APM Oil Pan!

Is 2:22AM (when i started writing this) and i can't sleep, so i decided to write this post.

I been working "hard" and the fastest i can to finish this thing, theres only one step remaining in order to start the "mass" production of this board.

This is how the last step will be done:

-A unique assembled board will be shipped from California to Spain.

-Then Jose Julio in Spain will open his "Surprise Board" and make us the great favor to all of us of checking the board integrity and the harmony of all the components on it (thanks in advance JJ).

-That includes checking some new components i can't mention right now.

-If Jose Julio says 'Si", then i will start the mass production.

By now SparkFun is straggling trying to supply all the components and test bed for APM. It will be released soon.

The board is kinda of big right now. This board will be called the "huge" version. When the code becomes very mature i will create a an smaller version of the APM and APM Oil Pan without the multiplexor on the APM and without the bulky FTDI chip on the APM Oil Pan, so i can save 30% of the size. This one will be called ArduPilotMega "In my way" version. ;-)

Oil pan stands for the old shield name. The difference is that the shield goes into the bottom of the ArduPilotMega board.

~~I decided to let the community guess the rest of the specs for ArduPilotMega Oil Pan, because write stuff in English make me drowsy. The best guesser will receive a $50 certificate. ;-)~~

Come 2:56AM!!

@ Mark; The datasheet doesn't mention an internal diode, nor does it give an equivalent circuit diagram :( However, it does mention the release times with and without a diode in parallel, so I assume it doesn't have one. Since this relay is powered from the high side (the microcontroller sources the current), it would generate a negative voltage on the controling pin of the micro... I think...

@ Jordi; When the current source to the coil is stopped the magnetic field of the coil starts collapsing to keep the current flowing, if this current can't go anywhere, it will create a large voltage potential. The coil of a relay is usually connected to a power rail and the other end is grounded (say, through a BJT) to energize the coil. When the grounding is disconnected (while the other end of the coil is still connected to power), the current keeps flowing due to this collapsing magnetic field. Since the effective resistance is infinity (or very high), the voltage will start rising at the disconnected end of the coil since the current can't go anywhere. Usually a diode is placed in reverse bias from the power end of the coil to the switched end to let this current recirculate in the inductor and prevent voltage spikes.

For this setup, the diode would would go from pin 1 (the powered side) reverse biased to ground.

I hope I'm making sense.

You need to be a member of diydrones to add comments!

Join diydrones

Comments

Developer

Mark Colwell March 3, 2010 at 12:09pm

Thanks Freddy for good description of inducing voltage spike to line vcc. Such a small current will create a small short spike. A larger size diode to span relay from pin 1 across traces to GND might work.
Admin

Morli March 3, 2010 at 11:48am

Jordi, the dip switches are 90deg types too( accessible from side)? Any chance of the sneak peak photo of the prototype sent to Jose for testing ? ;) Congrats Freddy
Freddy Gauss March 3, 2010 at 11:37am

@ Mark; The datasheet doesn't mention an internal diode, nor does it give an equivalent circuit diagram :( However, it does mention the release times with and without a diode in parallel, so I assume it doesn't have one. Since this relay is powered from the high side (the microcontroller sources the current), it would generate a negative voltage on the controling pin of the micro... I think...

@ Jordi; When the current source to the coil is stopped the magnetic field of the coil starts collapsing to keep the current flowing, if this current can't go anywhere, it will create a large voltage potential. The coil of a relay is usually connected to a power rail and the other end is grounded (say, through a BJT) to energize the coil. When the grounding is disconnected (while the other end of the coil is still connected to power), the current keeps flowing due to this collapsing magnetic field. Since the effective resistance is infinity (or very high), the voltage will start rising at the disconnected end of the coil since the current can't go anywhere. Usually a diode is placed in reverse bias from the power end of the coil to the switched end to let this current recirculate in the inductor and prevent voltage spikes.

For this setup, the diode would would go from pin 1 (the powered side) reverse biased to ground.

I hope I'm making sense.
Developer

Jordi Muñoz March 3, 2010 at 9:56am

Ok everybody Freddy won! ;-)

@Freddy congrats! Can you please PM me your email please?
Developer

Jordi Muñoz March 3, 2010 at 9:56am

@Chris, the oil pan is not "official" yet. But i will update the post and include the current parts and details, if any change occurs i will update.
3D Robotics

Chris Anderson March 3, 2010 at 9:44am

Jordi, now that Freddy has won, how about updating the post with the official list of all the components?
Developer

Jordi Muñoz March 3, 2010 at 9:33am

Hello Mark,

I heard about that, but where the spikes occur, inside the microcontroller?

Thanks
Developer

Mark Colwell March 3, 2010 at 9:29am

Jordi,
Sweet design, I really like logger option. You have taken a good methodical approach on this board hope there are no revs, but I doubt it.
Does relay have internal diode? if not I would add one across coil It will prevent spikes on +5v when relay releases. Can we add gimbal control ptz option next?
Developer

Mark Colwell March 3, 2010 at 9:19am

Freddy got the most first, now reverse engineer a Walkera UFO.
Developer

Sandro Benigno March 3, 2010 at 8:54am

My vote goes to Freddy Gauss.

(I believe that his only fault was about the AT45DB161B instead AT45DB161D).
The chip used on the Sparkfun's rocket is the same used on Ulog (AT45DB161D).

of 7

This reply was deleted.