Hi, I built the arm some time ago and while the construction took time I got everything complete. I used Dr.D-flo’s mods so I could house a touchscreen in a case and also hold the fans to cool the Slushengine. Ever since it’s been built the motors don’t have the strength to move the arm, the steppers keep stuttering while trying to move the body. It’s not as if there’s a major obstruction in the way, they simply don’t have the power. I’ve run a multimeter off the power supply and it’s a steady 24 volts. I also connected the double 24 volt outputs on the power supply together, thinking that there might be a current issue but that doesn’t seem to be the case, as the voltage never sags when i move the motors, the arm simply doesn’t have the power to move itself. Can anyone suggest a solution because I’ve very frustrated with this considering it cost so much money and time to construct it and the instructions left a lot to be desired. All suggestions appreciated.
I have a very similar situation - however most of the joints move adequately well except the shoulder joint (with the double motors).
I know it’s wired up correctly, because when upright I can move it both ways very well. However when the arm is pointed down, the shoulder joint can’t overcome its own weight to lift the arm. With some assistance it can make it, but it’s a long way from being able to make it on its own. Have experimented with setCurrent and a few different values, but nothing is sufficient to make it lift its own weight. the robots.py also worked great on all other joints but not sufficient power to lift shoulder.
Belts are tight. Voltage does seem to drop a couple volts when Raspberry Pi is hooked up (even though RPI is powered through 2.5A power supply), I assume it’s drawing from the 5v pins on GPIO. I may cut those lines on the ribbon cable to see if that helps… Otherwise we are many hours in and a significant investment and we’re stuck. Any help much appreciated.
What values have you tried for
setCurrent? Have you tried make changes to
setMicroSteps? Usually, if using full microsteps you get 100% holding torque. However, using a different value, the holding torque reduces and resolution increases.
The RPi doesn’t consume much amps, therefore it’s not worth cutting the line.
Have a read of this New robot arm 3D printed - I think they used a planetary gearbox for the shoulder.
I purchased a new power supply, and now I have adequate power. Two of the Power Supply units from roboteurs were insufficient / defective. Voltage would hover around 19v, and sometimes drop down to almost nothing. With the new power supply (not roboteurs) I now have steady 24.1V (and the motors / joints seem to have plenty of power) no matter what i’m doing.
Just to be safe I have supplemented the RPi power through USB - my understanding is that we are not supposed to power the RPi over GPIO as there are some current protections on the Micro USB port.
I’m sure the SlushEngine does have a power regulator and fuse protection for the RPi GPIO, to protect it from over-voltage, current spikes etc. However, if your voltage drops to zero, then the RPi will most likely reboot – saying that, your voltages should not really drop to zero!
I’ve had a similar issue with the arm. A few recommendations I have tried:
- Make sure the weight of the arm is on the bearing and not on the plastic. Make the holes larger at the 8mm rods locations.
- Increase the setCurrent(), but beware of not overheating your motor.
- Confirm how heavy the independent printed parts of your arms are. I was able to reduce the weight of many of them by reducing the infill to as low as 10% in some places. The arm was able to lift a 200g weight at maximum range at first. Somehow, it diminished after some time and I haven’t played with it more yet.
Hope it helps