[Diy_efi] trying to replace standard flap style afm with MAP sensor

[Lee M. Lemoine]

Damian,

My friend up here tried doing the same to his Talon, and basically the
short ver. of what we ran into was this:

The AFR you are modifying looks for the volume of air (mass air flow) 
entering the engine, whereas a MAP sensor is looking for the density
of air.   Two different units of measure.   your probably seeing the
vaccum bounce as it riches and leans out i'm sure, but i'm 99% sure
thats the problem, and i'm not sure how to rectify it to be honest....
(of course, without aftermarket EFI)

Anyone else see what i'm saying, or maybe have a little more time to
elaborate -- i'm gonna be late for work as it is lol!

On 7/20/05, Damian Long <bigcow_610 at yahoo.com.au> wrote:
> Hi, 
>   
> I am new to the efi digest. My first post! I eventually brought myself here
> after many nights of trial and error with no solution. I sense gurus live
> here 
>   
> What my problem is..... 
>   
> I have been modifying my vehicle and my efi system. I am fitting a piggy
> back, interceptor style computer to my standard efi computer. In doing this
> I also wished to do away with the flap style air flow meter on the intake. I
> was replacing this with a 3 bar Denso map sensor. I have discovered this is
> not a simple as first thought. 
>   
> The engine is a Mazda B6T 1.6l turbo with what I believe is a Nippondenso
> efi system. The flap style afm fitted to the vehicle is fairly standard and
> looks similar to those fitted to the Rx7 and miata. 
> I have a link at the bottom to a schematic of what I think the inside of the
> flow meter is. Well its what I have measured in resistances and from
> research on the net thats what I think is inside. So how I think it works
> is, there is a 12v supply coming in. A 100ohm resistor feeds a pin called Vc
> which is the 9v reference signal to the ECU. Pin Vs is the flow signal. This
> is a varying signal from 2 - 9v where 2v is no flow and 9v is 100% duty. the
> ecu takes a ratio of Vs/Vc to select injector duty. Makes more sense when
> you look at the picture. 
>   
> I measured the internal resistances across the pins and replaced them with
> standard resistors in the configuration shown in the picture. I used this in
> conjunction with the map sensor. The map sensor gives a 0-5v signal 0v at
> absolute zero and 5v at 3bar (ppl on here should know this anyhow). I then
> used the piggy back to trim this to the normal 2v-9v etc. 
>   
> Oh I also fitted 550cc injectors instead of the 340cc units. 
>   
> With the map sensor configuration the ECU does not produce any error codes.
> So I think as far as my knowledge goes that the ECU is accepting my
> modification. However the vehicle runs rich, very rich. With the piggy back
> computer I can force the cursor to stay on one map point. I can fully vary
> the output voltage (going into the ECU) from 0.6v through to 9v and the
> vehicle remains very rich at all voltages. It will not lean out. The plugs
> are wet and lots of soot. We have hooked up an oscilliscope to No. 1
> injector and the duty cycle is changing. You can tell by the sound of the
> car but whatever it is doing it just doesn't seem right. It hardly varies
> duty. Some other points 
>   
> * some time ago I intercepted the Vs signal with the piggy back, with stock
> afm and injectors and successfully altered injector duty 
>   
> * I did not fully check out the original afms workings before I butchered it
> for parts. Hence I can not go back to it to see it the car will alter duty
> properly with the afm (ie. my dodgey map mods are the cause) 
>   
> * my in depth knowledge of factory ECUs is fairly limited 
>   
> What I would like to know if anyone can is; 
>   
> * can you help at all? 
>   
> * can you enlighten me a little more on exactly how the flap afm worked and
> if my circuit should also work? 
>   
> * Anyone know how the stock ECU would alter duty? my guess was a look up
> table in which the ratio of Vs to the reference voltage Vc is looked up and
> an injector duty is selected. If this is the case then shouldn't my circuit
> work?  
>   
> any help at all is appreciated as I am at the point of giving up. 
>   
> thank you 
>   
> http://i7.photobucket.com/albums/y253/Jimbobtx3/ECU%20Code/bff1ae98.jpg
> 
> Send instant messages to your online friends http://au.messenger.yahoo.com 
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> http://lists.diy-efi.org/mailman/listinfo/diy_efi
> 
> 
> 


-- 
Sincerely,

Lee M. Lemoine
http://www.turbochargedsoul.com/