Haberdasher

Here's one for all the expert troubleshooters:

I'm running the AFC 'Hack' with all recommended parts. Settings are -40% across the board with the throttle switch over point at 71%/72% (I know it doesn't matter). I'm using a Tial wastegate for boost control, it has a green 0.5 BAR spring in it. The AFC is wired correctly. This is on a PR4 OBD0 ECU.

The problem is that I'm getting a CEL code 3 (MAP Sensor) right at max boost. I'm getting an indicated 8PSI on my Autometer /vac/boost gauge. The car goes into limp home mode, full rich which almost stall the car. I can clear it by killing and then restarting the engine.

I've tried 2 map sensors to rule that out. I've tried setting the AFC to -43% to see if I can knock enough signal off from the MAP sensor. Nothing has worked. I ran the problem past Liam821 and he was stumped too.

Anyone have any ideas on what the problem could be? Anyone with similar experiences?

Thanks in advance for the help

0xenocron

Dave, when am I going to be able to convert you to TurboEdit?

Is this something that just started or has this been happening since you switched to the Hack?

Maybe J. Davis while see this and chime in...

EE_Chris

What is the voltage that the ECU is seeing at max boost?

Haberdasher

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by EE_Chris &raquo;</TD></TR><TR><TD CLASS="quote">What is the voltage that the ECU is seeing at max boost?</TD></TR></TABLE>

On the MAP sensor signal wire? I don't know, I've been thinking of testing that. I'll try it.

I'm going to switch to TurboEdit but I only have $155.02 to my name (not kidding). I have a lead on a couple Software Engineer jobs so maybe that will change soon.

This has been happening since I switched but at first I was confident I could correct it. It's been 2 months or so and I'm about to give up and just live with it until I get TurboEdit.

Iceman05

idk about a code 3, but im getting a code 13 (BARO) from the hack. I hate honda MAPs! If anyone can help, id appreciate it!

EE_Chris

The MAP and Baro Sensor share the same output characteristics. When you turn the key 'on', the ECU does some sensor checks and compares the MAP and Baro. Since you have a VAFC installed, which has changed the MAP value during the key-on, engine-off time, you get this code.

TurboBumblBee

i was running into the same problem.. i had it wired correctly, and setup as its supposed to be and occasionally i would get CEL 3 not all the time but sometime.. something i did was put 3 check valves on the map sensor to bleed off the boost and it fixed the problem, ive heard more check vavles the better tho.. im also running the afc at -30 and the car does good.. ive done a little street tuning with it and drive it work sometimes and the car is nice.. waiting on the $$ so i can get it tuned on a dyno w/ uberdata.. remember every car is different thats why the hack works w/ some cars and some cars it doesnt!

*edit*
after reading this thread and having some free time today i wrote down all the settings on my vafc and initialized it so i could try to setup the hack as described here http://www.thedropshop.tv/vafc.htm

i set it at -35% across the board on hi and lo, throttle at 98/99, vtec in5k/out5.5k took the car out for a nice drive on some back roads so i could be able to adjust anything if needed and not worry bout other cars and it did fine.. infact i think its better then the previous setup.

i did get a CEL for 43, 1, and 41.. i have been getting those at random times ive come to the conclusion after going through a helms i need to replace the o2 sensor on the downpipe and this will fix all the codes.. 43 is directly related w/ the o2 sensor going bad.. but i did turn off the car, and start it agian and the code was gone, its intermittent so im ok till it does it all the time and ill replace the o2 sensor


Modified by Boostfedb18c1 at 2:01 PM 5/30/2004

M.A.R.C.

that doesn't make sense because the hack worked fine for me with the same setup. recheck the wiring on the vafc

Haberdasher

OK, ran the test.

The voltage going into the AFC is 4.4 volts at ~8PSI

The voltage after the AFC, going to the ECU is 3.2 volts at ~8PSI

So it would seem I'm getting about a -27% correction even though the AFC is set at -40% across the board.

What should the the voltage be?

Haberdasher

Anyone? Anyone? Bueller? Anyone?

Joseph Davis

Your math in re the voltage correction is accurate. You are getting -27% cut when you have the AFC set to -40%. Verify your double ground for the AFC is not damaged, and try another AFC to test if yours is damaged.

There's no way what you're experiencing will happen if the system were working AT ALL the way it should. Since you are OBD0, your MAP sensor is going to return a maximum of 4.5x volts to the AFC (or ECU) when it reaches it's max of 9.25 psi.

S2KP

for some reason. I been running the afc hack @ 15psi with no light until top of fourth gear.

swlabhot

Make sure you have setup your AFC right in the 'ETC' menu - ie., 6in 6out etc...and all the rest of it.

