[Diy_efi] EEPROM based efi compter, DIY style.

Wed Dec 20 18:39:14 UTC 2006

The PICs are horrible to program, unless you go to the 16 or 18 series 
and can use C, in which case you pretty much have to purchase a compiler.

The AVR is ok, though expensive for what they are, and if you wanted a 
large FLASH for tables, then you start to fight the architecture just to 
address the memory.

The MSP430 are good, but the higher end devices cost too much for what 
they have.

I've never used a Zilog Z8 micro, however Olimex also have a US$40 
development kit for the LPC2103 Philips ARM.  The JTAG connector is 
separate and costs another US$20 but is usable on all ARMs.  The ARMs 
also have available the GCC C compiler, which just can't be beaten for 
value for money.

The ARMs aren't that difficult to program, and if you're just learning 
are no harder than an alternative micro.  The ARM7s are also really 
really cheap, and you can generally get samples of them pretty easily.  
http://www.olimex.com have some good development boards.

I'd still recommend the ARM, it's nice to not have to worry about FLASH 
size or algorithm complexity, which for a beginner is probably the best 
thing.  A beginner can't approximate the compiled code size from lines 
of C code, nor can they truely comprehend the number of operations 
required for particular C constructs and so can't estimate the algorithm 
complexity in many situations.

Bevan
> For something this simple an 8bit chip would far easier to work with 
> than an ARM chip, especially for a beginner.  An ARM7 would be overkill.
>
> PIC is the most popular and then probably the AVR for hobby 
> electronics. There's plenty of Google info on these and include 
> ADC's.  I'd recommend the Zilog Z8 Encore!  $40 development kit from 
> Digikey or Mouser.  Comes with the programming cable and a simple 
> development/experiment board.  Free toolset including C compiler and 
> assembler, linker, in circuit debugger, and full development 
> environment from Zilog.
>
> Mark Higgins
>
> ----- Original Message ----- From: "Bevan Weiss" <kaizen__ at hotmail.com>
> To: <diy_efi at diy-efi.org>
> Sent: Wednesday, December 20, 2006 2:44 AM
> Subject: Re: [Diy_efi] EEPROM based efi compter, DIY style.
>
>
>> Hi Ben,
>>
>> This certainly is most easily done using a microcontroller.
>> If you obtain a micro with ADC inputs then you can easily convert 
>> from an analog voltage into the digital domain, the micro can then 
>> operate on this to provide a pulse width modulated (PWM) output 
>> signal (either via dedicated PWM hardware, or through a software PWM 
>> implementation).
>>
>> The main difficulty becomes conditioning the analog signal for input 
>> to the ADCs.  Things like O2 sensors don't have a very straight 
>> forward signal path and so additional control is required for them.
>> Most other sensors are relatively simple, being ratiometric to supply 
>> voltage, and hence only require an appropriate supply voltage to give 
>> a sufficient output signal.
>>
>> An EEPROM chip itself is capable of very little, it certainly can't 
>> produce a PWM output unless additional circuitry is used around it.  
>> I would strongly favour a microcontroller for this task.
>>
>> With micros like the ARM7TDMI readily available for very cheap this 
>> is not a difficult task.
>>
>> You would however want to pick up another programming language (C 
>> preferably, though some ARM assembler would likely be beneficial 
>> too). You would also want additional hardware development skills, but 
>> it would be an interesting project to learn on.
>>
>> If you're just after something plug and play, then you should look 
>> more towards the megasquirt range that Bruce et al designed.  If 
>> you're just after the end product it's the way to go.  If you're 
>> after the learning then doing it yourself would certainly be 
>> enlightening, however I would caution that it will never be as 'easy' 
>> as you first imagine.
>>
>>
>> Regards,
>> Bevan
>>
>>
>>> Hi people.
>>>
>>> Just another of my less crazy ideas, I was wondering what it would 
>>> take to build a computer that can make a pulsewidth output based 
>>> primarily on a voltage (ie air flow meter).
>>>
>>> The idea would be to have four or five tables:
>>> 1: Air-flow/Pulsewidth (with pulsewidth the output)
>>> 2: TPS/RPM (pulsewidth output, overrides table one while rpm AND 
>>> throttle position are in range)
>>> 3: TPS change/Time (pulsewidth output, added to table one/two)
>>> 4: Temperature/Pulsewidth multiplier (simple warmup strategy, or a 
>>> master rich/lean knob)
>>> 5: O2/Pulsewidth multiplier (simple closed loop under certain TPS 
>>> and temp conditions)
>>>
>>> I can see this being a simple computer for converting naturally 
>>> aspirated engines to single point efi, with a minimum of wiring 
>>> required, primarily for economy.
>>>
>>> so how would you go about building something that can do this? My 
>>> electronics ability is limited to assembling kits and understanding 
>>> how simple components (transistors, capacitors, diodes etc) work, 
>>> along with limited BASIC programming. My main interest is in how to 
>>> program the EEPROM chip, and how to get the inputs to the correct 
>>> format (ie frequency is easy enough to do, but converting a voltage 
>>> into a digital signal is something I know nothing about). I am 
>>> presuming that an EEPROM chip can make a pulsewidth output directly 
>>> (with the aid of a transistor of course), is this correct?
>>>
>>> thanks
>>> Ben
>>
>> _______________________________________________
>> 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 
>
> _______________________________________________
> 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
>
>