horsepower with a Megasquirt than with a stock ecu and an S-AFC, I'd definitely say it's not tuned as well as it could be. There shouldn't be a single point on that chart that shows the stock/safc making more power. I'm not even talking about the afm being in the system. If we want to talk about the afm, the best way to do it would be to do back to back runs on the MS equipped engine. One with an afm in place and the other without. That would be a valid comparison.

You aren't even making more power than a system that you couldn't adjust timing on so I will get picky enough to say that's not a very good comparison. Now to be fair I'm not saying you'd see a huge difference in power with one installed and without. It wouldn't be huge. It would be there though. However how much would be dependent on as you pointed out, how well the exhaust flowed. On a stock exhaust equipped car, you might not see any difference. The stock exhaust manifold only has a 2" outlet which is seriously choking the engine. Get a free flowing exhaust on the engine where it can breathe better and you'll see the afm have an influence.

Basically what I'm getting at is that "it depends". The afm may or may not have an effect on power. It depends what the rest of the system is like. I definitely wouldn't begin modding the car by worrying about the afm. There are far more important things to worry about. Do the exhaust first as well as other mods. Save the afm and/or ecu for last. The S3 afm on the GSL-SE is only about 2/3 the total area of the 2nd gen afm though so it will be a restriction much sooner.

 

I ran a 2.5" turbo exhaust. And I was trying to point out that the tune may not have been the best as it was not done by a rotary specialist. It was a long time ago man. If I could go back and do it again, I would use an rtek 2.1 , but that wasn't available at the time for the nonturbos.

The issue at hand is whether the flapper-type AFM holds the engine back, not whether a speed-density based standalone is superior to a stock ECU. Notice how the OP was suggesting a different type of AFM.

The point of the graph was to demonstrate that there will be no noticeable difference in every day performance from changing the airflow meter alone. Say in some situation you pick up as much as 5 horsepower on a nonturbo car and maybe 1.5 torque. So what? It's the curse of the nonturbo rotary owner or even the n/a Honda owner. After making all the easy gains, you begin to sweat every single additional horsepower not because it really matters for enjoying the car, but because it is some kind of scorecard or because you are simply running out of projects. On 99% of street 2nd gen n/a applications, is it worth the dollars and the hours of your life to switch away from the flapper type AFM? The answer is no. And it's not that big of a deal for moderate turbo setups either.

 

I adsolutely agree that for most applications it's a nonissue. Most people don't hit power levels high enough for it to be noticable. I actually run a stock air filter and box as I don't believe that a cone filter in the hot engine bay is beneficial and I used to have one installed.

The OP is mentioning 2 different afms. They are both flapper door style but the S3's is 4 sq in in and the S4 is 6 sq in. There's a difference there and the S3 is holding back power to a greater extent than the S4's is. The S5 cone style afm has less area than an S4 afm yet flows more due to it's smooth internal shape. The fact the cone is in the middle and air flows around it is irrelevant. Air flow is better. That's really the key. It's not so much how large is the opening but how well does air flow through it. Is there alot of turbulence? When there is laminar flow, which isn't going to happen through an afm, you can get tons of flow. Case in point being the restrictors on ALMS cars. They don't have plates but rather smooth venturis. The smallest point on the last rotary race car, the Courage C60 with the R20B 3 rotor installed, had a restrictor plate that was more or less equivalent to a 2-1/4" opening and it made about 450 hp. That's alot. To put that into perspective that's roughly the same total area as the S3 afm yet flows WAY more due to it's smooth profile.

This of course does prove that the afm shouldn't be much of a hinderance on a street engine with little overlap and relatively little power. It isn't a big deal. I am being picky when I say it does rob power. Technically it does. As you've pointed out it's not much and probably not enough for anyone to feel. Then again people remove the a/c and p/s all in the name of a few pounds of weight savings to "maybe" be faster so why not worry about that last few hp everywhere else too? The big problem is that most people try to start with the intake rather than the exhaust. 75% of your gains are going to come from the exhaust and that's where the greatest power per dollar spent is going to come from. I personally don't think anyone should worry about any part of the intake, not even the tb, until the exhaust is completely done engine to muffler.

 

Wow these are some great answers guys! In general I just think the hot-wire, mass air systems are better for driveability. If I recall correctly, they are much better at responding to differences in things like altitude, humidity, temperature and other atmospheric conditions, not to mention flowing better as the air doesn't have to move a mechanical linkage.

