Thread: Max servo cable length?

Tweet

I have a feeling I'm going to have to adjust my setup, but I have some servos in a prop I'm trying to set up for the holidays (controlled by a medusa dmx board) and ideally I'd like to have the board and power supply in the garage. This would require around a 30 foot cable to connect the 5 servos to the board. I'm guessing there is no way that that would work, would it?

Thanks for any thoughts.

Can you put the board in your prop so that it is close to the servos, and then use a USB extender to put your PC in the garage?

USB over CAT5 Extender:
http://www.cablestogo.com/product.as...1521&sku=39993

No USB extender would be necessary, since he's already using DMX.

Short answer

I'd recommend against it, if you can avoid it.

Long answer

It's a gray area. Depending on the specific wire, what's next to the wire, which servos, and servo moves you're using, you may have a problem, or you may not. Since this question comes up frequently, I'll try to go into a little more depth. There's two main places you could run into problems: the power and the pulse signal.

Power wires (potential for voltage drop)

If you're using long skinny wires for the servo power (e.g.: CAT5), the servos may pull enough current to cause a significant reduction of voltage at the servo. This would result in lower servo torque and speed.

A way to help prevent this is to use shorter and/or larger gauge wire for the power leads

Pulse signal (potential for electrical interference)

Long skinny wires also can act as antennae, picking up electrical noise (think, for example, any time you hear a 'hum' from audio equipment). When signal wires are run for long distances right next to other wires with electrical noise on them, the signal wires will get polluted with the noise from the other wires. In this case, the 'noisy wires' could be AC wiring, other motor or solenoid wiring, or - even the servo power wires. If the pulse signal wires get enough interference on them, the servos will have random, convulsive moves.

There's a few ways to help prevent this. 1) Separate the servo power leads and the servo signal wires (as above). 2) Wire the servo signal wires using the shielded versions of CAT5/6. Connect the shield to GND at one end of the cable. 3) Keep any other wiring from running next to these wires. If wiring has to cross it, do it at 90 degrees (so they don't run parallel to each other)

The bummer about this kind of thing is it might work fine.... until you move one set of wires. Or until you turn on all of your props. Or until you change your VSA routine to have more things moving at the same time.

Hope this helps
- Hook

Thanks Guys - that's what I figured, but I thought I'd ask. I'll just build a waterproof container and put the board and power supply in there and place it close to the actual props.

It may be too late, but I did run my servos through a 100ft CAT5 with no problems. The control lines were actually in the CAT5, and the power and ground were all grouped on one pair of 16gauge wire ran along side the CAT5. I had no problems. My results may be atypical, but I had no problems whatsoever.

Cat 5 cable is cheap. I recommend you try Cat 5 first and if it doesn't work go to plan B. We had a total wire monster crisscrossing all over our graveyard. We ran 32 servos on "stranded" Cat 5 with an average length of 35 feet. (I like stranded because it is much more flexible and lays flat on the ground) It crossed all over over AC power cables, DMX cables, 12 volt wires and other Cat5 cables. I had one jaw servo that moved when a different jaw servo in another prop moved. Interference? Nope, it was a cold solder joint on the ground wires for that servo.

Last year when I was experimenting and developing the 2010 routines, I got this "interference" problem. The wires were all piled up in a mess on the garage floor but after hours of testing, it turned out to be weak connections and not interference. Interestingly, it was always the ground connection.

I have no doubt that interference happens. For me though, it hasn't been the result of anything but poor solder joints. As far as starving servos for power, I never had a problem, not when there were 4 servos running on 1 Cat 5 and not even when running a 3 high power servos. I even tapped into the servo power for eye led's with no problem. In a 4 servo skull, I soldered together, the orange and white and the brown and white Cat5 wires with all 4 of the power wires for the servos and the + wire for the LED's. I did the same for the ground wires for the servos and LED's with the green and white and the blue and white Cat 5 wires. I wired the solid brown wire to the Jaw servo signal wire, blue to rotate servo, orange to tilt and green to nod.

There are a number of threads in the forums here about "potential" interference but I don't remember reading anyone posting that they actually encountered the issue.

With new props and wires everywhere getting ready for this Halloween, I did a little video of part of what we are working on and purposely captured the wire mess criss crossing everywhere. There is no signal interference or crossover signal at all. The video shows 2 DMX controlled skulls, 6 3axis skulls and 2 2axis skulls.

I have a theory that somehow the stranded CAT5 that we use somehow is less susceptible to transmitting or receiving interference. Near the end of the night last Halloween we noticed a jaw moving slightly on a prop when another's jaw was moving. We didn't test it all out back then but when unpacking things to get ready for this year I noticed that we had used a ready made CAT5 cable with solid wire for that prop. My theory may be off, but so far it seems like there may be some merit to it. I suppose too that it could be that the stranded cable I use is better quality. The stranded cable I use is from a 500ft spool made by Belkin (I crimp the connectors on myself), the ready made I used last year was Fry's el-cheapo.