rb26dett

Hi,

This thread is here to introduce and discuss DIYEFI.org, a new community with a difference.

For those of you who don't know who I am, I've been keeping a close eye on DIY EFI stuff since the late 90's and been involved with MegaSquirt for the last 3 years or so to some degree or other. Many things have come and gone and many things have changed in that time. The time has come for an even bigger change. Those of you who "just use" MegaSquirt and don't get too involved won't really care too much about this site for at least a few more months. On the other hand, if you like to participate and get involved, this site is probably for you.

As a way of testing whether you might be interested, I have assembled some questions here :

• Do you use GNU/Linux?
• Does engine management interest you?
• Are you a fan of Open Source Software?
• Do you enjoy printing your own boards?
• How about designing your own boards?
• Do you dislike closed door development?
• Tired of Lance deleting your posts on msefi.com?
• Tired of waiting for UMS, PWC, MS3, Router board, GPIO or Sequencer?
• Willing to put in the hard work to have the engine management system of your hearts desire?
• Clever with electronics or code and willing to work on an open source project for the greater good and benefit to all?

If you answered yes to any of those you may want to take a look at my new site, if not, there is no need to bother :-)

The site is totally non-profit. There aren't even any adverts (yet, probably never if I can help it). The designs will be community produced and available for free download just like FireFox and Linux etc. There are no businesses or companies behind it. Just me and a group of like minded individuals who want to make the DIY engine management world a better place for everyone.

If you are wondering why I have gone and put this much time, effort and money into apparently reinventing the wheel, take a look here http://www.diyefi.org/why.htm

If you are wondering who else is involved in the project, or indeed who has made MegaSquirt the runaway success that it has become, please see the appropriate sections of this page http://www.diyefi.org/contributors.htm

If you are wondering about some of the details of the history of Do It Yourself fuel injection, have a read through this partly incomplete history page http://www.diyefi.org/history.htm

If you want to dig straight in then have a of this thread http://www.diyefi.org/forum/viewtopic.php?f=17&t=90

The code is nothing special just yet, but some of the foundations are there and working flawlessly http://www.youtube.com/watch?v=jsXPi2KWRAU

It will be some time before it runs an engine, but that doesn't mean you should wait till then to get involved. If you express your opinion in a constructive way now, you will be happier with the end result when it finally arrives.

It was recently suggested to me that I was trying to recruit the type of people that complain a lot :

This is NOT the case. In fact, that is the LAST thing that I want. If you are just going to whinge, whine and complain about the past, please stay away. If on the other hand you have something positive to offer about the future, please sign up!

The forum is here if you would like to join our efforts : http://www.diyefi.org/forum/index.php

However, before joining, I recommend that you have a read of this first http://www.diyefi.org/forum/viewtopic.php?f=17&t=26

If you want to see some of my work that is more or less completed and you have DSL have a look here http://www.diyefi.org/forum/viewtopic.php?f=3&t=16

If you are wondering more about me, some would say that one word sums me up nicely

http://i260.photobucket.com/albums/i...g/100_7044.jpg

though, I prefer to prefix it with 'car-' ;-)

Thanks for reading, I hope you enjoy what you find :-)

Fred.

sotaku

iTrader: (4)

I love the idea of more DIY and OpenSource EFI projects out there but you might wanna reword a couple of your bullet points if you really don't want B&G Gripers:

# Sick of Lance deleting your posts on msefi.com?
# Sick of waiting for UMS, PWC, MS3, Router board, GPIO or Sequencer?

Those two really are a turn off. I could care less about the politics of MSEFI - I'm more interested in contributing to a cool project and being able to figure some stuff out than I am "sick" of anything.

Best of luck to you though.

Braineack

iTrader: (62)

The problem is, myself included, we are too stupid. We need things to be done for us and then we'll jump on the bandwagon.

rb26dett

Hi, Thanks for your input :-)

I think it is possible to answer those with a 'yes' and still be a positive person looking forward and not back.

Lance's heavy handed moderation over mentioning the word "extra" is a real problem. It was lance that split the forums apart in the first place. Now he gets upset when people post things such as "where is all the msextra info". Further to that is the appalling treatment that the MSExtra devs have suffered due to this anti-extra bias.

I think it's good that you can ignore such politics, but they do exist, and some can't. I am one of those.

