Servos vibrate when I stick a resistor on them?

Forum 2005-2010 (read only) Hardware Interfacing
1

So in my quest to get more out of a 9v battery when trying to power two servos and a ton of LEDs, I bought a new very LDO voltage regulator, and some resistors which could handle up to 1 or 2 watts so I could supply around 300mA or so to each servo, and no more.

The voltage regulator worked great, and I'm getting like 3x the life out of the battery that I was with the LM317 I was using. This new one has a dropout of like 0.35v, though it still seems like it might be more than that, since my circuit goes kaput when the battery reads 7v when unconnected to the circuit.

But the resistors worked terribly. Even with the highest wattage resistor with the lowest resistance... just 15 ohms... when I stick one on each positive power lead to the servos, the servos buzz like crazy. They move just fine... but they buzz even when stationary.

The servos work just fine even when the battery is almost down to 7v, but as soon as I stick those resistors on there, even with a fresh battery, I get the buzz.

Any idea what might be causing that?

I mean if i stick a 15 ohm resistor on a 5v power source, I get 333mA out the other side, don't I? I'm pretty sure servos only need around 25mA when idle, and 100mA when moving but not loaded, as is the case here.

So what gives?

2

What sort of value of capacitors have you got on your LDO? They are quite sensitive to having both too much and too little.

3

I'm using the mic29150:
http://www.micrel.com/_PDF/mic29150.pdf

And I have the reccomended 10uf electrolytic on the output and a 0.1uf ceramic on the input.

4

I do also have a 47uf electrolytic across the + - rails between the servos and the microcontroller, and a .1uf ceramic near the ATMega pins but I'm pretty sure I added the 47uf after I had the vibration problem.

5

what you've done is increased the impedance of the battery supply to the servo. motor tries to move, voltage collapses, servo drive from internal opamp is lost, motor stops trying to drive, voltage recovers, and now you're back to square one.

it might seem a daft question, but why are you sticking resistors in the servo supply line.

your last question illustrates that you are unclear about what a resistor does. when you fit a resistor in a supply line it simply consumes voltage in direct proportion to the current flowing through it to the load. with no load there is no voltage drop. with 15ohms and 100ma load you will drop 1.5volts. so your 5 volt supply now feeds 3.5 volts to the servo - is that enough to keep it working, or does my first paragraph suggest what's actually happening.

6

scswift, no offense but we need to roll you back to square one on electrical theory. I'll give you points for actually trying things while asking ceaseless questions, instead of just asking ceaseless questions...but I thought you were trying to complete a project on short time, rather than tinker around and learn? In that case it makes more sense to ask for solutions before trying something you don't fully understand.

7

it might seem a daft question, but why are you sticking resistors in the servo supply line

Because when a servo starts moving, or is in a stall condition, it draws a ton of current, and I would like to prevent such a condition from occuring.

I read somewhere that back in the 70's the reccomended way to limit current to motors was simply to use a resistor, and servos are a motor connected to a potentiometer, and resistors limit current, so it seemed like it might work. I figured it might waste some power, but if it allowed the 9v to keep going down to 6v, it might be well worth a little waste.

And before anyone suggests a better battery or a second battery, I've been down that road on the forums already. I don't have the space for anything but a single 9v alkaline, and I can't afford to stick custom NiMh battery packs in the props I'm making at this time.

I also don't have space or time to implement any more exotic solutions like maybe a current regulator, or a supercapacitor. But that's okay, because my testing indicates the servo current thing isn't a showstopper. I was just curious what was causing the problem.

scswift, no offense but we need to roll you back to square one on electrical theory.

Heh, I've only been been stuying this stuff for nine months, and I've been teaching myself by asking questions here and reading pages on the web, so I'm sure there's a few things I've missed. I didn't expect when I decided to use the Arduino to make some bargraphs animate that I would end up having to learn so much, or take on a project of this magnitude so soon.

thought you were trying to complete a project on short time, rather than tinker around and learn? In that case it makes more sense to ask for solutions before trying something you don't fully understand.

Yeah, I gotta get my PCB designed and out the door for manufacture by Friday, or I'm screwed. I've got most of the design nailed down, except the layout of the traces, but at the last minute I've decided I need to add a DAC, and I find now I need to know how to update it with interupts after I just learned how hardware SPI works ... Oy.

Testing whether the resistors would lower the servo current though was like a 5 minute job, and if it worked it would be worth it, so...

8

It's still preferable to let the servos draw whatever they want. An actual LM317 based current regulator circuit might do the trick, or speccing smaller servos, or using something other than a 9V battery. A few AAA batteries shouldn't be that much bigger than a 9V and would get you two or three times the capacity.

designed and out the door for manufacture by Friday, or I'm screwed. I've got most of the design nailed down, except the layout of the traces

Good luck with that...it's the part that usually takes the most time.

9

The most time? Over the rest of the design of the circuit? Maybe for someone who knows a lot about designing electronic circuits. But for me, this thing has been like two or three months in the design phase. :slight_smile: I don't think laying out a few traces will take quite that long.

But yeah, it's gonna be tight.

An actual LM317 based current regulator circuit might do the trick, or speccing smaller servos

I'm using sub-micro servos. They don't get any smaller. And I'd try the LM317 but I don't have the time, and I don't really have the space on the board for it either. But maybe once I get the board ready I'll try that and see what happens. I suspect that the reverse will simply occur to what is happening now though, and I'll still get a brownout.

or using something other than a 9V battery. A few AAA batteries shouldn't be that much bigger than a 9V and would get you two or three times the capacity.

:cry:

I looked at every battery case out there. Even considered N and 1/2 AA sized batteries. Can't make it work! (6 AAA rechargables in a prism form factor might just fit in the handle though. But only because there's no thick plastic case around them.)