Servo Board V4 - SBv4

[Rich Barlow]

Hi all,

It's that time again. With essentially no time to go before SR2015
begins I'm designing a board last minute.

This time it's the new and improved Servo Board. There are some notes
about the originally spec'd feature set on trac[1]. A few things have
changed and I'll be updating that page soon to reflect the changes. In
the mean time here's a summary of where things stand:

* USB control interface (separate to UART interface, unlike on MBv4b)
* UART control interface
* Isolation between control interfaces and servo outputs
* Built in 12A 5.5V SMPS with current/overtemp limiting
* Voltage/current monitoring of SMPS
* 12 servo outputs*
* Optional 'auxiliary' supply
* Oodles of blue LEDs!

* The 12 outputs are split into a group of 8 and a group of 4. The group
of 8 are always powered by the internal SMPS. The group of 4 can be
powered by either the internal SMPS or the optional 'auxiliary' supply.
This allows teams to either provide more 5V beef or a higher/lower
supply voltage for more exotic servos.

I've attached the schematic (since it's so small). I've just finished
assigning values to components so, in theory, it's finished. The next
step is to draw a few footprints and then begin the layout.

As usual, I would greatly appreciate any review of the design.

Rich

[1] https://www.studentrobotics.org/trac/wiki/Servo%20Board%204

[Jeremy Morse]

Hi,

This all looks good to me, although I'm not an excellent source of
electronics review. I particularly enjoy breaking all the IO operation
out to an expansion IC / whateveryoucallit.

Some of the datasheet for the 5v SMPS is devoted to it's behaviour when
hot; will air-holes in the case be necessary again? (I imagine it's not
as much of a problem as with the motor boards).

Thanks for pouring effort into this.

--
Thanks,
Jeremy

[Alistair Lynn]

Hi Rich–

Looks good, thanks for getting this done. Only a couple of points to make here:

1. The motor board had a large number of test points, but there don't
seem to be any in this schematic?
2. What's the procedure for flashing a bricked board in a school?
3. Will the board be damaged if a 5V output pin is shorted to ground?
4. What happens if a servo is plugged in backwards?
5. There's quite a bit of capacitance on 5V5_SMPS, could that lead to
any inrush current issues with the SMPS itself?

Alistair

> --
> You received this message because you are subscribed to the Google Groups "Student Robotics Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to srobo-devel...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Rich Barlow]

On Mon, 2014-08-04 at 13:41 +0100, Alistair Lynn wrote:
> 1. The motor board had a large number of test points, but there don't
> seem to be any in this schematic?



> 2. What's the procedure for flashing a bricked board in a school?

Either we replace it, they use a USB to serial cable or we ship them a
USB to serial cable.

> 3. Will the board be damaged if a 5V output pin is shorted to ground?

No. The SMPS has short-circuit protection.

> 4. What happens if a servo is plugged in backwards?

I don't know. It might be OK, it might not be. There's not really much
we can do to handle the crap connector design*. I suspect it'll be fine,
as the GND wire will be connected through a 330R resistor. If it tries
to draw current the GND wire will rise towards 5V. The only thing is
that the control line will see a negative voltage. Assuming (which is a
big assumption that these things are actually decent) the control line
has a clamping diode in the servo, it will actually clamp the GND wire
at 0.7V above ground. Either way, the control line will never see any
pulses and it won't actually try to draw a large amount of current.

> 5. There's quite a bit of capacitance on 5V5_SMPS, could that lead to
> any inrush current issues with the SMPS itself?

That's what the SMPS requires to operate correctly.

Rich

*The connectors aren't totally crap, only mostly. They do have some sort
of polarization. There are two variations: Futaba, which has a little
tab at one end, and all the others, which have teensy weensy chamfers on
two of the corners. It might be possible to design the case to take
advantage of these.

[Rich Barlow]

Rather ironically, I forgot to reply to this question.

On Mon, 2014-08-04 at 13:41 +0100, Alistair Lynn wrote:

> 1. The motor board had a large number of test points, but there don't
> seem to be any in this schematic?

I forgot to add them.

Rich

[Rich Barlow]

Now with test points.

Rich

[Andy Busse]

Hi Rich,

Thanks for the effort on this. I don't really have comments on the
schematic - it looks like a really robust, cracking design, and I look
forward to seeing them in the wild.

However, may I ask what the expected timetable is between the current
stage of development and when it'll be in the hands of teams?

I suppose a similar question applies to the whole set of kit we expect
to deploy this year [0] - as a relative outsider to the development
process, it does seem like there's a lot still to be done before
Kickstart. Or will some get deployed later? And what can we do to help?

Thanks,
Andy

[0] https://www.studentrobotics.org/trac/wiki/2015/Kit