Modifying Rb20 Afm For Greater Flow

[CEF11E]

Ok.

As we all know the all nissan AFMs have a limit of air flow they can measure before they reach their maximum voltage 5 to 5.1 Volts.

For an RB20 that means you are limited to 190RWKW or there abouts before you need to upgrade. this usually means the Z32 path as the Z32 has greater resolution.

as i am aware RB20, RB25 and Z32 AFMS are all 80mm diameter and the only difference being their resolution

However lets say you wanted to go a different path and rather than increase the resolution with a Z32 AFM you increase the diameter of your current AFM.

Reading one of SK's other threads. and kind of working from there.

If a 90mm hollow tube flows 26% more than a 80mm hollow tube due to the diameter,

if we could directly translate that to an AFM,

-if a RB20 80mm AFM provides enough MEASURED airflow to produce 190RWKW

-Will a RB20 AFM modified to a 90mm diameter flow enough to produce a shade under 240RWKW and still be measureable?

It is a given that the ECU would have to be tuned, just as it would if you swap to a Z32 AFM.

Here is an RB20 AFM taken apart. I am going to buy some 90mm PVC

I am going to run the following tests:

Measure the output voltage of my afm at different RPM all the way to redline and map the voltages.

Then put in the modified 90mm AFM and peform the same test and map the voltages.

to do this my ECU will be running EXTREMLY RICH to avoid engine damage.

If you have any experience in this area please feel free to reply...

I hope all this makes sense and my backyard science has some merit to it.

[Lazy-Bastard]

watching with interest

[T04GTR]

thats an interesting idea. i used to work for subaru and we had some mrt cars come in and they seemed to have the same idea.

a slightly bigger alloy pipe with the maf sensor in it.

also while i was at holden the ss inductions guyes did the same thing to the gen3 engines with their modified afm. they would remove the sensor and bore them out.

as a result it would make the engine run leaner across the board.and therefor more power.

[rob82]

The ratio of diameter to airflow isn't just as simple as what SK said if you increase the diameter from 80mm to 90mm thats not exactly 27% more flow as its area is 27% larger, its much more complex than that, this is why MAF tranfer tables are expotential not linear. That said your not dealing with a frequency based MAF so that approximation should get you somewhere close.

I have had alot of expereince tuning different MAF's for different cars including nissans. The problem with nissans is that there is not enough serial data you can compare you MAF transfer table to. This is because the maths behind the nissan computer is based on injector pulse width not desired air fuel ratio.

To get an idea of where the MAF table is going would be to watch your trim numbers in closed loop and modify the table until there is zero trim which is 100% on the nissan consult. Please understand that this is a big job for a chip tuner. You have to weigh up if it is worth the extra money spent on the dyno just to tune your experimental MAF or go with something that will just work like the Z32.

[blackrb20silvia]

still better to have tried and failed then have never tried at all

i for 1 am interested as im making around the 190rwkw with standard ecu, injectors and AFM

if by any cahance there is a gain for the price of a 90mm pvc pipe

ill have a crack

[rob82]

Hmmmm 5hours of dyno tuning at $200/hr for a 20kw increase in peak power while losing low to medium load drivability and perhaps usable torque, possible shit idle stability(due to loss in low flow resolution) and more common stall events. Thats everything 1 want in a $5 dollar PVC pipe.

I dont mean to be pessimistic but there is a reason that car manufactureres spend millions upon millions on designing their sensors. There is no reason at all that the standard ECU isn't up to the job its just a matter of using the right parts for the job and that would be the Z32 MAF. Your probly only looking at 1-2 hours on the dyno to tune a Z32 MAF.

[CEF11E]

Here is the finished [EXPERIMENTAL] afm.

I have since discovered the afm om my car to be a diameter of 70mm NOT 80mm.

The PVC pipe i used is 84mm.

Here is what I did.

connect a multimeter to my AFM output signal and drive the car monitoring the AFM voltage at certan load points in 3rd gear all the way up to 7000 rpm. noted the values and then repeated the tests with the 84mm AFM.

Results are as follows.

idle ofcorse was not stable as the car was not tuned for this afm, however it would hold an idle.

at 3000 rpm the afm showed 1.0v difference

at 4000 1.0v

at 5000 1.2v

at 6000 1.1v

at 7000 0.8v

maxing at 4.1v where as my standard AFM reaches 5.1v at 7000ish.

I do not have a wide band 02 sensor but the car is running at ratio of 10:1 so I wasn't concerned about enging damage.

the conclusion i have come to is, It's possible to increase the diameter of the AFM to provide I higher level of air flow.

Ovcourse you would not do this unless you were going to tune the car on a dyno.

My reasoning for doing this is as follows.

I am building an R31 for drift and want to keep it cheap but get some decent power out of it.

I have a freind that is a tuner and has chipped lots of ECUs for me before. he also has a wideband sensor so dyno time and cost is not a problem.

so having said that the AFM WILL be going on my RB20det powered R31 and i will be aiming to get 200rwkw out of it with out maxing it out

[midnight]

what happens when you upgrade your induction system in the future and are flowing way more air than you currently are at redline?

are you just going to estimate all values from 4.1-5.1 volts on you VQ table?

