1-2-Spark and 1-2-Squirt...

Discuss MegaSquirt, VEMS and other non-free hardware and software here.
User avatar
ToxicGumbo
LQFP144 - On Top Of The Game
Posts: 474
Joined: Tue Jan 03, 2012 8:37 pm
Location: Manhattan, KS. USA

Re: 1-2-Spark and 1-2-Squirt...

Post by ToxicGumbo »

Congratulations. If you're taking any action videos, can you link to them?


-Jeff
LPGo
QFP80 - Contributor
Posts: 32
Joined: Sun Dec 07, 2014 5:27 pm

Re: 1-2-Spark and 1-2-Squirt...

Post by LPGo »

ToxicGumbo wrote:Congratulations. If you're taking any action videos, can you link to them?


-Jeff
Thanks, to be honest I am amazed myself, did a little drive around the block... which I couldn't have dreamed off in the best case scenario...

The first video is accualy a video of the second run..

The first attempt took me about 5 minits befor the engine started (trying to get the pulsewidht right) and then I had to fill out all the cells 1 by 1 revving the engine more and more..

One thing I don't understand is that I need about double the pulsewidht compared to Megasquirt.. hmmm my timing is right, 6000 us is really 6000 us (0.006) sec. Don't know where it goes wrong. Whatever the engine runs on 1-2-Squirt..

https://www.youtube.com/watch?v=K6lTmu-R4Aw

The second video is really a video of the first drive of 1-2-Squirt (and 1-2-Spark ofcourse), yes I had a little drive on my very own programmable motormanagement system, EFI, ECU.. (sorry can't hold back, have to mention this)

https://www.youtube.com/watch?v=zyBSt_6jTDQ
After my video was made the engine leaned off, (do to going to normal heat)
so there is a lot of tuning to do..

To be honest, at some moments I had already the feeling it drove smoother then with Megasquirt (but this has probably something to do with "my own baby is the cutest baby in the world).........


Ps. offcourse the software isn;t right at the moment, it says still dwell up there (insted of injector-deadtime) and it lags behind here and there, but that has low priority for me right now..

Now please enjoy the videos of 1-2-Squirt my own injection system, did I mention that allready??
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: 1-2-Spark and 1-2-Squirt...

Post by Fred »

Based on that second vid, I'm glad you're not sharing the code! It could hurt one's eyes :-D There were some seriously concerning things in several areas, however the only one I'll comment on is how badly written the UI is. Flickering, slow rendering, unlabeled values, just abysmal, really. I sincerely hope you make your firmware more robust before you damage your sweet classic engine with it. Congrats on this first step. I remember 6 years ago the feeling of elation that I had from the same achievement. And now I think back to how basic and incomplete and dangerous that code was and glad it's in the past. Best of luck with your future.
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
Nige
QFP80 - Contributor
Posts: 61
Joined: Sat Aug 10, 2013 5:47 am
Location: New Zealand
Contact:

Re: 1-2-Spark and 1-2-Squirt...

Post by Nige »

I admire people who can do things like this. It is not my area of expertise however.

I do wonder what good could be done in an open collaborative effort though, peer reviewing/reviewed code and the like rather than one person's project. Taking constructive criticism is hard, but a worthwhile exercise, especially if those same folks can help to make it better or give direction.
I would far rather see an open project than a brag thread, no matter what stage it is at.
IRC nick: ignitionigel or Nige
LPGo
QFP80 - Contributor
Posts: 32
Joined: Sun Dec 07, 2014 5:27 pm

Re: 1-2-Spark and 1-2-Squirt...

Post by LPGo »

Wow Fred, you did take a look at the video's......

and looking at your comment your even jealous :D


See Fred, now tell me why do I need to give my code here?????

Now Fred believe me this is the best I can. You should know I have nothing to do with electronics, programming nor carmechanics, I studied for an architect, so I guess it is not bad at all...

Also I don't need a nice UI (whatever that is) I only tune for a day or 2 and then I don't need it (anymore) none of the users didn.t complain sofar

about this windows ide I programm in is for free, as free as it can be (wasn't that one of your anchor-points of this whole site)?????


Take a look over here download it and make your own windows software...

http://www.clickexe.de
clicksy.jpg
Now, Fred please tell me what did you do in the last 6 years, you are still struggeling to get a engine run....

and Fred please give me a link to your tuningsoftware and the software you used to write in (if you did write one), I will take a look at it......

I just did a quick look, I wrote the first injection-software (for the chip) the 23. of februari this year and yesterday I had my car running on it, I think it's not bad... is it...

Geo..
Last edited by LPGo on Sun Mar 15, 2015 12:30 pm, edited 1 time in total.
LPGo
QFP80 - Contributor
Posts: 32
Joined: Sun Dec 07, 2014 5:27 pm

Re: 1-2-Spark and 1-2-Squirt...

Post by LPGo »

Nige wrote:I admire people who can do things like this. It is not my area of expertise however.

I do wonder what good could be done in an open collaborative effort though, peer reviewing/reviewed code and the like rather than one person's project. Taking constructive criticism is hard, but a worthwhile exercise, especially if those same folks can help to make it better or give direction.
I would far rather see an open project than a brag thread, no matter what stage it is at.

The biggest problem is, everyone has is own choises, in hardware, software and so on. Mostly these choises are made by the things your familier with..

So if I tell you that all you've seen in these 2 video's (and all other vid's) is comming from Atmege 328 chips; (1 for ignition, one for injection) I know for sure that Fred needs to be reanimated.. he wouldn't even thinking about using 1 of those...

But again; it is free, free as free it can be, hardware software aso, so again I'm sticking to the rules of this website, my car is running on it ignition and injection......

