Upcoming Changes to Open-Robot's Main Controller Board

[abotics]

At this time I am in the process of making several changes to Open-
Robot's Main Controller Board. These changes include the following
items:

1). Add terminal blocks for connecting power-in, WW02 Wheel encoders,
Tx Rx & +5v to MatchPort board, and remaining free I/O. This will make
it very easy to connect up wires and additional hardware/sensors.

2). Remove 1k resistors that connect PIC18F4520 to SN754410 h-bridge.
Less soldering and small cost reduction.

3). Eliminate +5v regulator that supplys SN754410 and simply use
unregulated battery voltage for the (2)-GM8 motors. Cost reduction and
should extend battery life.

4). Replace the +5v regulator that supplies logic level voltage to the
board with a high-efficiencey RECOM regulator. Increase cost, but will
extend battery life.

5). Remove 6-position 90-deg header (J2) and replace with a RJ11 6-pin
jack. This will eliminate the need for a special programming cable and
allow CCS ICD users to directly connect the ICD cable to Open-Robot
for programming and debugging purposes.

In general, everything is going to remain similar to the version 1
board, but I think these enhancements will increase Open-Robot's
flexibility and provide the user with a better experience. I expect to
have sample PCBs by the end of this week and will post the results of
my initial testing with this new board design. At some point I intend
to perform a redesign of the MatchPort Wireless board, but that might
be a little while after I roll out the new Main Controller board
design.

[abotics]

I just posted a picture of the new board design assembled with
components, "open_robot_main_board_v2.jpg". To view this picture look
under Files. Everything appears to work well. Have not tested the new
program/debug connector with the CCS ICD and I still need to connect
up a pair of WW02 wheel encoders. Otherwise it all works.

http://groups.google.com/group/open-robot/files

I was hoping to be able to use the terminal blocks for the WW02 Wheel
Encoder connection, but I don't think there is enough room to get the
wires in. I might look for a 45-degree terminal block, but maybe we'll
have to stick with the polarized connector and corresponding housing
with crimps. I wanted to get away from this b/c you really need to
have access to a crimp tool. However, you can also solder the WW02
wires directly to the board.

Regards, Abe

[abotics]

Well, everything tested out well on the new board, that is except for
the RJ-11 program/debug ICD connector. I inadvertently miswired the
traces. The WW02 encoders work as expected, but I don't think we'll be
able to go with the terminal blocks because I was unable to find a set
of 0.1" pitch blocks that had a 45 or 55 degree angle, which is needed
to clear either the PIC18F or RFID module. At this time I will just
leave the old polarized headers. The terminal blocks are fairly
expensive in relation to the polarized headers, so this is not a big
deal. I just submitted an order for another batch of PCBs, which
should arrive by the end of this week. These will have the RJ-11
connector fix along with several other very minor changes. I moved the
PIC18F Tx, Rx, and +5volt terminal block over by the h-bridge, so that
it is closer to the MatchPort wireless board's connector. I also moved
the Gnd and +5volt headers so that they are 0.1" apart, so that end-
users can simply use a 2-position non-polarized header to connect up
and power additional hardware. As soon as I have an assembled PCB I'll
post a picture. I am hoping that this will be the last board iteration
before I order a production run of the PCBs.

Abe

> > design.- Hide quoted text -
>
> - Show quoted text -

[abotics]

Just finished soldering up the new PCB and everything appears to work
great! I posted an image of the board layout,
"open_robot_v2_board.jpg".

http://groups.google.com/group/open-robot/files

At this point I don't expect to find a problem, but further testing is
required. Once fully verified, I will let everyone know when to expect
the new boards to be available for purchase.

Regards, Abe

[abotics]

After some additional testing I feel that this version of the board is
good. I did end up having to retune the PID control slightly and also
while I was in there changing code I went ahead and added additional
user control by providing access to the position stop constant. This
parameter determines how fast the robot accelerates/decelerates when
performing PID position control. If you are performing larger moves
you can set this number higher and lower for small moves. Since I
removed the +5 volt regulator that previously supplied the motor
voltage we now use whatever is being provided by the battery, so you
do have to set the PWM Motor Offsets to around 160. Previously the PWM
offsets were around 195, so we have a wider range of PWM control and
in open-loop control you can really get the robot to move along
quickly.

I still have batch of the old boards remaining if anyone is still
interested in sticking with that design because maybe you have several
robots that you are using in your research or course projects and want
to keep everything the same. I have not yet placed an order for a
batch of new boards because I wanted to finish all tuning and be 100%
sure that the design is solid. I still have to revise and update the
Open-Robot Owner's manual with the new board design as well. I will
post more information as things progress.

Abe

> > > - Show quoted text -- Hide quoted text -