I think I am running into limitations of the MC68HC908.
I think I need another timer module for the dwell calcs.
Since I am doing another board anyway I am thinking putting a faster and
better
processor on it.
Not much different in cost and some of the later Megasquirt asm programs
will work.
The same processor is used on the daughter card. However there is other
stuff on the
daughter card we don't need. GM stepper IAC control not needed. CANBus
communications
for interface with GPIO Board, IO Extender, or other devices.
DIYautotune wants $199 for the daughter card but you still have to pay
another $250
for the mother board.
Why not? Just a few extra bucks for the processor. $10 total roughly. You
can never
have enough I/O. http://tinyurl.com/755wdzh
The MC9SC64 has no ROM in it. You flash the whole program in it.
It has 64K flash and 4K RAM.
Here is a list of feature.
-----------------------------------------------------------------------------------------
Freescale Semiconductor, Inc.
— Flexible identifier filter programmable as 2 x 32 bit, 4 x 16 bit or 8 x
8 bit
— Four separate interrupt channels for receive, transmit, error and
wake-up
— Low-pass filter wake-up function
— Loop-back for self test operation
• Timer Module (TIM)
— 16-bit Counter with 7-bit Prescaler
— 8 programmable input capture or output compare channels
— Simple PWM Mode
— Modulo Reset of Timer Counter
— 16-Bit Pulse Accumulator
— External Event Counting
— Gated Time Accumulation
• 6 PWM channels
— Programmable period and duty cycle
— 8-bit 6-channel or 16-bit 3-channel
— Separate control for each pulse width and duty cycle
— Center-aligned or left-aligned outputs
— Programmable clock select logic with a wide range of frequencies
— Fast emergency shutdown input
• Serial interfaces
— One asynchronous serial communications interface (SCI)
— One synchronous serial peripheral interface (SPI)
• CRG (Clock Reset Generator Module)
— Windowed COP watchdog,
— Real time interrupt,
— Clock monitor,
— Clock generation
— Reset Generation
— Phase-locked loop clock frequency multiplier
— Limp home mode in absence of external clock
— Low power 0.5 to 16 MHz crystal oscillator reference clock
• Operation frequency
— 32MHz equivalent to 16MHz Bus Speed for single chip
— 32MHz equivalent to 16MHz Bus Speed in expanded bus modes
— Option: 50MHz equivalent to 25MHz Bus Speed
• Internal 2.5V Regulator
— Supports an input voltage range from 3.3V-10% to 5.5V
— Low power mode capability
— Includes low voltage reset (LVR) circuitry
— Includes low voltage interrupt (LVI) circuitry
• 48-Pin LQFP, 52-Pin LQFP or 80-Pin QFP package
— Up to 58 I/O lines with 5V input and drive capability
— Up to 2 dedicated 5V input only lines (IRQ, XIRQ)
— 5V A/D converter inputs and 5V I/O
• Development support
— Single-wire background debugTM mode (BDM)
— On-chip hardware breakpoints
— Enhanced DBG12 debug features
----------------------------------------------------
Paul Lamar
I'm for the better processor. Like you said, you can never have too
many I/O's... especially if you foresee integrating an EFIS, etc.
Josh
The software for that is already done. Both in the megasquirt and the PC.
Over the years a tremendous amount of work has been done. It pays to take
advantage
of that.
Paul Lamar
G'day Paul, Josh
I 'm totally out to lunch when it comes to EFI but I have a few
points/questions
1. does an o2 sensor adjust the fuel to acheive clean burning ( to
pass smog laws) or does it just give us a reading? This may foil our
efforts to adjust AFR for lean burn and best power.
2. if the sensor is clogged by solid lead accumulation, perhaps we
could move the pick-up out of the direct airflow by placing it in a
recessed position or behind a shroud/deflector. this will still
analyze the exhaust but protect it from lead accretion.
So I thought I would search to see if I could find a lead safe O2
sensor and found this info. I reckon that this may mean more to you
Paul than I.
http://www.factorypro.com/dyno/dyno_ec997,eddycurrent,mc,dynamometer.html.
Also try the links at the bottom of the page.
Maybe if we used a TPS that would shift AFR to best power at wot and
best economy below 75%. Or perhaps different power maps as the race
boys use for power/economy modes. Back to you.
Graeme Riley
1. Usually the O2 sensor system are set to stoichiometric or about 14.7:1
air fuel
ratio. It depends on what kind of smog you are talking about. If you are
talking
about NOx yes this gives a low smog number but at a severe fuel burn
penalty. Lean
burn high NOs gives a better BSFC. Long story but here is a bit about it.
B.
Bruce-Biggs. The War Against the Automobile. Dutton. 1977 book explains it
the best.
Don't get me started :)
2. They might be one to something.
http://tinyurl.com/cps7v2o
Paul Lamar
1. If you're using a wideband O2 sensor, Megasquirt can adjust the
fuel to any target AFR that you specify. I believe you can even
specify different AFRs in a MAP vs RPM table so that you can target
different afr's in different situations. If you're using the narrow
band variety, it can target 14.7:1 only. In either case, the AFR
"seeking" function is optional. You can have it just give the
readout.
