Let's Make Robots!

Still alive, power problems

I'm not dead, I just had a little foul up with my Baby Orangutan and never got around to ordering a new one.

The Baby Orangutan is a very tiny robotics controller, complete with an AVR chip, motor controller and the Pololu library for robotics.  It's also really tiny, so if you don't need all the stuff on the larger Orangutan controllers, it's really ideal.  Just solder on the headers, get some female jumper wires and plug stuff in.  The problem I had was with the labels.  The board's labels are on the bottom of the board.  I thought "OK, I'll just solder the headers onto the board upside down!"  And that worked just fine, except the 2x3 header for the ICSP programmer.  I didn't consider the fact that that would be mirrored, and my programmer wouldn't work.  And I don't have the equipment to desolder those headers, at least without damaging the tiny board, so that one is just wasted (but I'm sure I can recycle it later).

So I ordered a new one and soldered the pins on right this time.  Or, by "right" I mean "not how the picture shows, but how I needed."  It's intended to plug into a breadboard, but I solder the long parts of the headers facing up, so I can plug female jumper wires into it.  It works just fine that way.  But now I run into a problem.  The first thing I try to control is the servo motor.  However, as soon as this servo motor moves, the microcontroller resets.  I'm assuming this is a power problem, and I'll have to make a small "power supply" board.  The MCU board already has its own regulator, but I'm assuming the power drop is too much for it to handle.  I'll need an array of smoothing caps.  But this helps me solve another problem I've been having:  how to provide a way to plug all the power connectors in.  I rigged up some horrible looking thing with wires, but having a board with headers to plug into will just be better.

The alternative is to have a second power supply, but from what I can see, that's not possible with the Baby Orangutan.  It takes a single power supply and doesn't expose the pins for an alternate motor supply from the H-bridge.  So even if I did power the servo motor from another supply, I'd have the same problem as soon as I tried to power my motors from my H-bridge.

But, at any rate, I'm back to working on this.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Board resets are not uncommon when a servo tries to pull too much current too quickly.

Things to check:

  • Are your batteries capable of providing enough current. (e.g., a little 9V battery might not do the trick)
  • Are the wires from your battery pack thick enough? Thin wires will limit your current.
  • Does the Baby Orangutan board allow you to separate battery supply from regulated supply?
  • Caps to filter the power, as you mentioned. Maybe a 10 or 100uF in parallel with a 0.1uF on the servo power line.

Are your batteries capable of providing enough current. (e.g., a little 9V battery might not do the trick)

I'm using 4 AA batteries, I think it should be OK.  I'll have to see if they're fresh enough though.

Are the wires from your battery pack thick enough? Thin wires will limit your current.

Maybe?  They're thin wires, if I keep having problems I'll have to look into replacing with bigger wires.

Does the Baby Orangutan board allow you to separate battery supply from regulated supply?

Sadly, no.  I was hoping to use two battery packs to avoid problems like this, but only after I got the baby orangutan did I discover that the motor supply to the H-bridge is the same as the supply for the MCU itself.  If I keep having problems like this, I might have to abandon the baby because of that.  Seems like a pretty big design flaw to me, but we'll see how many problems that really causes.

Caps to filter the power, as you mentioned. Maybe a 10 or 100uF in parallel with a 0.1uF on the servo power line.

That's what I had planned.  But what do you mean "0.1uF on the servo power line?"  Also, do I need to protect or couple the pulse line on the server in any way?  Or is that safe to go straight to the MCU?

Batteries: 6V (assuming you are using alkalines) should be OK, but if they are getting weak, that would explain your problem. The on board voltage regulator will drop some voltage before it supplies the MCU and motor controller. How much voltage depends on the regulator. So if your batteries are down to 1.3V each x 4 = 5.2V The Baby Orangutan's voltage regulator needs at least 5V. If the batteries are low (they can temporarily dip lower during current surges) the regulator will drop out, causing a reset. Consider adding another battery, giving you a 7.5V supply with more headroom.

Regarding the caps, OddBot explains it well in this comment.