[Coco] Re: 68B09P - Datasheet?

[Andrew]

Lumpin' a few post replys in one message...

> From: John Donaldson <johnadonaldson at comcast.net>

>    If these boards are configured as a CPU board and I/O boards, You 
> might be
> able to use them as is. You could make NEW EPROMS using Boisy's OS9 ROM
> and use the I/O boards under NITROS9. You may have to do some rewrite of the
> I/O drivers to match them up with the I/O boards, but you would get a 
> lot of support
> here on this form. Remember OS9 started out as a RTOS, which is a Real Time
> Operating System. A RTOS is what you want for a robotic project, if the 
> robot is
> going to be the type that runs on it's own and now via a radio/joystick 
> link.
>   You could even write the main routines in Basic09.

This sounds like a good idea, however, the boards are waaay too big for 
my project (they measure close to 15 x 17 inches), though they appear to 
have memory, the CPU, glue logic, and a separate EPROM - they are 
complete. Just too large. A smaller (about the size of a Disk Controller 
cartridge) board with just a 68B09P and some logic (though no RAM or 
EPROM, as far as I can tell) I also found, some kind of daughterboard.

Since I would have to reconfigure and rebuild a new configuration to get 
it in the size I need, plus purchase an EPROM burner (not a part of my 
kit, currently) - this may be a more expensive option than I want.

Furthermore, I still need the datasheets - without them, I am lost.

> From: Mark McDougall <msmcdoug at optushome.com.au>

> Depends on what sort of I/O you're going to need for your project. For 
> example, if you need A/D or D/A, a few timers, multiple interrupt sources, 
> comparator, edge capture, PWM, and/or GPIO, then you might be better off 
> finding a PIC or other microcontroller that suits your requirements.

My project mainly needs a small microcontroller to handle interfacing of 
PWM signals to the motors and servos on the robot. It is an ROV, using 
radio control right now with a wireless camera mounted on a custom 
pan/tilt servo mechanism (fun to chase the dog with!) - but I want to 
extend the range and functionality, plus give it some autonomy, via 
interfacing all of that with a small form factor PC motherboard and 
802.11x link. So, I figure a small microcontroller handling PWM for the 
motors and servos, interfaced via serial to the main motherboard. This 
machine will likely run Linux and have Apache on it, so I can use a 
browser to control the robot via a wireless CGI based interface (w/ 
video4linux vidcap from the camera, as well). Later, perhaps, a fully 
custom interface using python and networking (eliminating Apache and 
CGI, or supplementing it), and perhaps coupled with an HMD for 
telerobotics...

> If, OTOH, you don't really need any of that, but could do with plenty of 
> ROM/RAM, 2MHz is fast enough, and there's enough (S)RAM & (E)EPROM on the 
> PCBs then they may well do the job. There may even be suitable (digital) 
> GPIO if it's a gaming machine motherboard. Keep in mind you will probably 
> not be programming the EEPROMS in-circuit, which can be a pain (slow) even 
> with ZIF sockets on board. Having said that, a flash-based pic may not be 
> that much better if it's not in-circuit-programmable either.

I am not sure if the PICs I own are in-circuit-programmable, either - I 
know they are 16F84 PICs - but I think those have since come in other 
models which may have better support for ICP (I bought mine a while 
back) - not sure. However, since I am only using them as "slave 
controllers" (ie, basically homebrewing my own PWM servo/motor 
controllers - yeah, I could buy such things, but I want to build my own, 
and OTS controllers are expensive, and I am cheap).

> Software-wise, I'd prefer 6809 assembler over PIC assembler any day, though 
> I can't comment on PIC BASIC. AFAIK the Zaks 6809 book is considered the 
> "bible". Shouldn't be too hard to pickup 6809 if you're familiar with any 
> other assembler?!?

My greatest familiarity is with 80x86 assembler, but I have dabbled with 
6809 assembler in the past, as well as 68000 and 6502 - so I am not 
unfamiliar with low-level coding (a lot of the 6502 coding was on an 
Apple IIe typing hex codes into monitor/debugger to interface via BASIC).

From: Frank Pittel <fwp at deepthought.com>

 > Never cared much for pics. You may want to consider using a 
micro-controller
 > such as the 68hc11 or 68hc12. The core processor is very similar to 
the 6809
 > and the chips have built in I/O ports, eeprom and ram. Do a google 
search on
 > either 68hc11/68hc12 and prototype and you'll see a lot of good 
boards for a
 > reasonable amount of money.

I have considered other microcontrollers, but all of them seem "out of 
reach" for me from a money perspective. Since I am planning on using the 
microcontroller as a slave to a PC, and it will be a "one shot" 
programmed deal, a PIC seemed the way to go initially until I saw those 
6809s - but now they don't seem like a good deal for this project. I 
might investigate those other controllers you mention - my only hang up 
with them is that development/programmers from what I remember are a lot 
more expensive than for a PIC (for many PICs, you can build a simple 
serial port programmer from junk parts - a few resistors, a transistor 
or two, some diodes, etc). Plus, I already have the PICs, so it is a 
"sunk" cost, I consider them nearly "free" at this point.

All - thank you for your suggestions - I think at this point I will stay 
on course with the PIC, though I might look into Frank's 68hc11 
suggestions as well. A 6809 looks like overkill (from a setup, 
development, and design perspective) for what I need the microcontroller 
for (it would be a different thing if it was to be the main controller 
of the robot, but like I said, that will be reserved for a PC motherboard).

Thanks again,

Andrew