[CEF11E]

I have no experience tuning roms so i cannot comment on tables.

However the guy chipping the ecu for me has done diameter changes on AFM's before so i just have to take his experience as proof that it will work and he knows what he is doing.

the car the AFM was tested on is my daily driver and is not going to use the AFM it was purely for testing.

[midnight]

you might want to take a look at autospeed

http://autospeed.drive.com.au/cms/A_2632/article.html

article about doing the same thing.

[paulr33]

you can make 190rwkw on stock afm and injectors and ecu

[CEF11E]

you might want to take a look at autospeed

http://autospeed.drive.com.au/cms/A_2632/article.html

article about doing the same thing.

<{POST_SNAPBACK}>

I would love to read it but i dont have autospeed membership

[ebola]

I dont mean to be pessimistic but there is a reason that car manufactureres spend millions upon millions on designing their sensors.  There is no reason at all that the standard ECU isn't up to the job its just a matter of using the right parts for the job and that would be the Z32 MAF.  Your probly only looking at 1-2 hours on the dyno to tune a Z32 MAF.

<{POST_SNAPBACK}>

Manufacturers build all sorts of things into their products. Including tolerances...

Manufactures always have conflicting specs, too. The skyline is a passenger car, not a pure sports car. it does not come with 10kg/mm springs, as a concession to passenger comfort.

rob82 is right that the tune wont be as good as with a stock (or z32) item..., but if it gets you to the 200rwKw (slightly cheaper), congratulations

CEF11E, keep it up. This is the kind of mod that I would make, but its a whole lot easier when I know the end from the beginning. tuning the ecu (and related difficulties)...

So, what have you already done to the car to make it require this?

[Oosh]

Autospeed recently had an article on this, works fine, so long as you make the appropriate tuning adjustments.

[midnight]

dude, I already posted the autospeed link ...

[pnblight]

Hmmmm 5hours of dyno tuning at $200/hr for a 20kw increase in peak power while losing low to medium load drivability and perhaps usable torque, possible shit idle stability(due to loss in low flow resolution) and more common stall events.  Thats everything 1 want in a $5 dollar PVC pipe. 

I dont mean to be pessimistic but there is a reason that car manufactureres spend millions upon millions on designing their sensors.  There is no reason at all that the standard ECU isn't up to the job its just a matter of using the right parts for the job and that would be the Z32 MAF.  Your probly only looking at 1-2 hours on the dyno to tune a Z32 MAF.

<{POST_SNAPBACK}>

sorry to be like this but your wrong

15 years as an instrument tech

the afms meter works on air speed (cooling) so as long as you keep it near the centre of the pipe it will work fine. There nothing special about the z32 afm meter and if your running a pfc you can install a custom ramp graph for the afm meter and will not even need to adjust the maps(spark and fuel). Good to see someone willing to try some thing and not just be a sheep.

my 2c

[Oosh]

dude, I already posted the autospeed link ...

<{POST_SNAPBACK}>

Oops, that's what i get for skim reading,

[CEF11E]

So, what have you already done to the car to make it require this?

<{POST_SNAPBACK}>

Nothing as yet. I am slowly obtaining all the bits to build a dedicated car for drift.

I was sick of using my daily driver A31 Cefiro as a drift car. I was always breaking something on the track and having tien's in the car on the road didnt agree with me. plus i got sick of all the unwanted attention the car would get on the road.

So I bought a daily driver and decided to slowly prep an R31 for drift. I havent even bought the car yet. All I have is engine and a few boxes of bits.

The reason i did this is i wanted to see if it was possible. The R31 is going to have the following bits:

R31 sedan

RB20det with following support mods

RX7 injectors

Modified AFM

R34 Gtt fuel pump

Supra FMIC with waterspray kit

RB25 turbo with to4 V trim wheel and bored out compressor cover

remapped ECU

+ GTR clutch and pressure plate

JIC folly ajustable coilovers and other suspension parts

Welded or mini spool diff out of the automatic for closer ratio.

This combo should get me to 200 rwkw and with a stripped interior i should be able to get the dry weight down to a shade under 1200Kgs. giving me a cheap and semi compedative drift machine. and providing i can pick up an R31 shell in reasonable condition for less than $1,000. I have a nice weekend toy for less than $5000.

[Cubes]

Looks quite good.

I've heard of ppls doing something similiar with the Z32 sensor in a 90mm id pipe. The Z32 is the sensor of choice over the Q45/VH41 or RB20/25's due to the lower voltage reading for a given airflow.

[MaTBoY]

Good thinking mate

I did something similar to my old R31 (N/A). It definately helped give an increase to torque and power. I didn't have any idling problems either. You should take it further and get a larger throttlebody 2. and if you have a dremel thingy you could even try to smooth out the insides of your manifold... It's a good excercise if your bored one weekend