I'm doing full exhaust, and ignition, and I was just wondering where I could get a little more power on a relatively stock engine while at the same time increasing driveability (IE: not a map based efi where you have to retune it for any additional changes) So again its not so much the power I'm after (ok so my title didn't do a very good job at conveying that) but the adaptability and driveability that I would like to improve, which is also why I've considered designing my own maf based intake manifold as well.

 

https://www.rx7club.com/naturally-aspirated-performance-forum-220/s5-na-stock-port-dyno-830392/

s3 afm is smaller, so it might be a restriction

 

A good exhaust system will improve power and drivability so start there. Your stock intake systems full potential maxed out by someone who knows all the tricks and is using a flowbench is only about 10% greater than it's stock flow amount and that's it. A well designed full exhaust system can give you 35hp or so and tuning even more. The intake can flow enough to 200 rwhp so I wouldn't try to mess with something that isn't the main issue. Start with what needs it first and then worry about the small stuff. The intake is the small stuff!

 

I'd say squirtin' is an option if you are worried about the AFM.

But, maybe consider porting and a full RB exhaust. That would help a lot more.

I though someone said that the stock AFM was good to around 300 ish HP.

 

Well like I said I am currently doing full exhaust, but I guess the intake tract will be fine for my power goals...for now at least.

 

exactly what I've been saying. thank you K&N...

Altitude: factory barometic pressure sensor
temperature: two factory intake air temperature sensors, one by the manifold and one in the AFM
humidity: it changes the performance of the engine, but all the math and correction tables in the ECU account for changing atmospheric conditions.

The MAF sensor directly measures air mass (Rx-8). The volume flow sensor's output (FB, FC, Pre-Mk IV Supra) is used in conjuction with some combination of temperature and pressure (atmospheric, manifold) sensors to calculate mass air and engine load.

The MAF sensor is still superior and more modern. It's certainly simpler packaging. But the old Bosch L-Jetronic style of fuel injection (FB, FC, 280zx, etc) got the same job done, just in a different way.

 

Thanks, I did not realize the cars had barometric pressure sensors as well. I thought they were blind to density Mostly, I'm just trying to learn as much about as these 13bs as I can!

 

I believe that the 300 HP you are referring to is how much power you can be running before the AFM maxes out and then doesn't add any more fuel after that point of air flow.

 

No factory-produced RX-7 has ever had a MAF sensor. The 89-91 RX-7 had a linear-type AFM that somewhat looks like a MAF sensor. The FD RX-7 used a speed-density system that does not have an airflow meter of any type. The first rotary-powered car to have a MAF sensor was the RX-8.

The antiquated design of the stock ECU is more of a problem than the AFM. Most aftermarket replacement ECUs are of the speed-density type which do not use the AFM, so both problems are solved right there.

Nobody on this forum has successfully done this. Here is the most advanced project so far:
http://howto.globalvicinity.com/gv_w...=120&co=1&vi=1

The Bonneville 191mph record-setting 10AE was estimated at 390bhp. Nowadays, you would want to convert to a full standalone well before that power level anyhow, so any limitation imposed by the AFM is a moot point.

The stock RX-7 system does not account for humidity in its calculations. However, humidity makes little difference in air density. A MAF sensor does account for humidity when calculating airflow, so I guess that gives it a slight edge.

If you do not want to have a professional tune the engine, then simply stick to the stock AFM and ECU and focus on simple bolt-on power-producers such as the exhaust system. My convertible RX-7 still has the stock system because I don't want to deal with installing a better one, and the stock performance is just fine for me.

If you want max performance, you will need to install a standalone EMS and have it tuned by a pro. As for "drivability", I drove my racing partner's Wolf EMS-equipped non-turbo S4 RX-7 around a parking lot in all 5 gears without ever touching the gas pedal. The car accelerated evenly with my slightly-modified S4 TII, and it gets over 30mpg at cruise.

 

the afm/maf conversion is really popular in the german car world... i dont know why its easier for them, although cost is large and benefit is small

the old i'm getting the more fun it is to learn how to use the stock power better. especially at the track... (i used the stopwatch last time, results were surprising)

 

New RX-7 AFM: $900
Reman RX-7 AFM: $400
Reman RX-8 MAF: $140

A new replacement AFM or other outdated OEM component for a Porsche, BMW, Mercedes, etc., probably costs a small fortune, so it probably makes a lot of financial sense to slap on a cheapie Ford MAF sensor if they can get away with it. The Autobahn snobs are generally more reluctant to strip their expensive cars and install a standalone EMS, unlike the tightwads on this forum whose cars are usually worth less than a set of new wheels and tires.