Date:         Sat, 16 Nov 2013 16:20:07 -0800
Reply-To:     Brett Ne <brettn777@GMAIL.COM>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         Brett Ne <brettn777@GMAIL.COM>
Subject:      Re: Arduino and Vanagons
In-Reply-To:  <BAY405-EAS122ACAD2A442C187D599EF0A0FA0@phx.gbl>
Content-Type: text/plain; charset=ISO-8859-1

On Sat, Nov 16, 2013 at 9:14 AM, Dennis Haynes <d23haynes57@hotmail.com>wrote:

> Recently I installed a rebuilt Idle controller from GoWesty. I have to say
> they got it right. It works just as it should. The price in my opinion was
> reasonable for how well it is working. Time will tell if it lasts.
> For the questions below see my comments.
>
> Nice!  GoWesty definitely does nice work.  I'm hoping that our design
results in an ICU that is much cheaper and more durable than $190 (and
remember, that jumps to $290 if your old one isn't rebuildable).  I
certainly don't have an issue with Gowesty making a nice profit; I'm hoping
that we can design a solid replacement that costs $15 to make in quantity
and GoWesty makes and sells them at $60.  I'd buy one and happily give them
$45 for the convenience of not having to cobble one together myself.




> So, here's what we need to know:
> * The PWM duty cycle being feed to the idle air valve when:
>      --the engine is cold, the starter is engaged, and the coil is shorted
> I don't think that coil should ever get shorted. There has to be a map for
> cold start. There has to consideration for even warm or hot start all the
> way down to -40F.
>

By shorting the coil, I mean either short the high tension to the block or
else disconnect the primary power to keep the engine from firing, just as
is done when checking compression.  I just want some idea of what the duty
cycle is during a cold start.  Too much or too little air could throw the
air/fuel ratio out of the combustion range.  I don't think that the O2
sensor operates on a cold start, so the fuel injectors are just blindly
throwing a prescribed amount of fuel into the engine assuming that the idle
air valve is acting properly & allowing the right amount of air flow.


>      --the engine is cold & running
> Maintain 1,000 rpm up to one minute if coolant under 100F.
>      --the engine is fully warm & running
> Purely feed back controlled 850 rpm.
>      --the engine is fully warm & running with PS operating
> Add no more than 100-150 rpm
>      --the engine is fully warm & running with AC on
> Same as Power steering critical for automatics.
>

Good info.


>
> *  The current draw from the idle air valve at a known duty cycle.(or feed
> it straight 12v and measure the current draw.)
> 12 volts on that valve for any length of time will smoke it.
>

I assumed that someone with a multi-meter and the knowledge on how to use
it would only have it hooked up for the few seconds it would take to get a
current reading.  Perhaps I shouldn't have.  I need to know the current
draw in order to choose an appropriate transistor to drive  it.


>
> *  What is the voltage at the coolant temperature sensor when cold and when
> 160 F.
> Why 160F? The engine goes closed loop around 110-120F.
>

Semi-arbitrary number.  I figured by this temp the engine is pretty warmed
up & operating normally.  By measuring the voltage from the sensor the
microcontroller can at least distinguish between a cold start and a warm
start.  it'd probably be better to get temperature and voltage readings
both when it's cold and when it is fully warmed up.  From that we can write
an equation that will give us the temperature for any given voltage reading
coming from the sensor.


The target rpm's for the various operating conditions is helpful, but I'd
still like to have the duty cycle readings to go with them.  If the warm
idle duty cycle is 35% and jumps up to 85% when AC is turned on, I would be
much more aggressive with the reaction rates in the formulas than if the
same figures were 35% and 42%.  The readings would also give is a much
better idea on what should be happening outside of the idle range.  Should
the valve be closed during starting?  Should the valve be closed when the
van is going down the road at 3200 rpm?  Should the valve be fully open at
full throttle?  I don't know.  But the PWM readings under the conditions
listed  should give us a pretty good outline of what the ICU is supposed to
be doing under the varying conditions.





--
Brett in Portland, OR
"Albert" '82 VanaFox I4 Riviera