Hey guys,
I am currently working on a servo-based robotic arm project with a few friends and I am in charge of electronics. However, I am having some trouble with the powering of the various components. We are using 6-7 servos and they will collectively require about 15-20A at maximum torque. So, I have insisted on getting a 5V 20A power supply (https://www.amazon.nl/-/en/gp/product/B07Q2WQ8DS/ref=ox_sc_act_title_1?smid=A6FTR3WNTF6EM&th=1). My question is - how do I safely distribute the power from the power supply to each servo and the Raspberry Pi that we are using for control? I know that breadboards are out of the picture at this current, but what about perfboards? Or is there even a better alternative?

 

how do I safely distribute the power from the power supply to each servo and the Raspberry Pi that we are using for control? I know that breadboards are out of the picture at this current, but what about perfboards?

The power is distributed with wires of the appropriate gauge.
Don't understand what that has to do with breadboards/perfboards(?).

 

Hey guys,
I am currently working on a servo-based robotic arm project with a few friends and I am in charge of electronics. However, I am having some trouble with the powering of the various components. We are using 6-7 servos and they will collectively require about 15-20A at maximum torque. So, I have insisted on getting a 5V 20A power supply (https://www.amazon.nl/-/en/gp/product/B07Q2WQ8DS/ref=ox_sc_act_title_1?smid=A6FTR3WNTF6EM&th=1). My question is - how do I safely distribute the power from the power supply to each servo and the Raspberry Pi that we are using for control? I know that breadboards are out of the picture at this current, but what about perfboards? Or is there even a better alternative?

Make a distribution box to mount close to the supply. It should contain suitable terminal blocks or connectors for each device using the supply. Each block should be independently wired back to the supply +/- terminals. Along with each pair of power connections for the servos should be a the control signal connection which should be wired to a multi-pin connector for the Raspberry Pi.
If the terminal blocks or connectors are surface mount they can be assembled directly on the distribution panel.

 

Make a distribution box to mount close to the supply. It should contain suitable terminal blocks or connectors for each device using the supply. Each block should be independently wired back to the supply +/- terminals. Along with each pair of power connections for the servos should be a the control signal connection which should be wired to a multi-pin connector for the Raspberry Pi.
If the terminal blocks or connectors are surface mount they can be assembled directly on the distribution panel.

Thanks for the suggestion, think this could work the same?: https://www.amazon.nl/-/en/PCB007-2...1700321139&sprefix=power+distr,aps,235&sr=8-5

 

Thanks for the suggestion, think this could work the same?: https://www.amazon.nl/-/en/PCB007-2/dp/B0814B5P5M/ref=sr_1_5?crid=Q44W9797TMTM&keywords=power+distribution+board&qid=1700321139&sprefix=power+distr,aps,235&sr=8-5

That would work for the power but you still need to provide a means of getting a Pi signals to each of the servos.

 

That would work for the power but you still need to provide a means of getting a Pi signals to each of the servos.

Great, the PWM signal to the servo should be more simple as I can use wires (jumper wires probably) directly to the servo as those wires would carry a relatively small current

 

Great, the PWM signal to the servo should be more simple as I can use wires (jumper wires probably) directly to the servo as those wires would carry a relatively small current

It is not so much the current you have to worry about but the signal integrity at the receiver. You don't want to have an arc welder next to the robotic arm causing herky-jerky arm movements every time it discharges 100A into a weld.

I have to ask if you have any experience with this kind of thing.

 

Great, the PWM signal to the servo should be more simple as I can use wires (jumper wires probably) directly to the servo as those wires would carry a relatively small current

Best to ensure that all current carrying conductors to servo's etc are twisted pairs for e.g.

 

It is not so much the current you have to worry about but the signal integrity at the receiver. You don't want to have an arc welder next to the robotic arm causing herky-jerky arm movements every time it discharges 100A into a weld.

I have to ask if you have any experience with this kind of thing.

To be fair I don't have experience with this specific type of project. Could you please explain what the issue would be with the signal integrity and how I can avoid it?

 

Best to ensure that all current carrying conductors to servo's etc are twisted pairs for e.g.

I know this may be a stupid question, but why is that?

 

I know this may be a stupid question, but why is that?

