PCBs are ordered

[Crashmatt]

The PCBs are finally ordered. It will be a few weeks before they
arrive and a few weeks to build, test and ship.

The total count is 12 boards for 7 users. Many more than I expected.

This volume saved much cost per board of which I spent a little on a
silk screen print. Hopefully we will be able to see what connects
where.

There are two spare boards. One for first build mistakes (hopefully
none) and the second is a group spare. With this many people
experimenting there will be mistakes,

Thankyou very much for all of your support.

Regards Matt

[Peter Hollands]

Super Matt. Thanks so much.

[tshado]

Hello Matt,

These are good news!!
Thanks for your work

Best regards,
Jean-Claude

On 23 sep, 23:52, Peter Hollands <peter.holla...@gmail.com> wrote:
> Super Matt. Thanks so much.
>

[Crashmatt]

The PCB service I am using posts nice pictures of the PCB at each
stage of production. I will try to post all of them just for the
record.

Here is the first for the little CAN tranceiver board

http://rc-servo-interface.googlegroups.com/web/CAN_IF_V1+PCBPool+layout.pdf?gda=Eq6GB08AAADTCbOyoFv9XDIvMOn2bk5zPM98e99BCQkUEO0gGbeGp-OSoXQlLEZ8IIdkUF-1DPye_5ze5-Q9vaIAAttP-oV0nHMhSp_qzSgvndaTPyHVdA&gsc=mw3unAsAAAB5ucr8IV1lQOVLYzQEiHH9

[UAVflightdirector]

Here are the photos of the boards in production

https://docs.google.com/leaf?id=0B5SF4VCMDng5YjA3NDdjYmQtYTY1ZC00ZjVjLWIzYTktZTBiNjhhZWE3OTY2&hl=en

[Peter Hollands]

That's great to see. Thanks. Pete

[UAVflightdirector]

The PCBs have arrived. This link has a photo.

https://docs.google.com/leaf?id=0B5SF4VCMDng5ZWNiYjc5NTctNjU0YS00YjJlLThkZTYtYjg2ODg2Mzg1NGZm&hl=en

The first trick will be to solder down the ceramic resonator Y1. I am
starting to regret finding such a small component.

Fortunately the plan is to use the larger Y2 resonator. The smaller
Y1 is for 16MHz, the same as UDB2.

If we want to go faster, the resonator must be slower to fit the
minimum PLL frequency. The slower resonator is larger which makes me
happy.

Matt

On Oct 5, 8:56 am, Peter Hollands <peter.holla...@gmail.com> wrote:
> That's great to see. Thanks. Pete
>
> On Tue, Oct 5, 2010 at 6:48 AM, UAVflightdirector <
>

> uavflightdirec...@gmail.com> wrote:
> > Here are the photos of the boards in production
>

> >https://docs.google.com/leaf?id=0B5SF4VCMDng5YjA3NDdjYmQtYTY1ZC00ZjVj...

[tshado]

Hello,

It'a always a nice thing to get real hardware , like these PCb's
Very nice boards !
Thanks for keeping us updated

Best regards,
Jean-Claude

On 11 oct, 17:34, UAVflightdirector <uavflightdirec...@gmail.com>
wrote:

> The PCBs have arrived. This link has a photo.
>

> https://docs.google.com/leaf?id=0B5SF4VCMDng5ZWNiYjc5NTctNjU0YS00YjJl...

[J. L.]

Thanks for the pics cant wait till they are all ready. :)

[Peter Hollands]

Thanks for the update. Pete

[UAVflightdirector]

The first board is built.

For some reason two of the decoupling capacitors lost thier connection
to V+. They get an ugly but functional patch wire.

As predicted, I am having trouble with that resonator. It doesnt want
to run yet.

Debug works with the internal oscillator and the software is running
but at the wrong frequency.

Not too bad for a few hours work.

