0
\$\begingroup\$

I have 4 Tower Pro MicroServo 9g SG90 I want to power with a battery. This is the datasheet I found online: http://abc-rc.pl/templates/images/files/995/1428085018-sg-90-tower.pdf

I have a small breadboard and some resistors laying around (220, 1k, 10k), and a 9v battery.

I hooked up the battery leads on the power rail of the breadboard, and added my 4 servos in there, and they turned on and rotated properly when I issued commands from my Arduino board. The thing is that AFAIK it operates on 4.8v-6v from what I have read.

I did a voltage divider using some of the resistors I had and I could output 5v when tested with my multimeter, but the servos didn't work (probably because low current).

So my question is, can I power my 4 servos with a 9v battery? Should I add some resistors before? Bear in mind I am new to electronics.

\$\endgroup\$
1
\$\begingroup\$

Bare in mind? I'm supposed to imagine you naked? No thanks!

In any case, there are two problems with the data sheet: the voltage is specified as 4.8-6.0, which is clearly not compatible with a 9-V battery. The fact that it worked when you tried it simply means that you "got lucky" in terms of operating outside the specs.

The second problem is that the current required is not specified at all. Given the torque specification (which is surprisingly high for such a small, cheap unit), the current required will probably be much higher than what a common 9-V battery can deliver. Depending on how much torque you actually need in your application, you're going to need a much better battery than that, along with some sort of regulator.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Thanks for the answer. I don't want to fry my servos, that is why I am asking. I am afraid if I power them with a 9v they might overheat and break. Is that possible? Or is it current that could break them? \$\endgroup\$ – Pacha Jan 12 '17 at 15:25
  • 1
    \$\begingroup\$ The 9-V battery probably can't deliver enough current to fry the servo (part of "got lucky"), but a high-current supply at that voltage probably would fry them under heavy load. \$\endgroup\$ – Dave Tweed Jan 12 '17 at 15:34
  • 1
    \$\begingroup\$ SG-90 draws 0.75A peak. 4 servos would need 3A. Recommended battery is a 4 cell (4.8V) AA Nicad or NiMH pack. AA Alkalines might be OK if the servos are not heavily loaded. \$\endgroup\$ – Bruce Abbott Jan 12 '17 at 17:05
1
\$\begingroup\$

Voltage divider circuits using resistors are good when you are trying to sense the voltage at a point. That's the reason you are getting a 5V output. In reality, when you connect any load across the lower resistance, the equivalent resistance of the lower branch is the parallel sum of the resistor and the motors internal resistance. At this point, the output voltage to the motor will not be necessarily 5V. Try measuring voltage when your servos are actually drawing current while in motion.

| improve this answer | |
\$\endgroup\$
-2
\$\begingroup\$

It is an old post, so you might have found a solution, but I will still contribute nonetheless.

Have you tried to use a unit gain buffer beween the voltage divider and the servos?

In SHORT: Voltge divider - > Unit Gain - > Servos. - Unit gain takes power from an external source to provide the necessary power and current to the servos. - Voltage Divider set the voltage to whatever you designer it for and will remain almost unaffected by the servos tanks to the Unit Gain Amplifier. The unit gain takes a sample of the voltage divider an "forces" the same voltage to the servos.

In LONG: Unit Gain Amplifier: You would just need an opamp and connect the negative pin and the output pin with a simple jumper ( No Resistance or other ) to act as buffer. This way, the motor resistance would not affect much the Equivalent Resistance. The negative pin of the unit gain would also be connected to the voltage divider for the control of the voltage, output pin to the servo or servos ( Voltage at Vin would be the same as the Voltage at Vout thanks to the unit gain ). Then what feeds the opamp ( +V and -V ) could be an additional power supply with higher voltage, like it could go up to 12-24v depending on the opamp specification, hence it could provide whatever current the motors would need. ( Bare in mind that the 24v would not go directly go the motors given it is controlled by the 5V input pin via the voltage divider ) This way you would not suffer from the Equivalent resistance off the voltage divider affected by your servos in your design, and could provide extra current to the servos from another external power supply which could go up to 24 Volts. You would need to get an Opamp designed for 24V of power and whhich does not release too much heat like in a linear region at 5V of operation.

| improve this answer | |
\$\endgroup\$
  • 2
    \$\begingroup\$ The output current of typical op-amps is in the low tens of milliamperes (10 milliamperes, maybe 20.) That's not enough to power a servo. The battery itself (a typical 9 volt "transistor" battery) cannot deliver enough current to drive a servo for very long. The very reason the 9V battery didn't destroy the servos is that the battery can't deliver enough current to do any damage. \$\endgroup\$ – JRE Jun 7 at 11:33
  • \$\begingroup\$ Alright, thank you for the correction and sorry for sharing misleading information. So, my way would protect it indeed, but would also just not make it spin due to a low current ^^. But, how could he make it work with a 9V Battery then ? I have powered three of the same motors with a 2 cell LiPo for a delta robot 4 weeks ago and it worked just fine. When I connected just one motor to the battery, it did not get burnt. I did not use any electrical protection or other then, though I think I should have. Is it because I used a specific kind of battery which does not deliver enough current? \$\endgroup\$ – Cedric Jun 16 at 18:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.