Reduces and/or eliminates EMI, twisting produces a cancelation effect of any radiated energy.
Also if a GND conductor is ran at the same time, it should be wired un-twisted with any other conductor.
Also on the question of circuit boards, I prefer Vero strip board. See Stripboard https://www.futurlec.com/Protoboards.shtml

 

When data is transmitted on a cable there are at least two coupling mechanisms that could allow noise to corrupt the data being transmitted. One mode is capacitive coupling which happens when conductors are parallel to each other. The other is magnetic coupling where strong magnetic fields occur in proximity to a conductor with data on it.

 

When data is transmitted on a cable there are at least two coupling mechanisms that could allow noise to corrupt the data being transmitted. One mode is capacitive coupling which happens when conductors are parallel to each other. The other is magnetic coupling where strong magnetic fields occur in proximity to a conductor with data on it.

Thanks mate, really useful stuff! I'll look into it further once we have the actual physical components at our disposal.

 

Devices such as Servo/stepper motors radiate from the supply cables, others, such as encoders etc, suffer from effects of radiated noise, EMI, etc.

 

Reduces and/or eliminates EMI, twisting produces a cancelation effect of any radiated energy.
Also if a GND conductor is ran at the same time, it should be wired un-twisted with any other conductor.
Also on the question of circuit boards, I prefer Vero strip board. See Stripboard https://www.futurlec.com/Protoboards.shtml

Would it work to just twist the 3 loose 14 AWG power wires (+, PWM and GND) together (maybe like this?

), or do I have to go for specifc machine-twisted servo wires? All I can find online at the moment are servo wires with the 3 wires twisted together but they are 22 AWG (too small) and quite short too so I'm not sure.

 

You could possibly do that, the GND I was referring to was Earth GND conductor, if ran with others.
Not power GND.

 

Would it work to just twist the 3 loose 14 AWG power wires (+, PWM and GND) together (maybe like this?

), or do I have to go for specifc machine-twisted servo wires? All I can find online at the moment are servo wires with the 3 wires twisted together but they are 22 AWG (too small) and quite short too so I'm not sure.

Which servos are you using and how long will the longest leads be?
It really is best to keep the supply wires separate from the signal wires. Twist the power wires together (4 or 5 twists per inch) to minimize radiation. Twist each signal wire with a wire which is connected only to its digital group ground on the Pi. Leave the other end of the ground wire open circuit.

 

Which servos are you using and how long will the longest leads be?
It really is best to keep the supply wires separate from the signal wires. Twist the power wires together (4 or 5 twists per inch) to minimize radiation. Twist each signal wire with a wire which is connected only to its digital group ground on the Pi. Leave the other end of the ground wire open circuit.

We plan on using 7 servos - 2 20kg-cm servos (2.7 A each), 1 40 kg-cm servo (≈ 4 A), 1 6kg-cm servo (1 A), 3 microservos (0.8 A each), hopefully controlling them using a Raspberry Pi Pico (if possible) or Raspberry Pi 4. Now I don't have any specific details on the longest leads but I could see the longest reaching MAX 60-70 cm. These wires will likely be 14 AWG in accordance with current ratings.

 

This is an excellent pub. on current control wiring methods.

 

Here's the Servo list by the way:
Servo at base: use https://www.tinytronics.nl/shop/en/...otors/td-8120mg-waterproof-digital-servo-20kg 1st Joint: use https://www.amazon.nl/-/en/Digital-...60-be20-dbcbffc23490&pd_rd_i=B0C64NLF1Q&psc=1 and if possible gear ratios
2nd Joint: use https://www.tinytronics.nl/shop/en/...otors/td-8120mg-waterproof-digital-servo-20kg
3rd Joint: use https://www.tinytronics.nl/shop/en/mechanics-and-actuators/motors/servomotors/s3003-servo
4th Joint: use https://www.tinytronics.nl/shop/en/mechanics-and-actuators/motors/servomotors/sg90-mini-servo
5th Joint: use https://www.tinytronics.nl/shop/en/mechanics-and-actuators/motors/servomotors/sg90-mini-servo
Opening and Closing for the "hand": use https://www.tinytronics.nl/shop/en/mechanics-and-actuators/motors/servomotors/sg90-mini-servo
Total Current Needed: ≈ 12.8 A
Total Cost: € 59.74