2. I'm not sure if shielding the sensor would help with lead
accumulation or not. It would be interesting to see if it works.
Josh
The new chips are surface mount and a lot smaller. I was never really
fond of DB
connectors for aircraft use anyway. Ten times easier to wire. What do you
guys think
of a smaller, lower cost board with these connectors?
This board is 3.9 inches by 2.5 and is stackable. One rotor, two rotors,
three rotors
or four. The heat sink on the top will double as a spacer. In and out are
separate.
Input on the left and output on the right.
Paul Lamar
I believe there is a similar connector that is 2 piece. The base is pins,
the top
mating connector has the screw terminals. Makes servicing easier, instead
of having
to unscrew each individual wire, just pop off the top piece. Stands up
just fine to
vibrations, very snug mate, but easy to add a locking device, if it makes
you feel
better. Used in industrial applications everywhere.
C. Smith
I have seen those.
I have some space limitations. I can get 3 PC boards for $50 if they are
no larger
than 2 X 3.8 inches. The terminal strips are on 3.5 mm centers. Thats .137
inches.
Here is the latest version of the board. It has the best features of MS 1
and MS II.
It is able to run the latest software as it uses the powerful $10 MC9S12
chip.
We just don't need half the crap that comes with the latest Megasquirt
boards like
CAN, 8 cylinder injection and stepper motor drivers.
Not done yet but I will have a little real estate left.
Paul Lamar
As far as I can tell, they will have the same footprint on the board. Will
look for
the PN as soon as I can find one in the hangar, or get back to work on
Wed.
C. Smith
Is this it? Ost plug and socket. 0.15 inch pitch.
These are getting extremely popular. There are a zillion on the market.
How well does the plug lock in to the header?
"TE Connectivity's AMP brand products
- TE Connectivity AMP, formerly Tyco Electronics AMP is the world's leader
in the
development and manufacture of a wide variety of electronic/electrical
connectors and
interconnection systems."
Are you still in Michigan?
Paul Lamar
That looks very similar. The type I had in mind has the pin strip at
vertical to the
board, and the conductors 90deg out. It is a widely used style of
connector in
industrial applications. Many manufacturers have their own variations. The
ones I
have experience with are AMP manufactured, and the plug is quite secure.
takes a fair
amount of force to pull apart, and the pull has to be well aligned with
the pin axis.
Have never seen one that was properly seated come apart on it's own. Not a
lock, per
se, but a good deal of friction. A better method of termination would be
to tin the
copper, then a sleeve of heat shrink over the transition from tinned to
copper to
insulation for support.
No longer in MI, have relocated to Louisville. Hangar on Bowman field.
Building new Escape.
C. Smith
Paul,
I have no real amount of experience in what I am talking about
but..........
I have seen a lot of debate over "proper aircraft quality wiring" and have
to assume that it is indeed an important subject. I see the advantages of
these connectors but they are industrial and not aviation. Years ago in my
first RX7 I was a little too assertive in pulling the connector out of the
MAF sensor. My father, who was an electrical engineer working in radar,
navigation and weapons control systems for the Canadian government trouble
shot the MAF and after removing the potting material discovered that a small
wire had broken at a solder joint. Even though he was very experienced in
this sort of work he refused to do the repair himself , instead sending it
to Halifax to a colleague who was recognized as being the best at this sort
of detailed work. Apparently he recognized the importance of skill.
It seems to me that there are going to be several single piont failure
modes here regardless of how we go about making the connections so I wonder
if we are wise to not use the safest industry recognized connection method
available. Does Robin know of any aviation based connectors ? I am
interested in experimental aviation but not experimental electronics. I
would hate for rotaries to fall out of the sky because of something like an
industrial connector or an automotive hose clamp.
John
There is nothing wrong with industrial connectors for use on aircraft as all the
parts are soldered on the PC board as well. In many cases the industrail environment
is severe.
Look in any avionics box sold and you will see the same thing. Normal procedure is to
support the parts so they do not vibrate at the frequencies and accelerations they
are likely to see in a general aviation aircraft. Many use DB connectors with
soldered wires supported by heat shrink tubing.
According to NASA we are doing better by crimping the wires in these connectors.
However you are free to use AN connectors if you wish but you will still have to
solder or crimp the wires to the AN connectors and use heat shrink.
One can also use vibration isolators to shock mount the EFI system.
I shock mounted Tracy's computer in the RV4 using these mounts.
BTW the smooth rotary engine goes a long way to reduce the frequencies and
accelerations the avionics devices are likely to see.
Paul Lamar
All the aircraft programs that I have worked on use Mil-38999 connectors extensively
in the harnesses. They do have board mountable ones, but they are all round. That
would probably preclude stacking the boards on an enclosure
Josh
--
The Rotary Engine News Letter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
You Tube key word PaulLamar2
Copyright 1998-2012 All world wide rights reserved.