DFH - Defacto FreeEMS Hardware in KICAD

Jared's unmaintained and never-used TA based "Defacto FreeEMS Hardware" design.
jonr
QFP80 - Contributor
Posts: 46
Joined: Mon Aug 29, 2011 12:59 pm

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by jonr »

Sorry, I'm way behind - I found it difficult to find a good overview of this project - current hardware design, current software, etc. I've seen http://wiki.freeems.org/doku.php?id=freeems_dfh since and I love the stack/IO bus idea. Going forward, I think it would be interesting to maintain and develop the software on multiple CPU platforms (ie, imagine an adapter board that allows a LPC card to replace the Technological Arts card). This would help keep it portable. But it only works if the lead developer is committed to the idea of portability being a priority.

Another interesting thing about the EEC IV sw was that it made extensive use of fixed point fractional arithmetic. Quick example for non-programmers: 1.25 might be stored as 1010 bin 3 (1 + 0*.5 + 1*.25 + 0*.125). You can multiply by another bin 3 number (say 1010 also) and get 1100100 (1.5625 stored as bin 6). http://en.wikipedia.org/wiki/Fixed-point_arithmetic
Open5xxxECU.org
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by jharvey »

Yes the idea of a swap-able brain is one of the motivators behind DFH. At the moment the best developed brain is the TA card and Fred's code, however you are correct that other brains could be installed.

One idea I like about most ARM's is that you can DMA rake a piece of memory across the outputs, independent of the CPU cycles. Such that you can layout an obscure series of pulses in a piece of memory, then rake them across the IO to make a P&H injector, or what ever you have. You could add some fairly complicated features like softening the pintle's (sp) end of travel, or a PID like control when changing from the peak to hold transitions. Lots of handy architectural options if one were to ask me.

If you interested in thinking of an LPC approach, I'd be willing to chat about it some. Do you have software skills and a desire to write code that might back such an approach?
jonr
QFP80 - Contributor
Posts: 46
Joined: Mon Aug 29, 2011 12:59 pm

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by jonr »

Interesting, although when I do the math, it could take a lot of memory. Say I want a max period of 1 sec (engine cranking) with 1 usec (ignition timing) accuracy. I know that the LPC has some timers that can cause output pins to go on/off without any cpu involvement.

I am a software guy and I'm thinking how I could best contribute. I do believe that truly open/free software needs to be portable.
Open5xxxECU.org
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by jharvey »

Best contribute is probably to help Fred with code.

About the math, I see 360 deg per rotation with infamous .1 degree accuracy. The key is to figure out if that's on board or not, ect. The DMA clocking is based on the RPM input, so it clocks per RPM, not per XTAL.
jonr
QFP80 - Contributor
Posts: 46
Joined: Mon Aug 29, 2011 12:59 pm

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by jonr »

If I understand things correctly, Sean has written code to allow scheduling of any output event at .8 usec accuracy using a periodic interrupt service routine. So no need for DMA or special timers.
Open5xxxECU.org
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by Fred »

jonr wrote:I found it difficult to find a good overview of this project - current hardware design, current software, etc.
jonr wrote:I am a software guy and I'm thinking how I could best contribute.
I see that you've answered your own question (before even asking it! Impressive) :-)

The firmware is currently frozen. No help required on that right now. Focus for the time being is on docs and hardware details. It's unlikely that I can work legally before about 4 months time, so I've got no shortage of firmware hours for the time being, once I unfreeze it.
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
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by Fred »

Portability discussion moved to here.
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: DFH - Defacto FreeEMS Hardware in KICAD

Post by jharvey »

Some people have expressed interest in DFH. There are many items that should be addressed, which shouldn't take me long to do. Perhaps an hour to two. So I figured it would be shame if I didn't spend such a small amount of time to make it so much easier for other interested persons.

So far I removed that hacked hall sensor crap, as it should be done with the Allegro chip. I don't plan to put on the Allegro chip, as I don't know anyone that's planning to support it via software. I also started changing the RPM circuit to include the MAX chip instead of the LM chip.

I also plan to change the AN protects for the 470R and such as you noted in your empirical experiments.

Once that's done, it should be a matter of finishing the routing signals to TA pins, and double checking stuff.
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by jharvey »

Here's DFH for FreeEMS A.21 P.01 for review

https://github.com/jharvey/DFH_FreeEMS_1.0_hardware

For those with out KICAD, I've released a full PDF with the PCB layout. That can be found at these links.

Schematic PDF https://github.com/jharvey/DFH_FreeEMS_ ... f?raw=true
PCB PDF https://github.com/jharvey/DFH_FreeEMS_ ... f?raw=true

With it's 21st revision, I felt it was time for DFH to man up a bit. I think it's now matured enough that it could potentially be useful for trips to the local store, ect. I routed injector and ignition circuits to TA pins, I removed a bunch of junk, that was intended only for me as a science experiment, and updated the AN and DIGI protect circuits per Fred's experience with PUMA. Basically the brain is now connected to features.

Some key notes, until XGate makes more progress, I put the injectors in parallel with the ignition circuits, such that you only have 6 total channels of either fuel or spark, and you select if it's fuel or spark by populating the components. I plan to allow for folks to change that at a later date, by breaking some traces, and installing jumper wires to vias I've included.

Future planned improvements include changing the sub part LM VR chip. for JBPerf's VR board.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DFH - Defacto FreeEMS Hardware in KICAD

Post by Fred »

Jared, FANTASTIC to see you working on this! Changes sound TOP NOTCH too, I'd LOVE to see this man up and get some real use :-) I'd rock one with my dev TA for sure! :-)
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!
Post Reply