[Diy_efi] GCC/ARM LPC21xx assist needed for Donegan ECM/EFI :-)

Steven P. Donegan donegan
Mon Jan 15 00:22:10 UTC 2007


The hardware in question are the Olimex boards LPC2106 and LPCH2138 -
both using Phillips ARM chips. The OS I intend to use is FreeRTOS. Both
chips (2106 and 2138) have Capture/Compare registers and there are
'plenty' of available timers. What I want to do is the usual start
timer/enable interrupt, catch interrupt #1, save timer, capture
interrupt #2, save timer, subtract T1 from T2 and I have half the
frequency period (I think). My problem is that although quite saavy in C
and other assembly languages this is my first ARM project.

My toolchain is also Linux/GCC-ARM - FreeRTOS already
compiles/loads/tests out fine on the 2106 board.

Board info for both are at www.olimex.com

After I have the frequency capture part working I'll need to do some D
to A work (for those who want to use a GM MAF with ECU's expecting
flapper or other voltage type MAF's) and frequency generator out for
those that will use the first part of this project as an 'interceptor'
on GM vehicles (ie MAF frequency X in, adjust based on goal and output Y
frequency to spoof the ECU).

I will also need to read the ADC inputs as the project moves on and fire
on/off bits and PWM outputs (relay drivers, injector drivers etc).

TIA

On Sun, 2007-01-14 at 23:54 +0000, Ian Molton wrote:
> Steven P. Donegan wrote:
> > I didn't even really understand what FIQ was, now I do, since I will
> > only need 1 low speed frequency source for the current project (max Hz
> > is <20k) I likely just need a more standard IRQ setup on the CCP/TMR
> > setup.
> > 
> > Any further suggestions gratefully accepted :-)
> 
> Its possibly worth characterising the latency of the chips I/O lines. on 
> some ARMs (notably intels hacked up POS the 'XScale') the turnaround 
> from a state change on an input, to actually setting an output had a 
> quite alarming delay...
> 
> (A signal generator and a 2 trace 'scope would be useful here...)
> 
> if all you're after is a frequency source then yes, the chips internal 
> clock registers should be more than capable - in fact I'd have thought 
> any half decent RTOS could provide a garaunteed programmable timer ?
> 
> If you get some hardware to me, I'd love to help you hack on this one. 
> My development environment is all GNU/linux/GCC here. You can see some 
> of my stuff over on http://www.mnementh.co.uk/eseries/
> 
> Have fun :)
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> Subscribe: http://lists.diy-efi.org/mailman/listinfo/diy_efi
> Main WWW page:  http://www.diy-efi.org/diy_efi





More information about the Diy_efi mailing list