Go Down

Topic: From breadboard to perfboard (Read 4692 times) previous topic - next topic


Jun 24, 2012, 07:13 am Last Edit: Jun 24, 2012, 07:18 am by CrossRoads Reason: 1
Eagle lets you do some pretty complicated stuff. Just keep throwing stuff on the board until it breaks!
This all fit in a 10cm x 10cm board with components on 1 side.

It all comes down to careful parts placement to cut down on vias & traces that have to cross each other.
And good use of the ground planes.
Designing & building electrical circuits for over 25 years.  Screw Shield for Mega/Due/Uno,  Bobuino with ATMega1284P, & other '328P & '1284P creations & offerings at  my website.


Within limits the connectivity determines the layout. A working design will go to a pcb virtually without change with the exception of the power supply Frequently the overlooked Item Power traces should be as wide as routing the board will accept. By pass capacitors should be used. a .1uF cap/ic works well and if an onboard regulator distribute 100 300 uf in smaller capacitors. The last suggestion would be to do a "Copper Pour" and tie it's net to Ground and re-pour the net to include the ground points present in the netlist for the board. Finally whether hand or machine routed the same rules apply.
   The 'rules' I mentioned are only common sense and it is also common sense to route inputs to devices away from outputs and to route inputs away from power supply wiring. In the work I did which which was mostly mixed signal analog and digital circuitry split power supplies were necessary including switchers and those were placed at one end as far away from the inputs as physically possible and rc filtering was frequent done with 2 - 22 ohm resistors and 1 - 10uF bypassing before and after the filter resistor. This is done to "split " the Vcc supply line up so as to present a good low impedance Vcc source to all IC's. If your design only has 1 IC then follow the scheme used on the UNO board.
Experience is the finest teacher. When I began my designs lacked those things, were simple and worked mostly.. I was fortunate to begin with small boards with 2 - 3 - 4 IC's as the things I mentioned above came to me slowly. Frequently a circuit is designed from the function and data sheet pinout of the device and works great and little attention is ever spent in reading the design notes. This is where one learns the virtues of the simple things I mentioned above.
  This is where the real engineering begins. Where one learns that design notes are an engineers best friend. Where cation and conservative thought will make the difference in how and with what a design will work with.
  Batteries are a designers worst enemy. As they become depleted their internal resistance increases and this is one example of the mention of attention to power management is so important to a working design.
   And Finally a successful design will work under all conditions. Develop your own power supplies and only share ground and I/O with the Processor. You will loose a major part of any noise or analog voltage dependent measurement in accuracies. If an analog or any other circuit requires any supply voltage you supply it don't depend on loading the processor with your stuff/ Finally unless otherwise possible If you have an input going off board, terminate it, pull it up or dowm and let the next board do the pull ups and downs Never send a floating input or output off board, one cannot assume that the "other Guy" did his home work too.
   A good design works all of the time under any circumstances and with any equipment connected to it. It will be stable in a known state and be ready to connect and use or just use when connected, unconditionally. That is when I signed my name to a design... Only.

--> WA7EMS <--
"The solution of every problem is another problem." -Johann Wolfgang von Goethe
I do answer technical questions PM'd to me with whatever is in my clipboard

Go Up