Multiple ignition output

From DIY contraptions to sophisticated FreeEMS-specific designs! Plus general hardware development!
User avatar
longracing
LQFP112 - Up with the play
Posts: 140
Joined: Wed Jul 16, 2008 9:21 am
Location: NSW, Australia

Multiple ignition output

Post by longracing »

I know I suggested this in another post but I think it got swamped and forgotten about.

I suggested using a serial buffer but now I am thinking of using an 8-bit uC like a PIC12F, PIC16F or ATTINY to decoder a logic signal, similar to Ford EDIS.
This would only require one, two or three digital outputs to drive freeing up processor pins. It could be programmed to handle a number of logic outputs (up to 24 with the write choice of uC) with a very small code (written in assembly maybe), this would also free up the main processor.

I guess it would be a bit like a multi-coil distributor but only handling the low-side of the ignition drive.
Maybe one input for coil charge start,
one input for coil firing,
and one input for a home signal. It wouldn't care how many coils you had, it would just keep increasing the count until reset by the home signal. If written in assembly I believe it would have minimal delay time (maybe nS).
Any comments?
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: Multiple ignition output

Post by jharvey »

I'm not very familiar with Fords EDIS ignition system. Sounds like your an intermediate processing board could take ignition outputs and translate them into EDIS. Is that right? Do you have a link that explains EDIS? I can't access MS sites at this time, so non MS links are best for me.
User avatar
longracing
LQFP112 - Up with the play
Posts: 140
Joined: Wed Jul 16, 2008 9:21 am
Location: NSW, Australia

Re: Multiple ignition output

Post by longracing »

Here's a link about the Ford EDIS http://www.dainst.com/info/edis/edis.html

The Ford unit actually takes the crank sensor data and sends an RPM signal to the ECU ( called a PIP signal).
It then recieves a ( SAW ) signal back to tell the EDIS the timing, when to fire the ignition. The EDIS takes care of the Dwell and which coils to fire.

I was thinking the FreeEMS could still receive and process the RPM signal but intermediate processor could take care of the dwell and distributing the signal to the coils.
If you wanted the dwell could be set by the intermediate processor ( by serial coms or even set with a BCD switch since it would not require changing after setting up).

I suggested the ATTiny because you can build a serial programmer for them using a few resistors and diodes. The PICKit has also come down in price alot.

I beleive it could require a different structure to the main code so feel free to pick this idea to bits.
User avatar
longracing
LQFP112 - Up with the play
Posts: 140
Joined: Wed Jul 16, 2008 9:21 am
Location: NSW, Australia

Re: Multiple ignition output

Post by longracing »

Just had an idea. :idea:

Using a 20-pin PIC uC, BCD switch to set Dwell, 1-input for the ignition "firing" and 1-input for a home signal (reset cylinder count).

If this could be used for 4, 6, 8, 10 or even 12 cylinders would this free up enough outputs to enable the FreeEMS to have a couple more fuel outputs or would it just free up the pins?

Fred, I know you've already put a lot of work into the FreeEMS code so feel free to Veto this from the FreeEMS version but seeing how common 8 cylinder engines are it could open the project to a lot more users. (Yes I'm one of those gas guzzling V8 drivers :oops: )
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Multiple ignition output

Post by Fred »

Hey, I had a ride in SeanK's turbo V8 z28 camaro tonight, it's fcuking fast, about 600hp + based on his injectors and duty, i have nothing against them (yankee pushrod v8s) except in my mind the hatred of pushrods and rockers ;-) [i actually feel the same about 4 banger rocker/sohc engines too!!] {and i love the 1uzfe toyota v8}

Still, so long as I'm not achieving anything myself, i'm not vetoing anything. It's that simple.

The real time aspect of the firmware has been the biggest bitch so far, by far. If you think you can master good real time control, go right ahead. Whatever reads the input signals needs to do the output too, but something else could provide the parameters for that, I guess.

Feel free to play around on the bench, I'm not totally anti the idea, but I do think we can do a lot with the CPU we have already.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: Multiple ignition output

Post by jharvey »

That is a nice description of EDIS. Sounds like quite a project to interface to it. I don't understand what wires go to and from the device. It sounds like it controls all cyl's spark, sends an RPM signal to the ECU, and expects a SAW signal from the ECU. I don't really understand what they mean by a SAW signal. To me a typical SAW signal means one of the below.

http://en.wikipedia.org/wiki/Sawtooth_wave

http://en.wikipedia.org/wiki/SAW_filter#SAW_filters

However in this case this appears to be some kind of bit shifted digital signal. I'm not sure how many wires, or what impedances they have, ect. To me it sounds like a good source of issues, and if I were going start something from scratch, I'd likely go for connecting the VR's directly to the ECU.

