I am looking into getting DC power supplies for my computer controlled experiment.
I am going to use 32 of one of these servo types…
(amp values publised by hitec)
Analog HS-422 - 180 mA (No load) 800 mA (stall)
Analog HS-225 - 300 mA (No load) ??? (stall)
Digital HS-5245 - 300 mA (No load) ~1.1A (stall)
Does anyone have an idea how much these servos actually draw at typical higher load operation?
The quote from Mike Dvorsky at the bottom of the post outlines power input limitations for the SSC-32.
He states: VS1 and VS2 should be limited to 10A respectively.
He also states: 200 mA is a typical servo load.
From above this typical servo load may not be valid as that barely covers the No load current requirements…
By my math: that gives a total supply of 20 A over 32 servos. Hence 625 mA per servo.
Can any comment if 625 mA is sufficient to drive the analog servos or digital servos above without sacrificing performance???
Note: the stall torque of the standard HS-422 is 800 mA and greater than 625 mA. I fear that the HS-225 and the HS-5245 will be limited even further by the power supply.
thanks,
Stuart
A quote from Mike Dvorsky (SSC-32 designer) Regarding this…
There is no single “correct” number for the current draw, other than the no-load and stall values.
Are you going to move the servos fast or slow?
Are you going to put heavy mechanical loads on them?
Are you planning to move 32 servos at once?
In a many-servo situation, the total current draw will be very dynamic. You will most certainly need to run the logic from a separate supply from the servos.
Most folks end up being limited by their power supply, and the wiring and connectors that they use.
Note that the SSC-32 doesn’t turn servo power on/off, so it’s mostly a matter of the size of the PC traces and the little screw connectors on the SSC (I wish those connectors were bigger/better).
It would be possible to run the DC power to the servos with separate wiring, thus bypassing the SSC (the SSC would still provide the PWM signals).
Yes the servos will be moving as fast as possible at 6 volts.
The loading should be moderate to high for a standard servo. Say a torque of 4 kg-cm.
Yes the servos will all be moving in a simultaneous manner.
I agree, the logic will be supplied by a 9V battery as described in the SSC-32 manual.
I am interested in this. I am using hitec servos, with a yellow, red and black connection wire. I assume red and black are the power supply and ground wire, and the yellow wire is the PWM signal wire. Is this correct?
Should I consider buying a second SSC-32 to control half of the servos??? Will I have significant problems with delay between the two SSC-32 controlled servo motions???
The SSC-32 has all the power to controll all 32 channels with all of the features. Using two SSC-32’s would be missing the whole point of the power it has. You can move all 32 channels to new positions with a single command. They will all start moving at the same time (there is a short delay between each bank of eight) as soon as the SSC-32 receives a CR. You can even have them all start and stop moving in synchronization with each other by providing a “T” time for the move to take.
If current consumption of the servos is a concern, you can connect the black wire to the SSC-32 ground AND the supply ground, connect the yellow wire to the pulse output on the SSC-32, and the red wire to your external supply. This can be done with a “Y” connector.
The SSC-32 was designed to control up to 32 servos connected directly to the board with the expectation that not all servos would be drawing a lot of current at the same time. Most users of the SSC-32 will use 12 to 18 servos, some will use as many as 24 for bipeds, and yes many will only use 6. For the power user external power routing may be required. But hey, it’s a damn good controller for $39.95 Eh?
By the way…
When Jim mentions a short delay between the four banks seperate sendings, he’s means really short; small fractions of a second.
It’s not possible to see the delay in any of the servos.
I agree that I will probably have to use an alternate power source in my application because of the loading and the rapid movement of all my servos. Thanks for the info details on how to do that.
Jim… I don’t wish to modify the existing servo wiring.
**Do you recomend buying servo extender cables which can be cut and modified to connect servos to an external power source??? **
Modifications would be
Red power supply cable: Cut the red supply wire and connect it to the power source
Ground Wire: Solder on another ground wire to the existing (uncut) ground wire of the extender cable to the SSC-32. Connect the soldered wire to the ground of the power source
No modifications to the yellow PVM signal wire.
The extender wire would be connected to the SSC-32 board as usual, but the power supply wire would be coming from another source other than VS1 and VS2.
**Is there any need to connect a ground to VS1 and VS2??? **
or
Can I simplify this by grounding all the servos to the external power supply through the negative ports of the VS1 and VS2??? ie. not soldering anything to the ground wires of the servos…
Most importantly, Am I getting this right, or are there better options that this electrically uninclined mechanical engineer is missing?
I’ll post some info on current requirements of loaded servos for my experiment after I’ve got some data.
sorry if the question is old … but I wanted to know how much current (in Ampere) passes thru the SCC32 to reach the servos ( I mean … how many amperes feed the servos)
The SSC-32 can handle 15A per side for a total of 30A. This has been tested and verified for 18 servo walkers and 23 servo bipeds. But both of those applications didn’t even come close to the limits.
