If I am not mistaken the servos you are using are a bit more beefy than the ones typically supplied with an arm. you get more torque but they require more current at peak loads.
If that is the case, you should look into getting a bench supply for it.
Something like this:
tti-test.com/products-tti/psu/exr-series.htm
I have no idea what they cost, but it will probably cost more than your arm parts did.
That one is $309.
Here’s one for $165:
allelectronics.com/cgi-bin/i … PPLY_.html
- Jon
If I needed a foolproof system I’d use batteries AND power packs.
That way you can draw more power if needed from the batteries while not flattening them unnecessarily.
Also if the power goes out it will keep working.
Oh ok, I did not know those servos required that much current to operate. I was thinking they used far less current than that. I’m not familiar with the specs of that servo.
I use an old AT-style power supply for my “eggsperiments” which is rated at 5v / 15A and I can easily get it to struggle when all 18 servo’s on my bot are active…
Never underestimate the power of a servo! 8)
But you don’t need a variable supply, it only needs a fixed voltage. 
Sure, but I need a variable voltage bench power supply, so I like that one better 
- Jon
Hello again,
Something weird happened …
After working with my system for an hour or two, everything just stopped working as if I turned off the power …
Whenever I try turning everything back on, nothing happens … no response
from the motors …
Of course they’re not all defect, as I tried a spare one and it didn’t work too …
I guess either the chip is burnt, or its communication part is defective … What can I do to solve the problem ?
Except buying a new chip, is there anything I could do to fix that ?
Thank you for your concern !
fuse?