I see a potential benefit of getting 8 cyls, but other than that, what benefits does EDIS offer?
User avatar
EssEss
LQFP112 - Up with the play
Posts: 244
Joined: Thu Sep 10, 2009 9:23 am
Location: Dayton, OH

Re: Multiple ignition output

Post by EssEss »

jharvey wrote: I see a potential benefit of getting 8 cyls, but other than that, what benefits does EDIS offer?
a limp mode of sorts .. if (for whatever reason) the base ecu stops providing advance info, edis defaults to a hard tooth, and if you've phased the pickup correctly it'll fire @ 10bdtc. It's a completely standalone system that simply works off of 'advance hints'.

also understand that you will never go over ~9/10k w/an edis pack - this means kit car guys who like to drop in moto engines and moto guys are not going to use it. but to control it in existing systems (which are becoming less-n-less) is simple to accomplish.
User avatar
longracing
LQFP112 - Up with the play
Posts: 140
Joined: Wed Jul 16, 2008 9:21 am
Location: NSW, Australia

Re: Multiple ignition output

Post by longracing »

Ok, I probably shouldn't have refered to the EDIS system as what I was thinking of is quite different.

I've made a quick schematic (attached in .pdf format) of my idea using a PIC16F677. These only cost a few bucks each, have an internal oscilator and plenty of in/out pins.
Summary;
Running on +5volts (forgot the 0.1uf cap on the power supply)
Has an ignition input on PortA using an interupt.
Has a home signal input on PortA using an interupt.
Has 8 x Ignition outputs (for 8 coil ignitors) on Port C. (Could have more added on another port)
* Has a Binary Coded Decimal input on port B to program the dwell (I didn't have a BCD in the schematic program so there are just 4 switches there.

The home signal could even be dirrectly from signal conditioning.
If the BCD switch was used to program a known dwell time then the main processor would simply need to output an ignition signal at the required time - (minus) the known dwell. The dwell would have to be entered in the tuning software but this could possibly free up a timer? (Would have to check with Fred)

The other option would be to have the dwell start by the falling edge of the ignition signal and then finish (ignition fire) by the rising edge.

The potential is there to get more ignition outputs but the main reason for doing this would be if it could free up a timer or processing time to add more fuel channels.


I have to agree the Toyota quad-cam v8 is a well engineered engine. Did you know Holden developed a brilliant 36valve, overhead-cam variation of their V8 before canning it for the LS1. :roll:
Attachments
Ignition Multiplexer.pdf
(17.46 KiB) Downloaded 708 times
User avatar
EssEss
LQFP112 - Up with the play
Posts: 244
Joined: Thu Sep 10, 2009 9:23 am
Location: Dayton, OH

Re: Multiple ignition output

Post by EssEss »

I think I kinda of get it, correct any of the following assumptions if they are wrong:

1. so your dwell start is predicted for each particular cylinder ?
2. so you need at least one whole complete rev (2 'home' pulses) to actually begin prediction ?
3. you're depending on prediction to be well enough that the appropriate output is to fire when you cycle the 'input' pin ?

Essentially, you're using a single tooth crank wheel. Based on the above assumptions (again. correct me if I'm wrong) it's entirely possible. I personally wouldn't use it. I think there are better solutions out there.

I do have alternative proposals. I almost understand Fred's original system well enough (through his code) to know that the towel is not thrown in yet .. I don't see finished (or attempted) code which says that this is a hard restriction yet. Is this the horse before the cart ? I also see the the other core currently unused ?

I'm still reading up and digesting this all, I may have missed something. Online collaboration is proving painful for me :oops:
User avatar
longracing
LQFP112 - Up with the play
Posts: 140
Joined: Wed Jul 16, 2008 9:21 am
Location: NSW, Australia

Re: Multiple ignition output

Post by longracing »

Looks like It might make it too complicated (I'm still working my way through the code. I also downloaded 0.0.9 instead of 0.0.19 thinking it was the latest :oops: ). What I would like to know is if it is going to make the hardware + firmware more complicated it is a bad idea. If it can improve the firmware by freeing up resources for other outputs then it may deserve a bit more thought.
1. so your dwell start is predicted for each particular cylinder ?
2. so you need at least one whole complete rev (2 'home' pulses) to actually begin prediction ?
3. you're depending on prediction to be well enough that the appropriate output is to fire when you cycle the 'input' pin ?
1. No not predicted dwell, the Main ECU would calculate when it wanted to fire the ignition minus the known dwell time - it would create an ignition pulse at the known dwell time before ignition. Or it would see the falling edge to start the coil charge and the rising edge to end coil charge (fire the ignition).
2. It would only need to see one home signal to set the cylinder count back to ign#1, then it would increment to ign#2, then ign#3.... etc. until there is another home signal.
3. It would only be routing the signal, the ECU would still be controlling the ignition events.


Starting to see this as a bad idea, adding too much complexity and not adding that much to the output.
Post Reply