On Tue, Nov 26, 2013 at 7:23 PM, James <jk_eaton@hotmail.com> wrote:

> I think having this data is great - I could pretty much write the
> progamming now, if I make a guess/assumption about the ISV's behavior at
> speeds above idle.   ( Which one could do, as the amount of air being
> admitted is only a small fraction of the total air intake when the
> waterboxer is working.)  Thanks to those who took the time to make the
> measurements.
>
> Yes, once you have the data in front of you, the coding becomes pretty
straightforward.  I agree, the behavior above idle probably won't matter
much, but I still want to be consistent with the stock system.

> I wonder if the jump to 55% duty cycle is really to compensate for lower
> available voltage while starting, or to aid starting itself?  I guess I can
> better rephrase my question as, does the ISV open more at higher voltages
> (i.e., is there any connection between the amount of air it lets in and the
> voltage applied to it?).  If the ISV were a motor, of course there would
> be, but I wonder if it is more like a pneumatic valve - and the pneumatic
> valves I'm familiar with don't open more with higher voltages, as long as
> the minimum voltage is met.  (I teach pneumatics as well, using German
> Festo equipment, and consult on pneumatic factory equipment.)  Pneumatic
> valves are rather 'digital' in operation - either they're on, or off, with
> no intermediate positions.  The vibraion the ISV produces reminds me of a
> Festo pneumatic valve in 'flutter' operation.
>

Good questions.  I've been assuming that it operates like a small motor
torquing against a spring, but it could very well be an on/off valve
design.  Has anyone taken one of these apart or seen a reference on how the
innards operate?  I think that the cold cranking duty cycle is set to
compensate for the voltage drop and is just targeting the fast idle speed
for cold engine running, but that's just a lot of thinking and assuming
rather than knowing.  Fortunately, we're heading right into the dead of
Winter, so we'll have the perfect opportunity to test and adjust.

>
> The PIC microcontrollers I teach all have 10 bit (or 1024 step) PWM, so
> having a fine control over the ISV wouldn't be a problem, if we wanted
> something with a finer control than the 256 step.  For a cleaner design and
> possibly more reliability I'd avoid a separate PWM controller, if I could.
>
> If there were several pwm signals being output or if the microcontroller
were required to juggle multiple critical timing operations, then a
separate dedicated pwm controller would be in order.  We only have two time
critical operations, pwm output and engine speed from the Hall sensor,
which are pretty easily handled on a single microcontroller.  You should
give it a stab using a PIC.

Brett

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