Do all ECUs go into open loop at WOT?
George Reiswig
I have a '90 Isuzu 2.6L engine, and at or near WOT, there's a flat
spot in the power. This is common for this engine. If you back off the
accelerator a little, you can distinctly feel the acceleration improve.
One EFI shop that I spoke to said that this is a fairly common
affliction for many EFI motors. They claimed that, at WOT, the computer
goes into open loop mode, and bypasses the O2 sensor, so the mixture is
no longer set to stoichiometric. As a result, it goes rich.
Does that sound right? If so, it seems to me that it would be a
fairly easy fix. It seems to me that the TPS is what tells the computer
that the throttle is open all the way, and that jumpering or severing
(can't remember which right now) the appropriate wires on the TPS would
mean that the ECU never saw a WOT signal, so it would continue to
compensate.
Thoughts?
Also, why would the industry force open loop mode at WOT? Is this
just a cheapy way to prevent pinging? Thanks!
--
George S. Reiswig
Obviously, my opinions should not be construed as being anything like
those of my gracious employer, Intel. Their opinions are probably
better.
Advertisers!
It is illegal to use this e-mail address for unsolicited commercial
e-mail per United States Code Title 47 Sec. 227. I reserve the right
to assess a US $500 charge for reviewing & deleting each unsolicited
commercial e-mail.
Sending unsolicited commercial e-mail to my email address denotes
acceptance of these terms. Postings from me to UseNet neither grants
my consent to receive commercial emailings nor is intended to solicit
commercial email.
Mike Fahrion
George Reiswig wrote in message <35A234DB...@ccm.jf.intel.com>...
>Hi, all.
> I have a '90 Isuzu 2.6L engine, and at or near WOT, there's a flat
snip
> One EFI shop that I spoke to said that this is a fairly common
>affliction for many EFI motors. They claimed that, at WOT, the computer
>goes into open loop mode, and bypasses the O2 sensor, so the mixture is
>no longer set to stoichiometric. As a result, it goes rich.
> Does that sound right? If so, it seems to me that it would be a
>fairly easy fix. It seems to me that the TPS is what tells the computer
>that the throttle is open all the way, and that jumpering or severing
>(can't remember which right now) the appropriate wires on the TPS would
>mean that the ECU never saw a WOT signal, so it would continue to
>compensate.
> Thoughts?
> Also, why would the industry force open loop mode at WOT? Is this
>just a cheapy way to prevent pinging? Thanks!
As far as I know, all mfg's currently go open loop at WOT. The way I
understand it is that the EPA does not require WOT emissions tests, and this
way they can optimize for power instead of emissions. The fuel map being
used should still be compensated for temp and MAP, I would think that
usually the opposite of what you're experiencing is true. There also may be
some difficulty with closed loop engine management at WOT, although I think
I recall hearing that some future PCM's will stay closed loop at WOT.
A couple methods of fixing, reprogram the fuel maps at WOT or adjust the
fuel pressure. I like the adjustable pressure fix, when closed loop the O2
feedback keeps things in line (if you don't get carried away!), when open
loop, you can tune for power with slight changes in pressure.
Any EFI shop should be able to help you with the fuel pressure mods.
Good Luck
-mike
Mark Kovalsky
George Reiswig <george_...@ccm.jf.intel.com> wrote in article
<35A234DB...@ccm.jf.intel.com>...
> Hi, all.
> I have a '90 Isuzu 2.6L engine, and at or near WOT, there's a flat
> spot in the power. This is common for this engine. If you back off the
> accelerator a little, you can distinctly feel the acceleration improve.
<SNIP>
Something is wrong with it. I can't tell what, but something isn't right.
> Also, why would the industry force open loop mode at WOT? Is this
> just a cheapy way to prevent pinging? Thanks!
All cars go open loop at WOT because best power is made running a
little rich. If you run at stoiciometry at WOT you will lose power.
You can't use the O2 to run rich, either. All it can read is rich or lean,
it can't tell how much. Running 1% rich and 100% rich give the same
reading on the O2. There are much more expensive O2 sensors that
can read this, but they are much too expensive to use.
--
Mark
'30 Ford Model 'A'
'59 Edsel Corsair
'94 Ford Club Wagon w/7.3L Hypermax Turbo Diesel
'94 Sportsmen's 37' Travel Trailer
'98 Ford Taurus
The views expressed above are mine, and mine alone.
P. J. Remner
In a previous article, mfah...@bb-elec.com (Mike Fahrion) says:
>usually the opposite of what you're experiencing is true. There also may be
>some difficulty with closed loop engine management at WOT, although I think
>I recall hearing that some future PCM's will stay closed loop at WOT.
Correct. In closed-loop the computer relies on the O2 sensor to
"perfect" the fuel mixture. Unfortunately, current O2 sensors
can only tell the computer if the mixure is rich or lean, not
*how* rich or *how* lean. This is okay for light-load driving,
where it's good to be "stoichiometric". Under load, and WOT is
load! - you will want a ratio of roughly 12:1 or 13:1 air:fuel,
a bit richer than the perfect 14.7:1. So, computer has to go into
open loop, because all the O2 sensor is saying is "rich".
There is a "linear" O2 sensor being developed that is accurate
from 9:1 to 18:1 (or somesuch) air:fuel ratios, making a new
generation of lean-burn and constant-closed-loop control engines
possible, making the world a cleaner, more efficient, and more
powerful place to live, and all that other fairy-tale jazz.
--
I walked out to my car. I unlocked the door, smoothly opened it and sat into
the rich velour seat. I buckled the belt tightly and closed the door with a
solid satisfying "click". I twisted the key and the 429 roared and throbbed
to life, and I ratcheted into First and... then I woke up for real.