Does the word "Tired" go down more smoothly do you think? Perhaps.

All comments are appreciated.

Fred.

rb26dett

From what I have read of some of your posts in the past around the place, I'm not so convinced that that is the truth. Still, that is what I am doing anyway, but don't hold your breath :-)

Fred.

elesjuan

iTrader: (11)

I agree with Brain. Personally I'd love to contribute 200% to the project but my programming skills are non-existent and my knowledge of circuit design is very very very poor. Those two facts limit me to a sideliner who begs for new developments despite my huge urge to do things for myself. :(

SloS13

iTrader: (1)

I thought your setup sounded familiar. You've got some fab skills thats for damn sure.



http://www.homemadeturbo.com/forum/i...topic=81330.40

rb26dett

I'm still not convinced that you have to be a god like C coder to be useful here!

Just throwing your opinion and thoughts around a bit when stuff is up for discussion is important. It is that IMO that is missing from MS development. I.E. put me in my place! :-)

To rule out someone like yourself would be a big mistake for any start up like this. You just never know where good talent of various kinds might come from and what surprise uses various people might become :-)

At the very least, if you have a few dollars and hours spare, you can try to make my code not work! :-)

When I was studying at uni, me and my best mate used to do just that. Swap code and try to make it fail. It was an excellent ploy. We always managed to do better work faster than pretty much anyone else :-)

Fred.

rb26dett

Thanks :-)

Caution : beware of flame war (and of course the usual bad language etc that is par for the course on hmt) :

Fred

AbeFM

iTrader: (3)

I've been hanging out over at Fred's place a bit now, and I can say that even someone with no discernible skills such as myself has some ideas - we've spent a lot talking about things like "should we explicity support rotary", or "is logging internally a good thing". Of course, it should do everything and bake you a cake as well, but there are trade-offs, and that's where the talented folks have to weigh the input they get against the reality of putting things together into a coherent package.

Personally, politics aside, I just want to get something more capable. I want sequential injection, individually fired spark, AND things like boost control, water injection, staged fuel injection (or Dial-An-Octane as it's looking like it might be - very exciting), without running out of inputs and outputs. Could you imagine giving up fan control for an idiot light, or dropping EBC just so you can have a tach?

That's the situation I get every time I work on my MS. And it doesn't help that if I ask a straightforward question of the people who laid out the board or wrote the code, they act like it's greek to them.

Plus, I just like projects. So, it's worth my time to see what's going on, and I'm sure I'll be amongst the earliest beta testers... :-)

rb26dett

Thanks Abe :-) I hope not to disappoint!

The_Pipefather

iTrader: (5)

Hi Fred,

Looks like you guys are a very talented team of people who have the passion and conviction to make this happen. I am pleased to see that some members of the MS/Extra team is also on-board.

Anyway, I read the links that you placed in your post and you guys seem to be concentrating on the code a lot more than the hardware, atleast for now.

I have a couple of comments/questions.

1) Most OEM code is hardly hand-written these days, with rapid-control prototyping being the buzzword in OE circles. Usage of tools like Simulink and dspace is rampant, thus making the development of several thousand lines of code within the reach of people like me who hardly know anything about microcontrollers or C. Now I know that writing the code is part of the fun for you guys, but any intentions of using a graphical coding tool (like Flowcode, for example) in your development process? I ask this with a solely selfish intention, so that "code-impaired" mechanical engineers like myself can get involved too.

2) Since FreeEMS is intended to be the affordable open-source ECU of the future, I think that the hardware and way the system is integrated into the vehicle should also reflect that sentiment. I firmly believe that more and more people who use such a product are going to be leaving most of their OE vehicle systems alone, preferring to take control of only the fueling and timing. So would a piggyback-type version of FreeEMS, with the option of creating a "virtual-reality" around the OE ECU, be a possibility?

elesjuan

iTrader: (11)

Okay so I shouldn't rule myself out 100%, you're right. I do have a lot of ideas I could suggest for implementation and would be more than happy to help test out anything possible. On those rare occasions I have time tinkering is my number one hobby.

