Date: Sun, 7 Jan 2007 19:54:07 -0800
Reply-To: jon <jon@KENNEKE.COM>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: jon <jon@KENNEKE.COM>
Subject: Re: Idle Stabilizer Control Unit Reverse Engineering
In-Reply-To: <6da579340701071802h43ddbe5fj419a389f939bf6c0@mail.gmail.com>
Content-Type: TEXT/PLAIN; charset=US-ASCII
John,
I'd be interested in an untouched malfunctioning unit. I want to drill
down to the exact reason why it failed.
I used to do "fiddly" analog design for oceanoprghic instruments, so
op-amp wierdness used to be a daily chore. On one instrument, we hand
tuned everything. It was a PITA. Every time we had a new tech., I had to
spend almost a week training them how to select the tuning components
based on components tolerances.
At first look, the counting seems to be accomplished with a F to V
converter. Wow.
It would be very interesting to know why in the waterboxer application a
seperate IS unit was needed. There are Digifant cars that don't have them.
That also raises the question: How do megasquirt equipped WBXs idle? Given
they use a whole different air intake scheme, that might be the answer.
I've got an o'scope, so it has been in my plans to take a look at the ISV
waveform. I'll let you know.
Thank you!
Jon Kenneke
829 Washington SW
Albany, OR 97321
On Sun, 7 Jan 2007, John Bange wrote:
> >
> > How many folks out there have a *bad* IS control unit? I'd love one to
> > verify if it's the transistor that went south.
>
>
> I have a couple toasted ones Frank Condelli was nice enough to send me last
> year for similar purposes. My initial plan was to desolder the parts off one
> of 'em and map the PC board traces, then add in the components to get a full
> circuit diagram. After a little poking around, though, I figured out that
> it's actually a very touchy hand-tuned analog computer. Outstanding design
> for its time (mid 80's), but seriously yesterday's news. What takes a bunch
> of tense interwoven op-amps to do in analog, only takes a 25 cent
> microcontroller to do digitally. The engineer that designed this circuit was
> a genius, figuring out a way to "count" revolutions per minute and keep it
> steady without actually having anything capable of COUNTING. I have a
> partly done design for a PIC based replacement, but employment issues have
> conspired to keep me away from completing it. The big missing piece was the
> valve itself and how it's operated. I have no scope to hook up to it to see
> how the idle controller opens and closes it. It reads constant voltage on a
> meter, so I assume it's using some sort of pulse width modulation, but I
> don't know. Perhaps someday I'll have the time and the oscilloscope I need
> to look into it.
>
> Anyway, if you want a dead idle controller, I can dig one up and mail it to
> you, if you like. Just email me your address and I'll try to find it in my
> garage. I'm supposed to send Steve Sandlin four big boxes of foreign car key
> blanks I don't need (USPS will be sorry they made those $4.85 flat rate
> boxes!), so I'll be at the post office anyway and I'd probably be able to
> get it out this week.
>
> --
> John Bange
> '90 Vanagon - "Geldsauger"
> "Staubkess bolt? I have one of those?"
>
| 
