AL5D Elbow Servo

[sylvain]

Just finished building my new AL5D arm (with the wrist roll joint). This is my first robot ever, first experience with this technology.

When attempting to control the arm with the SSC-32U / Flowbotics Software, I observed some erratic arm behavior and after doing some troubleshooting, I’ve isolated the potential issue with the presence of the Elbow HS-755 HB Servo.

When I connected all the 6 servos together and used the SSC32 Servo Sequencer, the robot response was erratic, sometimes non-responsive, sometimes some servos would work, etc.

So I tested all the 6 channels with the pair of Base Yaw and Shoulder pitch servos (channel 0 & 1 / 2 & 3 / 4 & 5) and observed the 6 channels responded ok.

After a while decided to connect in sequence each servo to see if one was the cause of the problem. Connected the following sequence: Base Yaw, Shoulder, wrist pitch, wrist roll and the gripper and everything was ok, each of the 5 servos responded ok. At that point the shoulder link was in vertical position, elbow down(no power), wrist horizontal

I turned-off the robot power, connected the elbow on channel 2 and when I power-up and the SSC-32 servo sequencer was reconnected (all servos were enabled and at their sliders at the same position as before), I observed the following:

1. Shoulder servo rotate forward (by gravity?) and the arm just laid flat on the table…
2. No other servos becomes responsive
3. Saw jerk movements on the wrist roll servo which actually rotated when I was trying to command the shoulder to get the arm up …

I turned-off the robot power, disconnected the elbow servo,power-up again confirm I could control all other servos with the sequencer and turned-off the power. I then connected the elbow servo on channel 7 and observed the same.

I disconnected all servos but the elbow and could control it adequately. However, confirmed that this servo rotate in the opposite direction than the shoulder & wrist. Wen I commanded the servo sequencer slider to the right, the servo rotates CCW, when I commanded the servo sequencer to the left, the servo rotates CW. (Which is opposite from what I saw on the shoulder and wrist pitch). Did not hear noise or did not feel excess heat or vibration on the servo.

I then reconnected the base yaw and shoulder pitch servo with the elbow connected and observed same erratic behavior again after may be 10-15 secs.

Disconnected the elbow servo and connected the 3 remaining servos and the arm was responding ok.

What do you think?

[scharette]

Hi,

It does seem like your elbow servomotor is causing issues (and might be defective).

Please provide us with some pictures of your assembled robotic arm? Make sure to show clearly all components (especially the assembly of the elbow).

You can attach images to your reply by going into Full editor mode and clicking “Attachments” above the orange buttons (Preview, Submit, Save draft, Cancel). See the attached image for details.

We await your reply to further help you.

Sincerely,

[sylvain]

Been doing testing and think that the elbow servo rotation direction is a non-issue. Because of the geometry, if you command up on the sequencer, it is normal that the elbow goes down since the servo is constrained by the shoulder bracket.

There is pict in attachment.

To get my system to work, I’ve ended up using channel 8 & 11 and, may be more importantly, been careful to have all servos in the off position when powering-up with the Servo Sequencer. If I do this and IF the robot is in a neutral position (meaning the delta between the servo position on the controller and the actual servos),and if the robot is up, then the system is ok.

However, it looks as if the controller can’t handle the extra load. For example, if I extent the robot in an elbow down config meaning the angle of the shoulder is about 50deg with vertical, the robot falls forward, I loose comms with the controller (bluetooth). Not sure if the extra weight of the wrist roll servo cause the system to trip??

To regain control, I need to turn-off the servos, only then the comms get re-establish, and then I would enable the shoulder servo and control it again by enabling one servo at the time.
Is there a current limit on this system (controller card?) that trips it off?? If the card is suppose to handle 32 channels, I can’t understand how I’m getting into this situation.

I did not mess around anymore with the elbow servo meaning connecting and disconnecting cables.

Not sure if this is normal or not with this technology.

[sylvain]

I reconnected my servos to the standard pin config (0 to 5). Getting the same result as described in my last post, meaning that my elbow servo seems to be ok. The system will respond only if I positioned the arm in a config that is the same as the servo sequencer and then activate one servo at the time.

If I activate a servo controller on the sequencer, and then send a control signal, if there is too much delta on the shoulder or the elbow position, then the system appears to “trip-out”, and the robot just falls flat as per the picture.

I suspect that is not a normal behavior…

I’m unable to get the Flowbotic Studio - AL5 arm to work. If I position the robot in a shoulder vertical, elbow down, wrist horizontal and line-up up front (base yaw) with the control software in the same position and then turn the robot power-on, the robot connects for a sec, I see some jerking movement and it looses comms, reconnect for a sec, jerking movement, loose comms, etc.

Basically, I can’t use the robot as-is…

I don’t think anymore that it is a servo problem. When I first submitted the problem report, I was using the USB connection and now using Bluetooth, so don’t think it is the comms path either.

Is it possible I have a faulty SSC-32U controller card? Do you have any recommendation for further troubleshooting or is it normal behavior???

As info, I never keep the robot power-up for more than 15mins at the time. Also, none of the issues described so far involved putting extra weight on the arm: it is always load free.

Sylvain

[scharette]

Hi Sylvain,

Thank you for your detailed posts and images of your setup.

From your many tests and descriptions, it would seem like you do have a power issue, which results in a brownout of the SSC-32U (and therefore loss of communication, restarts, etc.).

The most likely culprit in this case would be the power supply.

Please open a support ticket with us here so we can proceed to further diagnostic and a (free) replacement if needed.

In your message, please copy a link to this thread so that the technician helping you has context of what has been done already. Please also attach a picture of the power supply you are using (the label with the information of the supply should be legible). We request this kits sold at different times may have a different power supply model.

Sincerely,

[sylvain]

Thanks Sebastien

Je crois que c’est le bloc d’alimentation de puissance. Je vais en recevoir un nouveau et on verra.