SOR Open Source Project: Servo Module

[Asellith]

Ok need some input working on the servo module for the project.

Check it out here:
http://www.societyofrobots.com/robotforum/index.php?topic=2765.0

or here:
http://www.societyofrobots.com/member_tutorials/node/69

Ok back on topic.

I want some input before I proceed. I have started design but I need some input from everyone.

Question:
1. How many servos do you want to control? I am aiming for around 20 but I need to make sure I have to board space for this without using SMD. (Plans for a future SMD pre assembled version might come later on the smaller module templates)

2. Does anyone think they will need a way to read the current position of a servo? The only situation I can think of is if more then the main controller will be adjusting the servos or two master controllers want to move servos and check to see if the other master has moved it before it proceeds.

Also any ideas or thoughts you might want to see in this module let me know.

[krich]

I have a comment regarding SMD.  Over the last several months, I have grown to really like SMD parts for homemade boards.

I've found that they are not as hard to solder as you might think as long as you stick to the "bigger" SMD packages (SOIC chips and 0805+ resistors/caps).  There's some really good tutorials out there.  SMD components will obviously save you a lot of space on the PCB.  For 20 servos, you might need it.  Heck, the fact that you don't have to drill all those dang holes is incentive enough to give it a try.

The traces can get a bit small if you're etching your own boards, especially if you're trying to run them between legs of an SOIC chip, but it's far from impossible with good etching technique.

I'm just saying don't completely count out SMD for your project...SMD components are well within the realm of home-brew electronics.  Even for a newbie like me.   

[Asellith]

You do realize my soldering skills are horrible I have done some stuff with SMD and its not fun. My plans are to prototype then do a small board run to test the PCB design. If that all checks out then I will release the design in a tutorial here. Then I may try to get either a group buy or depending on how much out of pocket I have to do I might produce a small run and sell them as kits or assembled version to those who do not want to bother with building their own. Then if the project as a whole takes off more I may complete a smaller SMD version. Because the module standard calls for several different sizes of modules I can have two different sizes.

Converting to SMD doesn't save me that much space. Other then a few stray resistors and maybe a crystal. The only components I have right now are 2 16 pin dip ICs and an ATmega16 MC. The space problem is in the fact I need a 4 or 5 pin connector for I2C and power. Oh and of course 60 pins for servo connectors. Those don't come in SMD packages Oh and a dip switch so you can address the servo module to let you use more then 1 in a design.

Several thoughts I have also had are the fact that it might be possible to do a background timer function to allow a command to increment a servo over a time frame. So the main controller would send a message like move servo 12 to position 180 in 12 seconds. So I might plan on adding that function to the controller.