Big problem with servo

Hi guys ,

I am trying to build a gadget with turns off knobs such as a gas knob etc automatically after a certain time. The gadget is programmed / controlled over bluetooth.

I use a hc05 bluetooth, a arduino nano, 3500mah mobile battery , one setup up converter to get battery voltage to arduino regulator compatibility. This is my hobby project for the month. Have spent quite a bit and now I am stuck. I am a programmer and not a electric engineer. The servo used is mg90s metal gear servo.

The problem is when on bread board everything worked ok ok. Then I made connections to battery and tested and servo rotation was flaky. Like a hit miss.

When I tested it with actual known same thing happened and after a while it stopped working. Arduino would turn on sometimes and not all of the time. When I removed servo it starts properly. I connected the servo to my other arduino which is on breadboard taking power from usb and it worked. But again when I connected it back nothing happened.

I used a 5v 1 amp dc power supply for the servo separately and still it won't move just makes little sound and heats. But again works on usb connected to PC. The known has resistance so I am sure it needs more power but then it is not taking power??? I can't even get it to work normally let alone rotating a knob.

Now I am stuck with all the investment into it and all it does is fancy stuff without the real thing. Please help me please.

Schematic?

rajdeeprath:
The problem is when on bread board everything worked ok ok.

Then I made connections to battery and tested and servo rotation was flaky. Like a hit miss.

When I tested it with actual known same thing happened and after a while it stopped working.

You seem to be describing 3 separate situations here and I don't understand the difference between the 2nd and 3rd.

And, as @DrAzzy has said, you need to post a schematic (and maybe 2 of them). A photo of a pencil drawing will be fine. Please don't use Fritzing.

...R

Code?

Attaching images if these will help . I have to get you code from my other PC and draw some schematics for you guys

Basically what I meant for 1,2 and 3 is:

  1. First I tested servo on breadboard with usb power from PC. Everything worked including the bluetooth operations

  2. Next I moved the project into a nano box where I connected stuff to 2 htc sensation batteries (in parallel) 3.7v 3500mah total. Here it was flaky in movement. When I sent command from android it received each time but did not move a few times. I figured that may be nano pins aren't providing sufficient power may be.

  3. When I connected the servo to the gas knob via a aluminium attachment I made, it only went half way and I had the same issue of flaky movement. After a few trial runs I saw that the system won't even start till I disconnect the servo.

image hosting 30 mb

image hosting 30 mb

free upload pictures

And here is a video demonstration

MG90 are only tiny servos.
Rated stall at 1.8kg/cm at 4.8v...you are trying to run it at 3.7v which is way under spec requirements.

Also, by the look of the part on top of the servo, it is going to require quite a bit of turning force.

Have you worked out how much load this is?

Looks like you are trying to turn the servo beyond its travel limits in the video.

When run from the 5v supply, did you use a common negative to the rest of your circuit?

Specification:

Weight: 13.4g
Dimension: 22.8×12.2×28.5mm
Stall torque: 1.8kg/cm (4.8V); 2.2kg/cm (6.6V)
Operating speed: 0.10sec/60degree (4.8V); 0.08sec/60degree (6.0V)
Operating voltage: 4.8V~ 6.6V

I am very grateful to you for your response. I am using a. Step up converter to get 3.7 to 8v so that it feeds into arduino nano vin. My idea was the battery has sufficient amps to provide so it should be ok.

Question: how do I know how much force is needed. Btw this servo just about turns the knob with little unease when Everything is powered off usb.

Yes I common ground everything. The ugly looking brown perf board next to nano is a custom breadboard with positive and negative in parallel rows. And all power is taken from the positive and all god connects to the gnd there.

Sometimes the servo reaches end and sounds like it's struggling I dont know why. If you say the servo is tiny , will his do the job ?

http://www.ebay.in/itm/TowerPro-MG995-Metal-Gear-Servo-Motor-/231685610685?hash=item35f189dcbd

Please help me out

Hi,
Have you measured your supply voltages while the servo is turning and in fault condition?

Thanks... Tom... :slight_smile:

Yes and here is the strange thing. So 3.7 v is pumped to 8v by step up and provided to arduino. The voltage at step up in is 4.0 and output is 8v, but then I connect the servo it drops to 0.98v !!! Don't know why!! Same setup on usb power runs.

Hi,

What voltage drops to 0.98V

Can you please post a copy of your circuit, in CAD or a picture of a hand drawn circuit in jpg, png?

Thanks.. Tom... :slight_smile:

I will update you in the morning. Also please suggest me a alternative servo for my use case. Mg995 or 996 ?

For the servo, 3.7v is too little and with 8v you'll smoke it for sure.

On the other hand the servo seems to be "struggling" because it appears you are trying to drive it past its physical limits.

It is not the servo it is your power supply that is not up to the job.

Also, gas knobs need to be pushed in to turn them. It's such a natural action for a human hand you don't realize you are doing it, until the servo fails to turn it without pushing.

At least for testing try powering the servos with a pack of 4 NiMh cells (fully charged).

Start with a simple reliable power supply before trying more complex solutions.

I would be surprised if the gas knobs need much torque to move them. They are very easy to move on my cooker once I have them past the safety detent.

...R

MorganS:
Also, gas knobs need to be pushed in to turn them. It's such a natural action for a human hand you don't realize you are doing it, until the servo fails to turn it without pushing.

Yes know that one :smiling_face:. I connect the servo to gas knob when it is in simmer mode so I from that pod it does not need a push. It needs a push tpcoto from off to on position

HI,
Have you got ALL your gnds connected together securely?

I am surprised you have not got a circuit diagram drawn, especially after prototyping then building your final project.
How do you know they are wired the same?

Tom... :slight_smile:

Thank you all for helping me. Here is a diagram i worked on today. Please don't mind the directional arrows, just see the color.

image

Now i also have a few questions.

  1. The servo i am using is MG90S - Working torque: 2KG/cm. should i move to TowerPro MG995 / MG996 Metal Gear Servo Motor (9 - 11 kg / CM torque) ? my current servo against the force almost manages it not easily at all.

  2. Why doesn't my battery support when i connect the servo to the perf board +5 and gnd. light on nano comes and goes immediately. But if i switch off the battery and connect USB from cpu it works.

  3. how can i drive a servo with external power ? battery / adapter nothing seems to work. Should my external power's GND also connect to GND on perf board ?

  4. Should i connect batteries in serial rather than parallel then use step down

My only shortfalls are the servo power and power supply for the servo. i don't want to use adapter though since its suppose to be battrery powered.

Robin2:
At least for testing try powering the servos with a pack of 4 NiMh cells (fully charged).

Start with a simple reliable power supply before trying more complex solutions.

I would be surprised if the gas knobs need much torque to move them. They are very easy to move on my cooker once I have them past the safety detent.

...R

Thanks i will be arranging that soon. but also tell me why i can't use LION cells ? cause i have a charging circuit for LIOn batteries which is already connected and charges the batteries using my standard android charger