Thread: My 3 axis skull...in progress

Tweet

Originally Posted by Deathly

In a nutshell, many RC and Robotics guys run their servos on 7.2 volt battery packs. You only need to watch the maximum voltage rating of the servos. Servos are able to produce more force with higher voltages anyway.

I believe the regulator on the SSC-32 will keep voltage in line for the logic. The controller and servos just need to share a common ground connection in order for signals to pass correctly.

I think 18 gauge wire is a bit of overkill. I know that several people including the Captn have used long lengths of cat-5 wire. The trick to that is to simply double up the wires for the power feeds only, the data/servo pulse wires carry very little signal. You can also run a heavier wire (like zip cord) along side the cat-5 dedicated to power feeds only (V+ and Ground) or even power the servos remotely only bringing back the common ground for signal transfer.

Cat-5 is made up of 8 24 gauge wires rated at .57 amps each, so double would give you better than an amp transfer capacity and less resistance than 22, triple even more yet.

Thank you for your input but I still have a big concern.
I understand that the SSC32 will run off of 7-9VDC so that is easy from my PC Power Supply.
And I was, from reading all the posts, assuming that my servo (HS-425BB) would be ruined by anything over 6VDC. Looking back at the servo box I read 6 volts which confirmed my fears. And right now I am looking at the servocity.com web site and I see that the HS-425BB (I use it for all 4: nod/tilt/rot/jaw) is in fact rated for (4.8-6) a maximum of 6 volts. So would not a 7.2 volt battery pack burn up or at least stress my servos?

On the wire issue I'm getting really confused.
Yes 24 AWG is rated at .57 amps for transmission (3.5 amps for chassis) if it is stranded, not solid. I was originally planning on using 22 AWG stranded which is rated at .92 amps for transmission. So yes if you double up a 24 AWG you would get over 1 amp but with 22 I'm starting at almost 1 amp and the servos only draw .5 amp hence I thought I would be safe.

I did not consider either the voltage drop or signal loss at 25' distance, hence HBs recommendations. Doing the calculations, at 25' I will have an almost 7% loss with 22 AWG, the same with 24 AWG so either way I'd like 6VDC to run the servos. This will give me 5.585 volts with 22AWG or 5.835 with 18AWG. What will happen to the PWM signal to control the servos I have no idea, that's WAY beyond me.

What I 'think' you are telling me is to run: 1 common power line, 1 common ground line, and independent signal lines. And to use heaver gauge wire on the two common lines and Cat5 on the PWM signal lines.... Yes? If that be the case I could run a one pair 18AWG for the power and one Cat5 for the PWM signals and be done.

Oww I think I've sprained a brain cell.

Originally Posted by Sparkey

What I 'think' you are telling me is to run: 1 common power line, 1 common ground line, and independent signal lines. And to use heaver gauge wire on the two common lines and Cat5 on the PWM signal lines.... Yes? If that be the case I could run a one pair 18AWG for the power and one Cat5 for the PWM signals and be done.

Exactly.


Very little difference weather it's stranded or not. In reality stranded and solid carry the same current. Unless you are running high tension wires with AC current, then it's a different story.

I would think that running the servos continuosly under a heavy load at 7.2 volts would cause them to overheat. In this case you are running them intermittently at various loads and at a voltage that will be less than 7 volts (from the voltage drop on the wire).

03.09.2010
Well, I said I would get back with a test. First I'd like to say thanks to all with ideas and help. I like the idea of using 18 AWG for a common power set (+/-) and then using 22 AWG for the PWM Signal so I may be doing just that to save on wire. But for those who have been following this thread here is the results. Using 90' of 22AWG on the jaw servo the skull ran just fine. I just left the roll of wire on my desk and tried it and it worked. I have a short vid but I don't know how to upload it.

I created a youtube (first) for you:
YouTube- 3AS-90FtWire.AVI

Have fun all.
Sparkey

Yep, looks good. You might want to look up Dudedrummer, I believe he used 75 feet of cat-5 in his multi-skull setup and it worked fine. I have over 2500' ft. of older cat-5 (not the E type) lying around so I find all kinds of uses for it.

I'm going back and rewiring some of our skulls and I like the idea of conserving wire by using a common power wire for all of the servos. My 3 axis skulls have 3 cables running to them. I could reduce that to two if I wire the servos to share the same power wire. I'm swapping out old cat5 solid wire cable for stranded cat5. I'm wondering if I can get away with cat5 to carry the power for 3 servos on one wire or I could run power over two of the 8 cat5 wires if that would work better. I've never liked the solid wire because I've had issues with it breaking. After reading all of the concern over what gauge wire to use, I just wonder if I'll run into any issues with conductivity with the stranded over the solid. Solid worked fine running 18 servos at lengths of 25-40 feet and each servo with their own power wire. We're up to 9 servo animated props with 21 servos and I'd hate to go to all of the trouble to wire all of this only to find out I need heavier wire.

BTW, which wire is power? Red, black or yellow?

Black is usually neg, Red is usually pos and the Yellow is usually the signal wire. If you look up the brand of servo you are using I think that will tell you what the color code is. Also I think if you check your servo controller it will have the pin designations printed on the board, at least it does on mine,a Parallax board.

For your servos, red (+) and black (-) carry the power, yellow is the signal wire.

I used stranded cat5 to carry the power and signals to all four servos in my skull (and I'm using a bucky with 645mg servos which is heavy and pulls more power than a lindberg would). I run a distance of about 25 feel without issue, and even power static LED eyes in the skull.

The cat5 has eight conductors, so you can double up on the power lines (two conductors for +, two for -, and one for each of the signal wires). Capt. Jack had a good schema for standardizing the conductor assignments.

I can take pictures of my wiring harness tomorrow if you want. I use servo extension wires and splice them into the cat5. Then I have a female rj45 jack at eachend. Then I can use a regular cat5 patch cable to connect the board to the skull. I just made another set for my second skull this year.

-DGM

Hmmm, that sounds even better. Ya, I'd like to see just how that is connected. I'm guessing that you connect all of the red with red and black with black at the SSC32, then to the cat5, and then split it out in the skull. But then the picture would eliminate guessing.

Originally Posted by buckaneerdude

I'm going back and rewiring some of our skulls and I like the idea of conserving wire by using a common power wire for all of the servos. My 3 axis skulls have 3 cables running to them. I could reduce that to two if I wire the servos to share the same power wire. I'm swapping out old cat5 solid wire cable for stranded cat5. I'm wondering if I can get away with cat5 to carry the power for 3 servos on one wire or I could run power over two of the 8 cat5 wires if that would work better. I've never liked the solid wire because I've had issues with it breaking. After reading all of the concern over what gauge wire to use, I just wonder if I'll run into any issues with conductivity with the stranded over the solid. Solid worked fine running 18 servos at lengths of 25-40 feet and each servo with their own power wire. We're up to 9 servo animated props with 21 servos and I'd hate to go to all of the trouble to wire all of this only to find out I need heavier wire.

The only compelling reason for using stranded wire in low voltage applications is for flexibility. Conductivity differences between solid and stranded would be very hard to measure since the surface area is the same for both. The maximum capacity for any wire really depends on several external factors like air flow, ambient tempurature around the wire... If a wire cannot dissipate heat caused by it's own resistance, it will catastrophically fail. At a point where current through the wire is sufficient to generate heat, the wire's resistance value increases drawing more current.... then you get the light bulb filament effect.

Fundamentals of Electricity

I hope this clears up some things.