Fred's firmware development diary

[Fred]

Commits are rolling in again. I've not been idle, but it's been difficult to complete any work for a while. I'm not semi setup with a desk and a few cushions to stop my wrists getting ruined, and a power supply that I stole from AbeFM, hooray.

Exciting news, BenchTest decoder is a reality:

viewtopic.php?f=56&t=1180&p=15142#p15142

That's good for anyone doing anything much on the bench. I'll expand it at a later date with data manipulation specific decoder emulation and other neat stuff. For now, it's useful to test the physical outputs, which is cool, as we really needed that. It would have helped get one more car running, if we'd had it a month or two ago. In any case, it's here now.

Hopefully I can make more progress in various areas from this point on. I've done quite a lot of stuff, but there is a lot more needing to be done to satisfy my requirements.

Give the bench test stuff a shot, and: Enjoy!

Fred.

[Fred]

Clearing out my browser situation...

These links show why the voltage vs. dead time curve is not sufficient for large injectors on small engines, our target goal.

http://www.injectordynamics.com/LowPulseTech.html
http://www.injectordynamics.com/fordcal.html
http://www.injectordynamics.com/GMDATA.html

So, I need to sit down and think about the best way to handle this in future. For now, though, the current table is good enough for most intents and purposes.

Fred.

[Fred]

A FAQ is up, give suggestions in the comments section of the site, in the thread on doc improvements, please!

http://faq.freeems.org

Fred.

[Fred]

I've been putting some time into the Mantis installation recently. Getting rid of old rubbish, adding new stuff, setting things up better, etc.

Today I branded it! So now it's all up to date, and branded with the FreeEMS logo! :-)

http://issues.freeems.org

I'm in the process of adding a bunch of issues to cover all known stuff with the firmware. Hopefully I can cut that down to size and keep it free of bugs and full of improvements and tasks :-)

Fred.

[Fred]

Yesterday I added about 53 issues of various types to the tracker! Most stuff is documented there now. Today I'm going to get some work done on some of them. I closed one today already and I've not even eaten breakfast! I will continue to add more as I find things and attempt to keep what is in there up to date too. I'm focussing on docs, easy stuff and low risk stuff until I have some testers helping me again. I've not heard any action from either of my current testers for some time. In fact, I've not heard from one of those testers at all for a while. I might phone him soon.

More as it comes to hand. Back to web sites and docs for me post todays simple efforts, I think.

Fred.

[Fred]

http://status.freeems.org/firmware/decoders/FreeEMS/
viewtopic.php?f=56&t=1204
http://issues.freeems.org/view.php?id=187

Work has begun, very exciting! :-)

Fred.

[Fred]

I've been getting a lot of overdue maintenance, organisational work and documentation stuff done recently, which has led to my decision today to do a partial code freeze on the firmware. One aspect of this is that I've discovered an area where the project is vulnerable, and I wish to close that gap and make it absolutely water tight, for the security of future users, and my own peace of mind.

Starting now, 4pm Spanish time, June 29, 2011, I'll only be doing firmware work for real world users, with hardware, attached to an engine. Additionally, I'll not be accepting any further contributions from any authors for the time being. This will give me the time that I need to bring the overall project up to the level of professionalism required to really stand out.

Along with the work that I've already been doing, I'm going to put some significant time into formalising the hardware interface documentation for those design and/or hardware vendors who wish to be "FreeEMS Compliant", and to allow new designs to begin without fear of a changing foundation.

As I see it, right now, our weakest area, by far, is hardware, with no solid, well tested, conservative, usable, reliable, documented platform readily available. This must change, and soon.

Please bear with me while I get this absolutely essential work done!

Fred.

[Fred]

Along with the work that I've already been doing, I'm going to put some significant time into formalising the hardware interface documentation for those design and/or hardware vendors who wish to be "FreeEMS Compliant", and to allow new designs to begin without fear of a changing foundation.

As I see it, right now, our weakest area, by far, is hardware, with no solid, well tested, conservative, usable, reliable, documented platform readily available. This must change, and soon.

Well, here it is, my first stab at a FreeEMS hardware design! Hopefully I can approach this with the same level of thoroughness and professionalism with which I approach software problems! If not, I'm sure that others can help! :-)

Introducing Project Pragmatic, AKA "Cheetah" viewtopic.php?f=9&t=1225

Fred.

[Fred]

Today I spent mostly on the forum! I read the entire 35 page DFH thread and took a few minor things from it. I posted here there and everywhere and got some good feedback from Jared, thanks! I still need to read the whole Puma dev thread, taking notes as I go, and check out dvisser's pdf and comment on that.

Dave has formalised his MTX todo list on the issue tracker, which is great! Some of his ideas sound excellent! :-)

Hardware progress is taking leaps and bounds with Dan on the scene and me taking a hands on approach and incorporating lessons learned.

I'm looking forward to getting some good work done on it in the near future.

I also bought a (filthy cheap) multimeter and a screwdriver today. I feel less naked now. I've got a few ideas on hw testing that I can do with the available equipment now, which is good. Perhaps tomorrow I'll do some testing along with more work on the new "cheetah" dev box and the usual array of FreeEMS chores.

Fred.

[Fred]

A neat utility that I just apt-getted in Debian:

http://srecord.sourceforge.net/

Great for manipulating firmware files.