> > This is total pulse width. Below about 15 inches of Hg only
> > the primary injectors work. Above 15 inches both sets of injectors fire.
> > Tracy adjusts the pulse widths so there is a smooth transition at 15
> > inches of Hg.
> >
> > The three curves represent what happens when you change
> > the injector size in Tracy's Mode 3. The stock injectors default pulse
> > widths is the middle curve. When you increase or decrease the injector
> > size or flow rate the curves shift right or left.
> >
> > Note; with the stock injectors you will run out of max pulse width
> > around 7500 RPM. That would be about the upper limit using the stock
> > injectors default pulse width. If the stock injectors will not flow
> > enough to provide the necessary fuel for your HP generated at that RPM
> > you will have to go up in injectors size or flow rate. This is why I
> > chose higher flow rate injectors for the P-port engine.
> >
> > Paul Lamar
> >
>
> >
> > Good point Chuck. I need to get some of those new skinny Bosch
> > or RX8 injectors and test them for dynamic range down below
> > 1 ms or so. Anybody got one they can send me? Either that or I need
> > to talk Tracy into supporting six injectors like the high power
> > RX8 engine uses. Obviously Mazda ran into the same problem
> > with the high power RX8 engine that has a 9,000 RPM red line.
> >
> > Paul Lamar
>
>
> Paul
>
> I tried asking Tracy if his injector drivers would handle two RX8 injectors
> wired in parallel as the secondaries, which is what I intend to do, even if
> it means modifying the output drivers to handle the extra current. (I can
> always go back to the michrotech.)
> No meaningful response, he insists on using one primary and one secondary
> per rota.
>
> Ian Beadle
>
> Hi Ian (or anyone),
>
> I need a little help here as I am running out of time.
> We need to do a little testing of injectors. I am sure you have a scope.
> Can you build a rig to test injectors? You will need a high pressure
> pump and some sort of pulse generator to output a variable width pulse.
> A 555 and would work for the pulse generator. http://www.electronics.50g.com/cosc555.htm
>
> 1200 RPM could be chosen for the bottom end test. Needed is a calibrated lab
> beaker and stop watch to measure the total flow for one minute or so.
>
> The top end is not a problem as MSD sells 50 pound injectors good
> for about 100 HP each at .5 BSFC. The big question is what happens
> below 1 milli second. The engine can run quite rich at idle without
> harm but we still need to know how much fuel is coming out of those
> big injectors at low pules widths. It would also be nice to know
> how much fuel is coming out at ANY pulse width :)
>
> Here is the injector driver schematic from Megasquirt.
>
> http://www.megasquirt.info/
>
> I don't know how hard it is to buy semiconductors in Perth.
>
> Paul Lamar
>
Paul,
Here's some information on how to set up a simple injector testbench.
This should work for both the low and high impedance injectors. The
LM1949 is a "peak and hold current" injector driver that will fit the
bill. Stick an oscilloscope (CH 1) at pin 1 to monitor the PWM signal
and (CH2) at pin 4 to monitor the current going through the injector.
(same as attached photos)
The injector isn't opened (typically) until the current in the injector
reaches a certain level. After which, the inrush of current increases
at a much faster rate since the pintle is now retracted inside the coil.
Note the "knee" in the first pic. The Second pic is a better overview.
I measure the actual opening of the injector based on the knee and not
the start of the pulse width. The "set up time" for the injector can
easily be determined and considered a constant.
Parts List:
- LM1949 is $4.23 at digikey
- I used the 2N6388 instead since digikey carries it as well...
- If you can't find the 33V Zener, a higher voltage is acceptable as
long as it is well below the breakdown voltage of the Darlington.
- For the 0.1 ohm resistor: a higher wattage is acceptable, don't
substitute the value.
Marc Pedings
Good work Marc. Thanks a bunch. Looking forward for the flow
as a function of pulse width.
Thanks again.
Paul Lamar
No problem... For pin 8: do not use the timer. Ground it instead (no
need for "Rt and Ct"). Datasheet is attached. If you need something
quick I can send you all the parts. Breadboarding this is a cinch.
Marc Pedings
Good work Marc.
How about doing a 555 pulse width generator on the same prototype
board for us. My guess is around 6000 RPM or 100 Hertz. Vary the pulse
width from .1 ms to 6 ms if possible. If not .1 ms to what ever
you can get on the high end. You might need two 555's to get the
low duty cycle. I understand the CMOS versions are wider bandwidth.
Here are a couple of scans from my antique copy of Don Lancaster's
TTL Cook Book :) Please ship it to me with the bread board and
I'll pay you whatever you have in it.
I'll supply the injectors and high pressure fuel pump and regulator
and take some fuel flow data. I also have the hand vac pump to vary
the fuel pressure. I also have some RX8 injectors to test I got from
Mazdatrix.
Paul Lamar
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Copyright 1998-2005 All world wide rights reserved.
