John,

I know Rabbit used to give away the RFU source. Probably not anymore, but
it's worth a try asking them?

-- Dave

j...@softsmart.co.za wrote:
> Hi,
>
> Does anyone know of the protocol used to reprogram rabbits (RCM3700)?
> I have seen some related threads here, but my problem is a little
> more specific (I think).
>
> We install equipment on the gold mines of South Africa, and the
> problem is two-fold.
>
> 1) Our equipment is not easily accessible, ie, gold plants are highly
> secure areas and no laptops are allowed. Underground (3kms) is just a
> pain in the rear end.
>
> 2) Our rabbit boards do not all have a direct connection to the PC.
> They are hierachical. There are rabbit based controllers which
> control populations of rabbit based boards (RS485).
>
> So, I need to try and implement a serial bootloader, but I have no
> idea how. We are placing atmel memory chips on the boards, so the
> plan is to have the PC software send the new program to the
> controller, and it will just write the new bin file to memory. Once
> this is done, we have a small pic that will take over and start the
> programing sequence. Once this is done, the updated controller will
> then 'pass on' this program to its population, and they will
> reprogram themselves in a similar way.
>
> I am stuck on the programing protocol though. I know that the SMODE
> pins should be pulled high. I know that a bootloader is sent to the
> rabbit at 2400 baud, then a pilot is sent at 57600 baud, followed by
> the application at 115200 baud.
>
> But I don't know exactly how to send it, and what replies come back
> from the rabbit.
>
> Does anyone know how to do this, or where I can get some info on this?
>
> Regards, and thanks,
>
> John



(You need to be a member of rabbit-semi -- send a blank email to rabbit-semi-subscribe@yahoogroups.com )

Would need a long programming cable if laptops are not allowed. I
found it can be easier to go into the site with newly programmed
Rabbits, power down, install new Rabbits, power up again. Faster than
most bootloading methods, and if you find the new code isn't working
as it should just switch back again.

PROS:
* Provides backups if things go wrong, just switch the Rabbits back
* Very simple if this is a task not done often and in an environment
where access is a bit tricky
* The process of removing one Rabbit and inserting another is often
faster than reflashing chips
* FAR easier to program

CONS:
* Difficult when the processors themselves are hard to reach
* Requires purchasing extra Rabbits
* It helps to get setup parameters from the existing Rabbits and
transfer them to the new Rabbits prior to instalation
--- In r...@yahoogroups.com, "Dave Moore" wrote:
>
> John,
>
> I know Rabbit used to give away the RFU source. Probably not
anymore, but
> it's worth a try asking them?
>
> -- Dave
>
> john@... wrote:
> > Hi,
> >
> > Does anyone know of the protocol used to reprogram rabbits
(RCM3700)?
> > I have seen some related threads here, but my problem is a little
> > more specific (I think).
> >
> > We install equipment on the gold mines of South Africa, and the
> > problem is two-fold.
> >
> > 1) Our equipment is not easily accessible, ie, gold plants are
highly
> > secure areas and no laptops are allowed. Underground (3kms) is
just a
> > pain in the rear end.
> >
> > 2) Our rabbit boards do not all have a direct connection to the
PC.
> > They are hierachical. There are rabbit based controllers which
> > control populations of rabbit based boards (RS485).
> >
> > So, I need to try and implement a serial bootloader, but I have no
> > idea how. We are placing atmel memory chips on the boards, so the
> > plan is to have the PC software send the new program to the
> > controller, and it will just write the new bin file to memory.
Once
> > this is done, we have a small pic that will take over and start
the
> > programing sequence. Once this is done, the updated controller
will
> > then 'pass on' this program to its population, and they will
> > reprogram themselves in a similar way.
> >
> > I am stuck on the programing protocol though. I know that the
SMODE
> > pins should be pulled high. I know that a bootloader is sent to
the
> > rabbit at 2400 baud, then a pilot is sent at 57600 baud, followed
by
> > the application at 115200 baud.
> >
> > But I don't know exactly how to send it, and what replies come
back
> > from the rabbit.
> >
> > Does anyone know how to do this, or where I can get some info on
this?
> >
> > Regards, and thanks,
> >
> > John
> >
>



(You need to be a member of rabbit-semi -- send a blank email to rabbit-semi-subscribe@yahoogroups.com )