Society of Robots - Robot Forum

Electronics => Electronics => Topic started by: hoshiro on July 17, 2012, 12:13:23 AM

Title: Smoking Servos Aren't Much Fun
Post by: hoshiro on July 17, 2012, 12:13:23 AM
I'm going to put my noob factor on full display here. What in my ignorance did I do wrong here?

I'm learning how to operate servos off of an Arduino. I find I can run the servos fine if I power them off of a home-built variable-voltage power supply. However, this means I have to be plugged into a wall. I want to unteather this circuit from the wall, and so want to run off battery power.

I reached for a small 6V 1.3Ah AGM (lead acid) battery that happens to be handy.

(http://www.batteriesplus.com/images/Product/Large/32639.jpg)

I used the multimeter to discover it's I measured 6.3V across the terminals. I assumed that the 0.3V over wouldn't be catastrophic.

I connected the battery to the power pins of a HiTec HS-55.

No joke, I let the magic smoke out of my poor HS-55.

I know software well, digital hardware reasonably well, and mechanical things like servos not at all. What's the obvious thing that I'm missing? (After you stop laughing. I warned I'm putting my noob factor on display.)

Is this an over voltage problem?
Am I dumping too much current through?
Do I need some sort of regualtor?

I'd rather not smoke more servos on trial and error.
Title: Re: Smoking Servos Aren't Much Fun
Post by: Soeren on July 17, 2012, 11:46:16 AM
Hi,

I used the multimeter to discover it's I measured 6.3V across the terminals. I assumed that the 0.3V over wouldn't be catastrophic.

I connected the battery to the power pins of a HiTec HS-55.
[...]
Is this an over voltage problem?
I have no experience with that particular servo, but it sounds like it, given that you probably didn't have anything connected to the signal pin??


Am I dumping too much current through?
If you give a too high voltage the device may draw too much current (you don't dump current into it).


Do I need some sort of regualtor?
A simple silicium diode capable of the current would do (1N400x should be fine).


I'd rather not smoke more servos on trial and error.
You can measure the currrent at eg. 5.6V (which is about what your 6.3V battery with a diode in series will deliver) and see if anything acts up.

You can get some data for the servo here (http://www.servodatabase.com/servo/hitec/hs-55)

If you open the dead servo and post sharp close ups of anything that seems burnt, someone may be able to help you repair it - remember to take lots of pics along the way, so you can put it together correctly once you're done :)
Title: Re: Smoking Servos Aren't Much Fun
Post by: hoshiro on July 20, 2012, 05:19:28 PM
I am grateful for your reply. I've waited until after I could do some more experimentation.

Quote
you probably didn't have anything connected to the signal pin??

Correct. Nothing was connected to the signal pin.

Quote
the device may draw too much current

Yes, forgive my imprecise language. :)

Quote
A simple silicium diode capable of the current would do (1N400x should be fine).

I appreciate this tip.  I have several 1N4002 diodes.

I've gone back and made notes step by step from the known configuration (powering the servo from the variable-voltage power supply).

The end of the long story is that this works fine. :D

I programmed the microcontroller to move the servo 22.5º every half second, and the voltage at the diode's cathode varies between 5.53V and 5.59V — with no smoke. ;)

I do not know the cause of the problem, but I don't want to destroy another servo trying to discover what I did wrong.

Again, many thanks for the ideas.
Title: Re: Smoking Servos Aren't Much Fun
Post by: Soeren on July 20, 2012, 05:32:32 PM
Hi,

Great!

Since this could be of help to others, perhaps you could shoot an email to The Servo Database, about this servo dying screaming if seeing 6.3V?
(Might help keep more of them alive :))

Don't throw the smoked servo. Either fix it, or save it for spare parts (like the gears), for future repairs.