Replies to This Discussion

3D RoboticsPermalink Reply by Chris Anderson on December 2, 2011 at 9:48pm

1) Yes. 

2) Yes, but don't go over that.

Permalink Reply by spagoziak on December 2, 2011 at 11:45pm

Thank you Chris :)

I've been crawling through search results regarding this and haven't yet encountered an article that mentions power would travel from BEC > RX > APM > SERVOS :)

Now I can get back to drawing out my wiring scheme!

Permalink Reply by John Campen on December 10, 2011 at 12:19am

The problem I am having is that the one BEC/ESC power feed gets that ESC very hot because of the load it has to supply to the APM,GPS,Xbee and RC receiver. Its a lot of load from a small circuit that isn't intended for this.

I was wondering if some how a  regulator can be made up so the four BECs can be used to supply the power and so it can balance the voltage to avoid the one ESC over heating. This I would suspect would be the best way. just I'm not sure how to make this up electronically.

Permalink Reply by John Campen on December 10, 2011 at 1:05am

Update: Ok now, Ive  connected a separate ESC direct to the battery connection at the PDB and it works very well and the ECS/BEC that was the power source before doesn't heat up. . However this separate ESC as a power supply still get quite hot. I see the two regulators on the ESCs PCB still get hot which is expected, but the three lines of chips under the heat sink on the reverse side of the PCB also get hot even though there is not load going to a motor. I assume these chips regulate the electrical pulses to drive the motor.  This heat is no doubt drawing current unnecessarily so is there any way to isolate the 5v regulator and disconnect these motor drive chips so that the ESC is just a 5v regulator?

Permalink Reply by spagoziak on December 10, 2011 at 10:29am

Probably best to buy a dedicated UBEC for that thing.  That's my plan; I'll use a bigger one to drive all the servos and probably power the ardu along with it.  If I need 6v then I'll use a second smaller UBEC to drive juice to the ardy independently.

Not sure I'll use the bec on the esc at all, actually!  

Permalink Reply by John Campen on December 10, 2011 at 3:34pm

i don't think you need to go up an extra volt to make it 6 volts to avoid brown outs. You just need more amps. Look at it this way. a small 12 volt battery wouldn't likely have enough current to start a car where as a large acid lead 12v battery will. Its not the voltage making a difference but the amperage.

DeveloperPermalink Reply by R_Lefebvre on December 10, 2011 at 4:30pm

Or more precisely, the internals resistance of the large battery is much lower than the small battery, which is why it can supply the higher amperage with the same voltage. ;)

As to the original question:

I think I just came up with the ultimate solution to the power question.  First, perform the "Jason power hack", which is to separate the 5V rail between the input pins and output pins.  Then unsolder JP1.  Now the APM, output rail and input rail are ALL seperated.  Then solder a wire from a 5V source on the APM to the 5V input servo rail.  Now the input rail is connected to the APM so your receiver is powered by the APM source, but the servo outputs are seperate.

Now, my own contribution to the solution:  Simply purchase an L7805 5V voltage regulator.  I found it for $5 amongst the small supply of parts still available at Circuit City.  It can supply 1A which is more than enough to power the APM, IMU, sensors and receiver.  I bring that power into the servo input rail and it back-flows into the APM.

Then I use a UBEC set at 6V to supply the servos.  It all works perfect so far in bench testing.

Permalink Reply by spagoziak on December 10, 2011 at 6:47pm

This is pretty interesting!  Can you post pictures or a tutorial?  

Permalink Reply by spagoziak on December 10, 2011 at 6:46pm

I didn't mean to suggest that my move to 6v was for brownouts.. .that's for more powerful servos.  I'm going to use a 10A UBEC and add a 3300uf cap on the line.  That'll do plenty to avoid brownouts :)

I was thinking of a different bec for the APM because someone earlier in this thread suggested it may not be so good an idea.  Chris said it'd be ok, though, so .. we'll see.  I don't actually have any of this stuff yet lol

Permalink Reply by John Campen on December 10, 2011 at 7:38pm

well Ive now  bench tested with a separate ESC and its even lower amps at 25amp where the others are all 30amp and the 25amp ESC now works really well as a 5v regulator and it doesn't get as hot as the one 30amp that was sharing its bec with the APM and motor controller.  This ECS was quite cheap. What Id like to do is isolate the ESCs 5v regulator circuit from the motor control chips so it is using less current and less heat.