Matt

On Oct 11, 10:22 pm, Peter Hollands <peter.holla...@gmail.com> wrote:
> Thanks for the update. Pete
>

> On Mon, Oct 11, 2010 at 8:08 PM, J. L. <vwyodap...@gmail.com> wrote:

[UAVflightdirector]

A happy treat before bedtime.

Fitted the 4MHz ceramic resonator with the 4X PLL and it works. The
servo frame is at 25ms where we put it.

Next step is to complete the half built CAN TRX board and get it all
working with UDB.


On Oct 11, 11:23 pm, UAVflightdirector <uavflightdirec...@gmail.com>
wrote:

[UAVflightdirector]

and a picture of the built board to really finish off the day:

https://docs.google.com/leaf?id=0B5SF4VCMDng5OGUxOTQxYzQtNDUwZC00MDFmLWI0MzItN2M1M2Y4Y2UyNTY3&hl=en

Please note the use of large coins and cunning camera angle to make
the product look quite small.

Thats a Swedish Kronor, a euro and a 2 pound coin. Unfortunately my
leftover USD coins are not at home so USA is not represented.


On Oct 11, 11:55 pm, UAVflightdirector <uavflightdirec...@gmail.com>
wrote:

[William Premerlani]

Hi Matt,

Cool. Its really great having you on the team.

Best regards,

Bill

[UAVflightdirector]

Thanks Bill

I dont know if the crystal made it onto the UDB4. With a bit more
testing this might prove that the 4MHz ceramic resonator would be a
smaller working solution.

Best Regards
Matt

On Oct 12, 12:45 am, William Premerlani <wpremerl...@gmail.com> wrote:
> Hi Matt,
>
> Cool. Its really great having you on the team.
>
> Best regards,
> Bill
>
> On Tue, Oct 12, 2010 at 2:12 AM, UAVflightdirector <
>

> uavflightdirec...@gmail.com> wrote:
> > and a picture of the built board to really finish off the day:
>

> >https://docs.google.com/leaf?id=0B5SF4VCMDng5OGUxOTQxYzQtNDUwZC00MDFm...

[UAVflightdirector]

More good news this morning.

The whole system communicates and getting servo output.

No faults on the CAN trx board.

Inputs are not tested yet. For that I need to collect the spare RC
receiver that is waiting in the post office.

Happy days

Matt

On Oct 12, 6:25 am, UAVflightdirector <uavflightdirec...@gmail.com>
wrote:

[Adam Barrow]

Matt,
very cool progress, thank you for sharing. I look forward to flying with the new CAN boards :).

Regards,
Adam Barrow

[Peter Hollands]

Matt, All very exciting. Congratulations on the progress so far. Pete

[William Premerlani]

Matt,

The UDB4 design is still in flux, so the resonator is still eligible. Could you send me a link to the part?

Bill

[UAVflightdirector]

The resonator is a Murata Ceralock 0.5%. The best accuract grade is
required for CANbus

Digikey part number 490-1200-1-ND
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=490-1200-1-ND

Cost is not so bad at ~0.25Euro at 100off quantity. Not quite so
cheap as a crystal but quite a bit smaller.

I have not figured out if I should have added a series resistor to
prevent overloading the resonator.

One danger is the PCB pad layout. My pad design is good enough for
prototypes but not good enough for production.

Matt

On Oct 12, 2:54 pm, William Premerlani <wpremerl...@gmail.com> wrote:
> Matt,
>
> The UDB4 design is still in flux, so the resonator is still eligible. Could
> you send me a link to the part?
>
> Bill
>
> On Tue, Oct 12, 2010 at 8:25 AM, UAVflightdirector <
>

[UAVflightdirector]

Correction, that should be the 0.1% version, not the 0.5%


On Oct 12, 3:27 pm, UAVflightdirector <uavflightdirec...@gmail.com>
wrote:

> The resonator is a Murata Ceralock 0.5%.  The best accuract grade is
> required for CANbus
>

> Digikey part number     490-1200-1-NDhttp://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=490-...

