12 Servo's together does not seem to work, Power problem

Printable View

- 14th November 2006, 03:09
macx75

12 Servo's together does not seem to work, Power problem

Hello All
Am currently making a robot with 4 legs and 3 servo's each leg,
am using the 16f877 to control it.
At the start I used the same power supply which I used with the Microcontroller but the servo started shifting positions if there was load on the legs,
Now I have a diffrent power supply for the servos, and ground is common, the shifting is stopped.
the problem is
1) with the same power source 5volts 500Ma if 2 servos are joint they move freely (use a push button to make it go clock wise and anti clockwise)

but the servo does not have enough power to get up.( have to lift it up manually and then it get is position)

2) if connected to 3or 4 servo then the other 2 does not have any power to even move.

question is that
1) do i need many power line to fulfill the power for 12 servo's ?
2) or do i need a transformer with more amps?

I have seen many hexapod designs with just one power supply and it controls 18 motors. and also the weight, how do they do that?
mac
- 14th November 2006, 06:09
Darrel Taylor

Hi mac,

Every servo is different, but if we take a "Standard Servo", under full load, it can draw as much a 700ma @5V (stall current). So if there's a lot of weight that they are trying to lift, 2 servos = 1.4amps, 2.8amps for 4, and a whopping 8.4amps for 12 servo's.

Hungry little boogers.

Even under NO-LOAD conditions, you'd still need 1.9amps for all 12. (160ma each)

Time to break out the Heat Sinks :)

<br>
- 14th November 2006, 06:18
sayzer

Even if you use a stronger power supply, you may get the same problem along with some other problems.

Even though all V+ and GNDs will be parallel for all servos, the last ones in the power line will still be getting low voltage.

I suggest you make a connection as below.

In Figure1, S1 does not get the same voltage as S4 or S5.
In Figure2, all servos get the same voltage.

-----------------------------

<img src="http://img6.picsplace.to/img6/23/ads__305_z.GIF" alt="Image Hosting by PicsPlace.to" >

--------------------------
- 14th November 2006, 23:19
BigWumpus

Good plan !
- 13th December 2006, 02:09
macx75

12 servos at a time

Thanks
I have been using 1 amp input and thanks Sayer for the design now they get equal power.
I have tried it with 6 servos and they work perfectly will join 6+6 more.
maybe to control all I will need diffrent power supplies.

New Ideas come only by Experiments.
Mac
- 13th December 2006, 04:06
emavil

Servo Motor Schematic clarification

Quote:

Originally Posted by sayzer

Even if you use a stronger power supply, you may get the same problem along with some other problems.

Even though all V+ and GNDs will be parallel for all servos, the last ones in the power line will still be getting low voltage.

I suggest you make a connection as below.

In Figure1, S1 does not get the same voltage as S4 or S5.
In Figure2, all servos get the same voltage.
--------------------------

I am curious about the schematic diagram in figure 2, can u give possible PCB layout for those servos. including power supply, location & connection and a simple microcontroller like PIC16F84A? Do i have to design the pcb following the arrangement of servo connectors found in Figure 2?

I stopped the servo motor programming 'coz i also encountered the same problem. I have only 2 servos, one for steering and the other for acceleration. Both of them mounted perpendicular to each other. As I turn on the circuit, instead of performing the desired action which is move forward, the steering servo tends to turn left, sometimes, right. I still have to do some series of power on, then off, then on, then off until it behaves according to what is expected.
- 13th December 2006, 08:10
sayzer

emavil,

Why don't you try this power supply connection by hand first?

Instead of having it on PCB, try and see if this is indeed the case for you.

Separate the cables of Servos from each other, and apply the voltage as in the schematic above.

Even then if you still get the same issue, then you will know Power Supply was not the case.

Also, are you using battery or what?
---------------------------
- 13th December 2006, 08:32
emavil

Quote:

Originally Posted by sayzer

emavil,

Why don't you try this power supply connection by hand first?

Instead of having it on PCB, try and see if this is indeed the case for you.

Separate the cables of Servos from each other, and apply the voltage as in the schematic above.

Even then if you still get the same issue, then you will know Power Supply was not the case.

Also, are you using battery or what?
---------------------------

Yep, I'm using 4 pcs 1.2vDC 2000mAH rechargeable batteries.
- 13th December 2006, 09:18
malc-c

The alternative is to use digital servos rathet than "standard" servos. These have the advantage of providing feedback so that the controlling software can find exact position of the servos are in, rather than having to "reset" them everytime the thing is powered up. The disadvantage is the cost...
- 13th December 2006, 16:50
sayzer

1 Attachment(s)

Ok, here is something I have just designed.

A concept PCB design for servo power supply should be in similar manner.

<img src="http://www.picbasic.co.uk/forum/attachment.php?attachmentid=1261&d=1166038578" >

More servos can be added keeping the same concept here.

Any question about the concept?

Edit: Servo pins corrected upon Alain's warning!
----------------------------------
- 13th December 2006, 17:55
Bruce

Quote:

Any question about the concept?

What's the difference electrically?
- 13th December 2006, 18:22
mister_e

Well to me, if those connections are close, or simple header it doesn't make any difference as the PCB trace impedance will be pretty low. OK maybe a little more stress at the PCB trace source end but... Few mOHM @ few amp, don't make any difference to me. and we talk about few amps DC and/or low frequency stuff so...

If you had 1 single pair of cable from the PCB to multiple external servo, and each servo are hundred meters (or more) separated... maybe... but who wants to do it like this?

The theory is fair, but i don't believe it's as this critical here.

Sorry Sayzer :(
- 13th December 2006, 19:00
sayzer

1 Attachment(s)

I knew this was going to come out.
I was so lazy to draw all 12 servos. Here it is.

Now, it should make more sense.

In Figure3, S1 and S12 do not get the same voltage although they are parallel.
But in Figure4, all servos get equal voltage.

<img src="http://www.picbasic.co.uk/forum/attachment.php?attachmentid=1260&d=1166037872" >

Edit: Schematic corrected upon Alain's warning.
-----------------------------------------------
- 13th December 2006, 19:13
Acetronics2

Cold shower

Quote:

Originally Posted by sayzer

I knew this was going to come out.
I was so lazy to draw all 12 servos. Here it is.

Now, it should make more sense.

In Figure3, S1 and S12 do not get the same voltage although they are parallel.
But in Figure4, all servos get equal voltage.

<img src="http://www.picbasic.co.uk/forum/attachment.php?attachmentid=1259&d=1166036361 " >

------------------------------

Hi,Sayzer

Sorry for the cold shower ...

But the signal pin is ALWAYS at pin 1 or 3 of the connector...

The two power lines are ALWAYS side by side !!!

Alain
- 13th December 2006, 19:30
sayzer

Thanks Alain.

I corrected the pins in post #10 and #13.

------------------------