Importracer_001

I'm also using the AFC hack on a B16A1 ODBO
I'm gonna try to see if I can hit 10 psi with teh hack
BUt I've been hitting 8-9psi without any problems

USDM 4G VTEC

since u are OBDO
what MAP are u using?
the one on the firewall
or the one on the TB mounted one
???

i would try to find a TB mounted MAP and see if that one will do it to ya

i am pushing 9.5 psi on mine and i havent see a CEL

turbo89civicb18hatch

set the map setting to 6 in 6 out

gotboost95

What's the AFC reading for pressure on the screen. There is a conversion

web site on the internet to change that to PSI you can also check it that way.

USDM 4G VTEC

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by gotboost95 &raquo;</TD></TR><TR><TD CLASS="quote">What's the AFC reading for pressure on the screen. There is a conversion

web site on the internet to change that to PSI you can also check it that way.</TD></TR></TABLE>

http://www.convert-me.com/en/convert/pressure

Haberdasher

At idle with 18in hg I'm getting 1.2 volts into the AFC and .97 volts to the ECU with an indicated -545mm hg on the AFC.

This means that there is a -19% correction here. Why would it be -19% here but -27% at ~8PSI?

At ~8PSI I'm getting an indicated +.4kg cm^2 on the AFC.

The Pressure sensor in the etc. menu is set to 6 in 6 out

I double check the double ground wires and they are well connected (soldered).

I have a GSR throttle body and throttle body mounted MAP sensor on the car. It's what I've been using all this time. I have also tried the stock non-TB mounted MAP sensor with the same results I've always had.

I do not have access to another AFC to test.

I did find something that was interesting in my test drive today. In one of the test runs I went up to ~8PSI but didn't hold it there for very long. This run did not trigger the CEL or limp home mode. So it would seem that it only triggers if I hold it at 8PSI for longer than a few seconds.

Any who...I'm 'bout ready to roll it off a cliff...


Modified by Haberdasher at 5:04 AM 6/5/2004

EE_Chris

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by CivicRyda2k &raquo;</TD></TR><TR><TD CLASS="quote">that doesn't make sense because the hack worked fine for me with the same setup. recheck the wiring on the vafc</TD></TR></TABLE>

I hear ya'. Some people get it, some don't. But the only common denominator in the equation has always been the VAFC when the Baro code came up.

True manifold pressure isn't found by just the MAP alone.

Haberdasher

What should the rest of the etc. settings be? I noticed there was some kind of Cr% in there.

What would happen if I put my Missing Link back on the MAP sensor?


Modified by Haberdasher at 11:16 PM 6/5/2004


Modified by Haberdasher at 4:10 AM 6/6/2004

Joseph Davis

Wierd electrical **** happens when circuit resistances are off... bad connections to grounds, sensors, actuators, yes? Your AFC wouldn't operate correctly in that case, yes?

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by J. Davis &raquo;</TD></TR><TR><TD CLASS="quote">Verify your double ground for the AFC is not damaged, and try another AFC to test if yours is damaged. </TD></TR></TABLE>

Quit ******* around and start poking around with an ohm meter. Do it with power OFF, preferrably the battery disconnected, as power flowing along a wire throws off resistance measurements.

kpt4321

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Turbo_Siii &raquo;</TD></TR><TR><TD CLASS="quote">for some reason. I been running the afc hack @ 15psi with no light until top of fourth gear. </TD></TR></TABLE>

If you're running any more than about a -40% correction, you will never get a CE light no matter how much boost you run on the stock sensor.

However, DON'T DO WHAT YOU ARE DOING! Once you hit 10-11 psi, the stock map sensor is maxxed out. Past there, the ECU is not going to increase your fuel level, AT ALL!

You're lucky you still have an engine.

kpt4321

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by swlabhot &raquo;</TD></TR><TR><TD CLASS="quote">Make sure you have setup your AFC right in the 'ETC' menu - ie., 6in 6out etc...and all the rest of it.</TD></TR></TABLE>

Why 6 in 6 out?

kpt4321

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by gotboost95 &raquo;</TD></TR><TR><TD CLASS="quote">What's the AFC reading for pressure on the screen. There is a conversion

web site on the internet to change that to PSI you can also check it that way.</TD></TR></TABLE>

This won't help you with anything.

The pressure reading on the screen of the SAFC is totally banjaxed, because the Honda MAP sensor isn't scaled the same way as the 2BAR the SAFC thinks you are running.

For example, it will tell you you're at like 1.0 BAR at full throttle, even if you're... not.