[UAVflightdirector]

What is wrong with servo input 6? Nothing!

Just spent 4 frustrating hours trying to work out what is wrong with
the new hardware
Turned out to be a badly copied line in the code. I had used a
workaround on V1.

The good news is that everything really does work. Its time to order
the components and commit to a weekend of construction.

Need some test code to prove the hardware is good. This will put a
test pattern on all available outputs. Something that will test for
both connections and shorts between pins. Any ideas on this would be
welcome.

Matt

[tshado]

Hello Matt,

Nice to see your bringup and validation tests !
Bravo !!

Jean-Claude

On 12 oct, 22:47, UAVflightdirector <uavflightdirec...@gmail.com>
wrote:

[UAVflightdirector]

Here is the first complete, inspected and fully tested production
board
https://docs.google.com/leaf?id=0B5SF4VCMDng5NjU0NjVkMWEtOTI3ZC00YmM2LWFjNGEtOTlmZWRjZTYwMWFk&hl=en

The test software is complete. I will be shipping with the test
software installed. I will also release the test code, project
and .hex when I get around to it.

The test software outputs servo PWM on every connection.
The PWM is modulated so if you attach a servo it moves.

There are two PWM signals that go in reverse directions. The
directions are assigned to alternating pins. This nicely checks for
shorted pins.

The production qualification involes probing every pin with a servo.
If every signal connection moves the servo, it passes.

DO NOT CONNECT YOUR INTERFACE BOARD TO ANYTHING WITHOUT REPROGRAMMING.

The dsPIC solder joints are being visualy inspected by way of digital
camera.

The production line (office) needs tidying a bit before full
production gets under way
https://docs.google.com/leaf?id=0B5SF4VCMDng5MjYxYWQ4NGUtY2YzOS00Y2ZmLTk3MjQtZWU5ZWUwZjlmZjMw&hl=en

Matt

[UAVflightdirector]

After 14 hours and a packet of HobNobs, all of the boards are finished
and tested 100% working.

Photos of production are here
https://docs.google.com/leaf?id=0B5SF4VCMDng5YjA3NDdjYmQtYTY1ZC00ZjVjLWIzYTktZTBiNjhhZWE3OTY2&hl=en

Photo of all the complete boards here
https://docs.google.com/leaf?id=0B5SF4VCMDng5YjBjNzQyMGEtOTUxZi00NDg1LTk3YmYtYmFmOTA2YjdmNDMy&hl=en

Time to start building the CAN tranceiver boards.
It is almos ttime to send money. First I need to work out the postage
and what the packaging will be.

Hobnobs: http://en.wikipedia.org/wiki/HobNob

Matt

On Oct 16, 10:24 am, UAVflightdirector <uavflightdirec...@gmail.com>
wrote:

> Here is the first complete, inspected and fully tested production

> boardhttps://docs.google.com/leaf?id=0B5SF4VCMDng5NjU0NjVkMWEtOTI3ZC00YmM2...

>
> The test software is complete.  I will be shipping with the test
> software installed.  I will also release the test code, project
> and .hex when I get around to it.
>
> The test software outputs servo PWM on every connection.
> The PWM is modulated so if you attach a servo it moves.
>
> There are two PWM signals that go in reverse directions.  The
> directions are assigned to alternating pins.  This nicely checks for
> shorted pins.
>
> The production qualification involes probing every pin with a servo.
> If every signal connection moves the servo, it passes.
>
> DO NOT CONNECT YOUR INTERFACE BOARD TO ANYTHING WITHOUT REPROGRAMMING.
>
> The dsPIC solder joints are being visualy inspected by way of digital
> camera.
>
> The production line (office) needs tidying a bit before full

> production gets under wayhttps://docs.google.com/leaf?id=0B5SF4VCMDng5MjYxYWQ4NGUtY2YzOS00Y2Zm...