Searching \ for '[PIC] Absolute -> Relocatable' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'Absolute -> Relocatable'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] Absolute -> Relocatable'
2008\07\08@203003 by Jinx

face picon face
Notes for HV_PWM_proto project. It's incomplete as a whole
but the core generator is reasonably finished (well, it's working),
and PWM is being generated. It's proof-of-concept only at this
stage, some functions are not running yet

Variable duty cycle and variable frequency 6-bit PWM that retains
set duty cycle as frequency is varied within a 100Hz - 279Hz range

Basis is a count of 64 TMR0 interrupts

Duty Cycle pot sets the number of consecutive interrupts that are
a '1' on the output pin

Frequency pot changes the reload value of TMR0

On the 64th IRQ, duty cycle is updated

At other values of the IRQ counter, other ADC channels are (will
be) read and the LCD is updated from a print buffer

As the LCD's Busy Flag is not being used, two interrupt periods
separate any data write to the LCD. Writes should complete in
40us, and minimum IRQ period is 56us. Outside of the IRQ system,
LCD processing delays are used

Two "mysterious characters" appear at positions 0 and 1 as soon
as the LCD is updated for the first time under interrupt. Yet to
figure out where they're coming from. They aren't ASCII, look like
random data, maybe their origin will become clear when redefined
characters are used. Over-writing gets rid of them for now

Schematic - voltage limited to 28V for now. Battery maintenance,
regulators, efficiency, 4017 drive etc will be re-visited

2008\07\08@203142 by Jinx

face picon face
part 1 41 bytes content-type:text/plain; (decoded 7bit)

Code


part 2 22611 bytes content-type:application/octet-stream; (decode)

part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2008\07\08@203143 by Jinx

face picon face
part 1 46 bytes content-type:text/plain; (decoded 7bit)

Schematic


part 2 14086 bytes content-type:image/gif; (decode)


part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2008\07\08@203251 by Jinx

face picon face
Olin, code and schematic sent separately as two together are just a
little too big for one post

Also

RAM_START        EQU  H'0000'
START_VECTOR     EQU  H'00000'
PROG_START       EQU  H'00020'
LOW_IRQ_VECTOR   EQU  H'00018'

added to P18F2520.inc

2008\07\09@070302 by olin piclist

face picon face
Jinx wrote:
> Olin, code and schematic sent separately as two together are just a
> little too big for one post

OK, got it.  I've got a customer deadline Friday, so I probably won't do
much with this until then.  I did take a quick peak, and this looks like a
good candidate project as a example of using relocatable mode and my
environment.  1000 is about the right length too, small enough to not be too
much work but big enough to be a meaningful example.

I'm going to take the time to not just convert it, but attempt to really
bring it into the philosophy of my development environment.  I will posts
results here when I have them, which probably won't be for a few days.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2008\07\09@073451 by Jan-Erik Soderholm

face picon face
Olin Lathrop wrote:
> Jinx wrote:
>> Olin, code and schematic sent separately as two together are just a
>> little too big for one post
>
> OK, got it.  I've got a customer deadline Friday, so I probably won't do
> much with this until then.  I did take a quick peak, and this looks like a
> good candidate project as a example of using relocatable mode and my
> environment.  1000 is about the right length too, small enough to not be too
> much work but big enough to be a meaningful example.
>
> I'm going to take the time to not just convert it, but attempt to really
> bring it into the philosophy of my development environment.

Fine.

Then anyone can compare that with the plain-reloc-port I made and
posted just a few minutes ago (hasn't showed up on the list yet).

:-)

Jan-Erik.


2008\07\09@075550 by Jan-Erik Soderholm

face picon face
Jan-Erik Soderholm wrote:
{Quote hidden}

Well, *this* message arrived before the one with my ZIP'ed
reloc project. I've no idea why. It was a 47 KB attachement.
I've mailed Jinx my ZIP'ed MPLAB project privately in case
it will not make it throught the list server.

Jan-Erik.

>

2008\07\09@080321 by Jinx

face picon face
> I will posts results here when I have them, which probably won't be
> for a few days

Absolutely (or should that be relocatably) no hurry at all

If you need any info just holler


2008\07\09@080629 by Jinx

face picon face
> I will posts results here when I have them, which probably won't be
> for a few days

Absolutely (or should that be relocatably) no hurry at all. A couple of
things in it that I'll probably tidy up now I've had a think

If you need any info just holler



2008\07\09@081813 by Jan-Erik Soderholm

face picon face
Jan-Erik Soderholm wrote:

> Jan-Erik Soderholm wrote:

> Well, *this* message arrived before the one with my ZIP'ed
> reloc project. I've no idea why. It was a 47 KB attachement.
> I've mailed Jinx my ZIP'ed MPLAB project privately in case
> it will not make it throught the list server.
>
> Jan-Erik.
>

He, *now* I know... :-)

> "Your message was too big; please trim it to
> less than 40 KB in size."

Anyway, Jinx has got it directly now.

Jan-Erik.

2008\07\09@192118 by andrew kelley

picon face
On the first update FSR1 is not initialized.

        clrf    irq_cnt           ;IRQ counter
                                  ;update duty cycle every 64 IRQs

I do not see any setting of FSR1 before or after that exclusive of the
interrupt, which doesn't occur first around.


Simple solution: add in a comf irq_cnt,f after the clrf.

This was looking at the dead listing, so if its not right, sorry.
--
andrew


On Tue, Jul 8, 2008 at 8:26 PM, Jinx <spam_OUTjoecolquittTakeThisOuTspamclear.net.nz> wrote:
> Two "mysterious characters" appear at positions 0 and 1 as soon
> as the LCD is updated for the first time under interrupt. Yet to

__FIRST TIME__ thats a hint you gave.

> figure out where they're coming from. They aren't ASCII, look like
> random data, maybe their origin will become clear when redefined
> characters are used. Over-writing gets rid of them for now

2008\07\09@200943 by Jinx

face picon face

> I do not see any setting of FSR1 before or after that exclusive of the
> interrupt, which doesn't occur first time around

Well spotted Andrew. Added

        lfsr    fsr1,print_buffer
        clrf    irq_cnt

before gie is set. As you mentioned, and I hinted at to myself, it happens
on the first LCD update. The characters shown was also a clue, as they
varied in shape. I'm usually pretty good with variable initialisation,
having
being caught before, and probably would have found it sooner or later

Thanks for making later = sooner. Help yourself to a gold star

2008\07\10@203414 by Charles Rogers

flavicon
face
Sent: Tuesday, July 08, 2008 7:29 PM
Subject: [PIC] Absolute -> Relocatable


> Schematic
>
Jinx:

How do you make your PCB's ?

CR

2008\07\10@205353 by Jinx

face picon face
> Jinx:
>
> How do you make your PCB's ?

http://home.clear.net.nz/pages/joecolquitt/0makepcb.html

2008\07\10@223149 by Justin Richards

face picon face
>
> OK, got it.  I've got a customer deadline Friday, so I probably won't do
> much with this until then.

Olin (and others),

I really dont know how any of you can meet any deadlines with all the
effort you put into the Piclist and for that I thankyou all.  The
amount of effort you folks put in must far out weigh what you get out.

I spend a great deal of my spare time (and I am on holidays) just
reading the emails.

I am looking forward to the development of this thread.

Thanks Again.

Justin

More... (looser matching)
- Last day of these posts
- In 2008 , 2009 only
- Today
- New search...