555 EFI again...

[Walter Petermann]

> ...
> > >http://brokersys.com/~corsaro/wiring_diagrams/555fe.htm
...
... The voltage across
> > the capacitor rises in an exponential fashion, whereas we require that
> > the voltage rises in a linear ramp.  This can be acheived by using a
> > constant current source to charge the cap. 
..
> > With this technique the desired pulse width can be set by altering the
> > threshold voltage, or by keeping a fixed threshold and making the
> > current source into a voltage controlled one.  This can be done by the
> > conventional op-amp and transistor VCCS, (ground referenced) with a
> > current mirror used to provide the current from the V+ rail.
> > --
> > Chris Morriss

Clare Snyder wrote:
> How do we implement this - modification to the referenced circuit?

There is an extra op-amp in the LM358. 

i) Use it with the same resistor configuration as the MAP op-amp with
   these differences:

     The negative input goes to ground instead of the vr2 wiper.
     Use a 2.5K fixed resistor instead of vr3
     Replace the MAP input with a 2K 10turn pot.
     Connect a 3.9K (or the closest higher available value) between
      the high side of the 2k pot and 5v.

This will make the op amp into a 0 to ~0.7ma current source. Controlled
by the 2k resistor.

ii)  Remove vr1 and the 2.7k resistor.
iii) Replace the cap with a 2.2uf 'metallized polyester' (sorry, this is
     a bit on the large size..).
iv)  Connect the current source's 2.5k resistor to the cap 
     (555 pins 6 & 7).

You should be able to adjust the 2k pot to give you a max pulse of 
0-11ms.

I'll put up the modified circuit up this evening.

Someone sugested a way of multiplying RPM * MAP using the 555. The math
didn't quite work out, but it was a great lead. I posted the question
to an electronics design group  sci.electronics.design and received back
a possible solution. I'm thinking that eventually this could turn out
to be the simplest fe ever (that's why I tried to leave out the current
source..). If the MAP * RPM works out we can put a shift register on the
output of the 555 and make a sequential fe.

 Walter

PS

For those interested, here's the original MAP * RPM suggestion along
with the correction that might make it work:

If you set up the 555 so that the time between sparks is the cap charge
time,
cap discharge time is the on time for the injector, and the tps controls
the
CAP charge voltage, is that not the analog electronic equiv. of the
mathematical operation of adding the logarithms of 2 numbers to multiply? 
I hav not thought this through - it was just an off the wall idea, but...
sometimes a 555 can be the basis for a really elegant simple circuit. Is
this
one of those times??

: equations don't seem to come out correctly. Has anyone 
: tried this?

Yes, there is a problem:  the time between pulses is proportional to
1/rpm, not to rpm.  Let's see...try this:  Generate a current proportional
to the sensor voltage (perhaps just the sensor voltage thru a resistor, 
depending on required linearity/accuracy!).  Discharge a capacitor to zero.
For each ignition pulse, let the current source charge the capacitor for
a fixed time.  (Make that time just less than the minimum time between
ignition pulses!  A 555 in one-shot mode would do it...)  After a
fixed time has elapsed, the capacitor will be charged to a value
proportional to the product of sensor voltage and number of pulses
received, and number of pulses is proportional to RPM (+/- 1 count).
You can then let the capacitor discharge through a constant current sink
and generate a pulse equal to the time it takes the capacitor to discharge
back to zero.