So if several people would work at it, I'm not sure it would benefit from it, 10 people, 10 opions, 10 executions which doesn't work..

Doing it on my own I can build in the things I want and need, and not what others think I need.....

Take this topic, someone is already citisism that I need a triggerwheel, the other is mentioning bad tuning-software, (ok I can do with beter software :D :D ) but in my filosofy I don't need it...

Geo.
LPGo
QFP80 - Contributor
Posts: 32
Joined: Sun Dec 07, 2014 5:27 pm

Re: 1-2-Spark and 1-2-Squirt...

Post by LPGo »

Today I wanted to start the cold engine at 1-2-squirt but it didn't start... (just for a few short moments only), so I put Megasquirt in its place and again nothing, so my first guess was I blew up the injectors running them without pwm, (but I know some people who run them all the time without pwm).....

So I changed to petrol to get the engine going,,, after a while I came to the conclusion that my LPG tank was empty Thats why it was leaning out yesterday evening.....

Here is a pic from "the day after"...

Attachment:
the day after.jpg

Next week I'm going to programm cranking enrichment, I think I will need it....

and now I'm off for another tuning session....
User avatar
ToxicGumbo
LQFP144 - On Top Of The Game
Posts: 474
Joined: Tue Jan 03, 2012 8:37 pm
Location: Manhattan, KS. USA

Re: 1-2-Spark and 1-2-Squirt...

Post by ToxicGumbo »

If you're not too keen on criticism (not necessarily negative, but honest opinions based on experienced or, at the very least, initial reactions), your project may not gain much fanfare. Most of the people that are likely to take notice here enjoy the development aspect and collaboration. I hope you'll continue to post about your thoughts, goals, and any possible lessons learned regardless of interest or criticism. An open exchange of ideas and progress is extremely valuable even if the code and design as a whole aren't available.

I've been running Fred's firmware on DeuceEFI's hardware via malcom2073's tuning software for years now and almost a full year actively on the road (without a decent tune due to my own failure). There have been some substantial improvements over those years with safety and robustness in mind as a goal for these projects is to eventually be used by nearly anyone with little-to-no-risk of damaging their engines. It says a lot about their achievements in this regard that I made the list and haven't had a single problem despite traveling substantial distances and bouncing between -3F to 110F weather.

What you've done is definitely a great accomplishment. If you're hoping to broaden who's driving on it, there will definitely be some strong statements here to encourage further development to protect your users--something that isn't a strong focus of the Mega$quirt team and which may take some getting used to.

I'm not a skilled as a developer and enjoy seeing what everyone brings to the table and where they're heading with it. That aside, I do encourage you to consider a license which protects your work while possibly exposing more of the functionality and design. And it never hurts to consider criticism, particularly if your plans are to develop a commercial product for others to use. The reigning projects here are heavily transparent and peer-reviewed and that's where many attitudes will be coming from.

So again, congratulations on pulling it all together! I'm enjoying the videos and commentary. :)


-Jeff
User avatar
sim
LQFP112 - Up with the play
Posts: 112
Joined: Thu Jun 02, 2011 8:17 pm

Re: 1-2-Spark and 1-2-Squirt...

Post by sim »

LPGo wrote: Now what does a regular 36-1 tooth chip do... It measures lets say 5 or ten times per revelution (6000rpm) where the crankshaft is at the moment and does then its calculations.. Then a lag factor goes on top of this, now tell me how accurate is this????
It is so wrong that the software even has a prediction build in otherwise the ignition would be to far of...

Please correct me if I'm wrong.
I don't think you have this right.

Every tooth in a 36-1 wheel causes an interrupt in the processor.
So that means it measures exactly 35 times per revolution,
regardless of RPM. The more teeth the wheel has, the higher the
resolution. Higher resolution means greater accuracy calculating
instantaneous RPM and also the rate of change of RPM. The more
accurate those two data points are, the more accurate the
calculated spark event will be.

Will a single tooth work? Absolutely! My Volvo ran just fine for
32 years with a condenser and points; effectively a single tooth
system. It has run a FreeEMS system for the last four years with
a 12+1 wheel (counting in crank degrees) and it runs way better
that way. The tooth count is far from the only change made.

Please do not assert that a different set-up is inferior simply
because you have made up your mind to not use it.

For what it's worth, I don't think you need a multitooth wheel. I
believe it would overcomplicate your system, and you may not have
the processor power for it. Getting the code right is non-trivial
also.

Nice work getting an engine running on a spark and fuel system of
your own devising, that is an accomplishment, make no mistake.
<@TekniQue> but in the end, it's code that makes a computer useful
LPGo
QFP80 - Contributor
Posts: 32
Joined: Sun Dec 07, 2014 5:27 pm

Re: 1-2-Spark and 1-2-Squirt...

Post by LPGo »

Now please explain to me... taking a 36-1, ok time is measured every single teeth, maybee there will be looked if it is the missing tooth, but how many times gets the advance calculated, or the injectorpulsewidth. maybee 500 times a second (if you have a good chip).... or am I very wrong here?????

Points in your car work slightly diffrent, the moment they split then your spark occurse.. So always spot on (if adjusted correct)

I'm working diffrent..
1. The first sensor passing the timer resets to zero (45 degrees btdc)
2. The second sensor passing the time is measured by the timer, Rpm is calculated, advance is calculated too in time, by the tables temp and map) and when the time has gone the
Spark occurs (between (almost) 45 and -6 degrees as calculated)...

1-2-Spark..

The same procedure I do with 1-2-squirt right now...

So I need only 1 tooth (at 45 degrees b.t.d.c.) at my crankshaft.
Post Reply