Fred's firmware development diary

[Fred]

Sorry about the long wait! It was a fairly large effort to get everything in order in my current situation. Thanks to all of those who helped in various ways!

BrickEMS.com DIY HW: viewtopic.php?f=62&t=1780
BrickRPM Dual VR Board: viewtopic.php?f=62&t=1781

Fred.

[Fred]

Interesting case of dodgy hardware the other night:

viewtopic.php?f=15&t=1783

I feel sorry for Peter! :-(

Fred.

[Fred]

Congratulations to Peter for engine number 12 on FreeEMS, and also for being the first person other than me to FreeEMS more than one engine! Wicked! ♥

http://www.youtube.com/watch?v=nwwUW7BUV1g

viewtopic.php?f=55&t=1787

Love it!

Fred.

[Fred]

Just six weeks in, and already we've got live tunable 2D tables with visualisation. They need some polishing, but this morning is the first time that I've ever even seen them outside of source code and open office!

Upper left: Dwell vs Voltage
Lower left: Dead time vs Voltage
Upper right: Priming PW vs Temp
Lower right: Engine Temp Enrichment



This could be the official tuner in as little as two weeks if things keep going the way they are! Exciting times.

Fred.

[Fred]

I realised earlier today that no matter how good and usable and awesome EMStudio turns out, I still need to do my own tool! I need it for rapid prototyping of interface changes and development of data formats. I can't reasonably or efficiently ask someone else to assist in these matters. The only reasonable thing to do is continue development of my tool into something stable and usable.

In the unlikely event that Mike throws a giant child-like tanty and accuses me of trying to cause him to divorce it'll come in handy :-)

Fred.

[Fred]

I stayed up late last night to see a friend off to the airport, so while waiting, I got stuck in and did some firmware work. I probably shouldn't have, as I can feel it today, but I did, and it's nearly done. Push soon, I guess, and then some more commits leading towards the Deuce running right.

[Fred]

Finally finished and pushed that sensor commit and another couple of commits, yesterday. Today I've been working on a few more, just need to test and push, all cleaned up and ready to go already. Might hold off posting until I've pushed it and suggest trying it. It comes pre-configured to turn on two PORTK pins above 100C CHT and 5k RPM with 10C and 1k hysteresis, respectively. Change as you please, but keep in mind interactions with other code/settings. There is nothing to stop you setting up two things on the same pin or one pin on a specific port on interrupt code and another not. There will be later, but not until after 0.4.0 is released and work on 0.5.0 has started. Beware! :-)

Waited all night, test rig issues, 220am, will test tomorrow, somehow, and push. Sorry for cop out :-/ DB37 issues? Wouldn't that be ironic! LOL. :-/

Fred.

[Fred]

Puma fixed, dry joint on regulator, dodgy Marcos-spec solder joint + lack of strain relief = fail. Soldered well now. And working again. New code works too. And is pushed.

Tracking my vague TODO here: viewtopic.php?f=8&t=57&start=590

In addition to those, the missing tooth suspected bug needs dealing with: http://issues.freeems.org/view.php?id=594

58% of 0.2.0 is currently completed. More soon.

Fred.

[Fred]

Just fixed this with a tiny tweak for Peter: http://issues.freeems.org/view.php?id=570

[Fred]

Another epic skype session and Mike has 3D tables represented in tabular view with colouring. Nearly 5am for me. Dedication, I tell you. Incidentally, this is the tune that runs Sean's LT1 Camaro, IE, it's a non-tune tune... and yet it runs well ;-)



Fred.