Date: Wed, 4 Dec 2013 20:07:02 -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: <CAHbJSdVR27Eh+6pFfgy5qj0dw86CEdGgiXo04zN+acrhNKVhLg@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Yah, whenever I take a measurement with a microcontroller, I take several
readings and average them. In this case, I measure the time in
milliseconds for 8 pulses and then calculate the engine speed. By
unstable, I mean the rpms were mostly jumping around 1300-1800 and
occasionally over 2k. My ear and the tach on the dash agree that the speed
was pretty steady at about 1100.
I drug out my laptop & USB scope today and got a quick reading. Here is a
link to a screenshot:
https://dl.dropboxusercontent.com/u/4930890/testvan1.bmp
Kind of a hybrid square/sawtooth wave form. The peak voltage is 2.82 v,
while most of the flat part is around 2.4v. For the Arduino, that puts the
entire peak region below the minimum 3v required to guarantee a positive
reading. I'll probably have to add an op-amp to buffer and raise the
voltage a bit. I was in a hurry & now wished I had stretched out the time
scale to see the noise on the peak region better. If that noise is right
near the pin's high/low voltage reading cutoff, it could be causing
multiple readings and throwing my measurements way off. If I run a voltage
doubler using an op-amp(with high rail of 5v), the entire peak would be
over 4v, and then trigger on the drop to zero, I should get a pretty clean
signal to work from.
The duty cycle is at 70%, so the shorter off cycle corresponds with the
smaller gaps in the slotted distributor shield thingy, which lends credence
to
Ryan's remark that most Hall sensors are open drain. That means that the
Hall sensor connects the circuit to ground (drain) when in the presence of
a magnetic field, but opens the circuit (disconnects it) if there is no
magnetic field. The electronic ignition feeds it a small amount of current
at 5v, and when the slot rotates near the Hall sensor & exposes it to the
magnetic field, the sensor connects that tiny current to ground, dropping
the voltage level to zero.
I'll try adding an op-amp to the circuit in the next couple of days & see
if I can get a reliable tachometer out of it.
Also, there is no guarantee that a vanagon Hall sensor has the same peak
values as the VW Fox Hall sensor that I'm using, so it would be really
great if someone could get a scope reading on an actual Vanagon engine to
compare.
Brett
On Wed, Dec 4, 2013 at 3:47 PM, Jim Akiba <syncrolist@bostig.com> wrote:
> Also keep in mind the disty output is very unstable just because of
> the slop in the gear drive/powerpulses as it's rotating. I saw the
> same thing when we built our distributor replacement (replaced the
> stock hall sender with a better square wave output hall sender) and we
> saw this slop in the readings. Our pickup was given 72 pulses per rev
> of the disty so I had lot's of granularity to see it and it needs some
> filtering to be used for any real accuracy (like how we were using it
> for fueling and spark control). But just rpm readings though you
> should be good with some averaging.
>
> Jim Akiba
>
> On Wed, Dec 4, 2013 at 4:09 PM, Craig C Forney <craig@opus.com> wrote:
> > David is right. Multiply pulses per second times 30 to get engine RPM.
> 30
> > pulses is 900 RPM.
> >
> > Target Vanagon idle RPM is 880 plus or minus 50, which is 29.33 pulses
> per
> > second plus or minus 1.67 pulses.
> >
> > ICUs working properly typically perform at 850 to 910 RPM, with most
> > performing at 860 to 900.
> >
> > Craig in Cupertino
> >
> >
> > At 11:33 PM 12/3/2013, James wrote:
> >>Distributor rotates at half crankshaft speed.... idle is 850 rpm....
> >>half is 425 rpm... one Hall Effect signal per crankshaft rotation ( or
> >>is that two per rotation...??).... nah, I'd better try to calculate
> >>this tomorrow!
> >
> > One pulse per (piston @ TDC** on firing stroke) so two per crankshaft
> > revolution.
> >
> > 850 rpm / 60 secs/minute = ~14 rev/second x 2 pulses/rev = ~28
> > pulses/second.
> >
> >
> > ** Really at the fixed timing advance point which is a few degrees before
> > TDC on the 2.1l.
>
--
Brett in Portland, OR
"Albert" '82 VanaFox I4 Riviera
| 