You mentioned already something about a digital dash which is something I've really wanted for many years since Ex Vi's Skyline R33 build, and SiTune's Euro Spec Escort. Something like this, only cooler:



I had high hopes for something similar when I saw the touchscreen megaview, but was discouraged at the price tag and limited display. Limited display is also something I dislike about Megatune. Sure the 8 gauges can show a lot of info, but maybe I want more displayed that I could take a real quick glance across? Plus the gauges in megatune have a lot to be desired cosmetically.

rb26dett

Jump on and pm ababkin about it. He has a similar master plan, you guys could start a thread and get started drawing it up :-) He codes for a living too, so after him for code, you for ideas, you just need a hardware guy to make it happen :-)

rb26dett

Thanks, I certainly hope you are right. It seems to be true anyhow :-)

Yes, I am also pleased to see them there, 4 so far have suggested involvement, or participated already. However, after receiving a PM a few minutes ago, I feel that I should point out that James and Ken are NOT involved. I was inspired by the great thing they have turned Al's code into, and wanted to acknowledge that inspiration publicly. I certainly did not intend for it to appear that they are involved, and I have made repeated changes to that page to make it more clear. I will make some more drastic changes now to illustrate the point.

I feel that the code is the most important part at this stage. Experience has shown that if there is code around, everyone wants to knock up a board for it to run on. On the other hand, there is no point building a new board without good information about how the cpu will be hooked up and used. Hence, code first, boards later.

Most likely no. There are a number of reasons for this. As you stated, writing the code is half the fun (or more), however the chip is still somewhat underpowered compared to what OEM's would use, and thus hand coding is necessary to extract the performance required to do what we want to do.

If you have been scared off by reading the MS code at all, you should download my code and have a read. It is VERY different to theirs. IMO, much more understandable to an average observer. Together with my initial supporters we have come up with some advanced techniques to make sure the best possible transient response is achieved. However, everything is documented to the tiniest details, so most anyone that can read english can get at the very least, the gist of it.

Another reason would be the lack of free cross platform tools to perform such automated code generation.

Also, I am yet to see any significant generated code of any complexity that is particularly good (exposes his naive view of the code world)

Hmmm, I'm not certain that I agree with that, however, it certainly is a possibility. For a start, OEM style wiring looms have grounding techniques that can cause difficult to track noise issues. Secondly, if you are going so far as to use a "stand alone" surely you want to do it right and replace the old ecu? Because we have 3x the IO than ms2 it should be possible to control all of your accessories fairly easily and still run the engine as well as is humanly possible.

I hope that clears it up a little.

I'll try to fix that page up right now.

Fred.

Arkmage

iTrader: (3)

once my work schedule slows back down this sounds like something I'd be interested in helping out with. my circuit design skills are rusty and were never too great to begin with, but I'd like to learn more so maybe I can post schematic of ideas and get feedback for improvement and robustness. I used to be fairly good with C, C++, and VB but it's been 6 years since the last time I tried. I'd like to learn to code for processors so maybe that's something I could pick up around the forums as well. In the very least someone might recommend a good book to get me going.

See you on DIYefi in a month or so.

rb26dett

I think until January I hadn't written any C in 7 years :-)

Grab the 1300 page pdf for the chip and read that :-)

Also, grab the code 0.0.11 is the latest, and read the links that accompany everything that isn't obvious.

You should be OK. Good coding is a mind set IMO :-)

See you in a month!

sotaku

iTrader: (4)

Yeah I certainly don't mean to discount with my reference to politics, more that the phrasing gives a feeling of politics which is a turn off. These things are the very reason why we do need another DIY EFI solution. MS is great for many things but it's getting weird and limited for some of the crazier stuff out there.

Hell yeah, I love projects, which is why I'll keep an eye out and possibly jump in when I'm done with my current stuff.

timk

At the end of the day, you still need a standard hardware platform to work from or it will all fall in a heap. I think this is why MegaSquirt has worked so well; the boards are all the same with the exception of minor hardware hacks. In saying that, I'm going to check out your site and I may even contribute in the future.

rb26dett

That is true, and a std base simple design will be priority number one as soon as the code reaches a suitably mature state. In the mean time this is just a fun project to work on with a sky high goal :-)

Shoot for the stars, fall on the moon?

The Adapt card from Technological arts is a fairly nice little unit. It will be pretty easy to design a pair of boards to sandwich around it that can be printed and populated by hand.

Think of it like a ms2 card, but MUCH bigger :-)

(and not auto temp rated, not that that matters)

I look forward to seeing you there.

Fred.