Boost Controllers: Risks and returns: Risks (long)

["Geoff Jenkins" <jenkinsg@pilot.msu.edu>]

Since the boost controller thread started, I've received a large number of
questions as to the safety and effectivenes of these devices. In order to
answer some of these, I've put together two posts (the other should be going
out simultaneously) which reflect my interpretation and opinions of the
risks and returns which these devices offer.

================
How Risky Is It?
================

Three significant concerns have been raised over the use of boost
controllers on our cars (though they may not be _real_ problems, or indeed
there may be other issues which have not yet been brought up). The issues
are Altitude, Heat and the ECU's response to Knocking.

Altitude:
---------
Problem: At higher altitudes, where the air pressure is lower, there is a
risk that a "dumb" boost controller will attempt to make up for the lack of
ambient air pressure by piling on the boost (i.e. it will try to maintain a
pressure reading without considering the drop in _outside_ pressure). This,
naturally, will lead to risks of running unacceptably high boost, or of
overspinning the turbo.

Resolution: The straight HKS "EVC IV" _doesn't_ deal with this issue. The
"EVC IV atm" allegedly _will_, but I am informed that it is not yet on sale.
If I was living in CO, I suspect I would wait for it to be available. Here
in Michigan, it doesn't seem like a worry.

Heat:
-----
Problem: If you're running high boost under high load in a hot climate, the
air intake temperature may become "too high" (note: this is a value which
remains unknown at this time), at which point the ECU will attempt to dial
back the boost. The EVC, however, will ignore this and keep responding to
the driver's right foot. The consequences of overly hot air entering the
engine include detonation, with all its expensive consequences.

Resolution: So far, _in_my_environment_ and _with_my_driving_style_ and
having maintained a relatively low boost level (+ 1.25 bar), this does not
appear to be a problem. The car has been taken to the track (prior to the
EVC) and after 45 minutes of lapping in 80+ degree weather, boost had not
been perceptibly reduced by the conventional ECU. That would indicate that
the "too high" temperature is _very_ high.

Further methods of clarifying this could involve an intercooler-exit air
sensor, warning the driver to lift off when the temperature exceeded X
degrees. IMO, this is either: (a) not very relevant, if the "X" temperature
is drastically high or (b) a very sensible mod for all high-boost cars, if
the "X" temperature is easily reached.

An alternative might perhaps involve a larger intercooler, hence delivering
cooler air for any given boost level. My own solution is somewhat simpler.
Because of the EVC's effect on the car's reflexes, a boost level which was
previously merely "OK" now supplies ample response, so I simply keep the
boost lower than I otherwise would have wanted. Lower boost -> lower heat,
all things being equal.

Knocking:
---------
Problem: When the engine knocks, the ECU attempts to both retard timing and
cut the boost. The boost cutting signal will be ignored by the EVC. If the
driver attempts to maintain full throttle with the turbo delivering high
boost while the engine is fighting against knocking, at the very least,
temperatures will climb dramatically.

Resolution: Currently, there seems to no technical fix to this. (a) You have
to use good gas and (b) if you hear knocking, back off.

In the longer term, there are two additional answers. In the near future, a
rework of the ECU code may be available (according to Scott Mo.) which would
allow the "Check Engine" light to come on as a warning that the ECU was
attempting to back off. Secondly, it _should_ be possible to isolate the
signal which the ECU sends to the WGFV to ask it to open up the wastegate in
an emergency, and use that signal to switch the EVC automatically from its
"high" boost mode to "low" boost. A driver who ignored such warnings would
have nobody but himself to blame.

For myself, I'm _concerned_ but not worried about this issue. However, I
probably _will_ obtain the reworked ECU code when it's available, and
_would_ be interested in the signal-switching solution, though it lies far
beyond my technical scope to implement.

So... How Risky Is It?
======================

If you encounter situations where the ECU would be attempting to dial back
the boost: High Altitude, Overheating, Knocking, then the EVC's failure to
adjust will pose a risk. To answer Frank Amoroso's original question: No, it
does not retain all of the factory safeguards.

If you use good gas, if you don't reach the "magic temperature" and if you
live in the flatlands, the odds are that you aren't going to encounter
enough risky situations to make it a serious threat, but that's ONLY if you
don't run excessive boost, and even then, a _somewhat_ higher risk potential
will now exist.

To my mind, the moment I start to tweak a car at all, I have to consider the
balance of risk vs. reward (which I'll come to in an accompanying post).
Given my driving and my estimation of the risks, I view the rewards to be
more than sufficient to justify installing an EVC in my car.

Going up the Rockies in a fully laden car at max boost on a hot day with
dodgy gas, the EVC will add to the risk, but there's no good in pretending
that under those circumstances the EVC is responsible for all of the risks
that are present.

=====================

Geoff
'94 S4 with HKS EVC IV