Society of Robots - Robot Forum

Electronics => Electronics => Topic started by: littlebananas on November 07, 2006, 04:49:27 PM

Title: need help on 30 servos' total power consumption
Post by: littlebananas on November 07, 2006, 04:49:27 PM
Hi everyone,

I'm a new inductee into the world of robotics and would like to get some advice on the following issue.

Basically, I need to be able to control 30 servos simultaneously. This is achieved through the use
of the Basic Stamp 2, and two Parallax Servo Controllers which will be able to control up to 32 servos
at the same time.

However, I have a question regarding the power requirements. My servo is the Hitec 785HB, and at 6V, it
has a rated no-load current draw of .25A. The stall current for the servo is 1.8A.

My question is this: if I am going to have 30 servos, that will ramp my power requirements all the way
up to 54A if I were to calculate it based on the stall current. That's a lot of current and I wonder if
I'm doing this calculation right. Additionally, my 30 servos will not be lifting heavy loads at all - they will
just be turning a lightweight rod attached directly to the output shaft. Does this mean I can get a power supply with a lower rating? (If I assume that each servo will draw just .5A, that will mean getting a power supply of 15A)

I've read about how servos can 'spike' the current draw, and would like some help and advice in my picking
of a reliable power supply. It would be great help if anyone can provide some experience with dealing with these matters.

My current preference for solving this problem is looking at those enclosed AC-DC switching regulated power supplies on www.jameco.com.

Thanks!
Title: Re: need help on 30 servos' total power consumption
Post by: Admin on November 07, 2006, 07:26:14 PM
The best way to do this is just to buy the servo, test it out under the right conditions, and see its current draw. I have a similar problem operating 20 servos that I am working on now. All I can do is test the servos under the desired conditions, then build my power supply with that knowledge.

If you expect low torque, an estimate of .5A average sounds about right to me. I am using a digital HS-5995 servo at medium/high back and forth motion torque , drawing about .35A average, with a peak current of 1.7A.

Current spikes can however be surpressed with a very large electrolytic capacitor between power and ground . . .
Title: Re: need help on 30 servos' total power consumption
Post by: littlebananas on November 07, 2006, 09:41:59 PM
Thanks,

I haven't had my HS-785 servos delivered yet, but I'll try that out when they arrive, and post the results here. I guess that's the best way to get an idea of how they'll behave. Your .35A average is a good sign though. How often does the peak of 1.7A occur, and does the spike happen when you initially turn on the circuit to all servos?
 
Funny how I didn't realise I could test it that way!

Regards
Title: Re: need help on 30 servos' total power consumption
Post by: Admin on November 07, 2006, 09:55:20 PM
Being DC motors with high inductance, the largest current draws occur during the highest acceleration.

For example, suppose your servo is moving clockwise at full speed and is suddenly commanded to move counterclockwise at full speed. This sudden and drastic change in acceleration would yield the largest possible current spike for a servo.