I have my SSC-32 connected to my hexapod. I then use Lynxterm to control the servos. When I initially flip the switch “ON” and click the “ALL=1500” button, my servos twitches.
I then continue to click the “ALL=1500” button for a few times and the servos finally positions to the 1.5msec position.
Why does it take a few clicks (commands) of the “ALL=1500” for my servos to actually respond?
My servos on the hexapods are all HS645MGs (18 of them), I have SSC-32 V1 with 1.06E firmware (this phenomena was noticed even with older firmware), and for power source I’m using 5xAA NiMH 2500mAH Energizer cells that I soldered myself with copper battery bars.
The battery packs are soldered to connector with beefy 16AWG stranded wires and there are mating connector/wire with the same gauge wire on the SSC-32 soldered onto the PCB of the SSC-32 on VS+.
The strange thing is, I pass the command “serout” on the botboard to do a group move that moves all of them to position “1500” (1.5msec), and this does not seem to have any problems.
I’ve only experienced this problem when I ONLY use the SSC-32 by itself and Lynxterm to send serial to the SSC-32 via DB9…
This is the oldest problem in the book. lol Your SSC-32 is resetting. Your battery pack’s voltage is dipping when you turn on all 18 servos at once. Go figure… The fix is to:
Get a battery pack with higher current delivering capacity.
When trying to move 18 servos, the instantaneous current draw forces the voltage to drop, been thinking about slapping a 47uF cap between the VS terminals… I’ll probably give that a whirl and see if it’ll improve… I’ve also been thinking about doubling up two 5xAA packs to double up the current source… Thanks for the reply!
I measured my 3DOF CH3-R with my DMM today. I’m using 18 HS645MGs with SSC-32 (v1 w/1.06E firmware), and powered by a single 5xAA 2500mAH Energizer battery pack hooked up in series (roughly 6.8V-7.0V open circuit after a full charge). The body of my round hexapod is propped up with a plastic bucket, so all six of the legs are “dangling” sort of speak. Here is some useful data:
“ALL=1500” Command on Lynxterm after “ALL=0” (Peak power with no loading on the tibias):
min Voltage = 5.5V <–this is what’s killing my SSC-32!
max Current = 1.7A
When all the servos are at 1500 and holding (Average power flow while maintaining pulse width/position on 18 servos since the servos are holding position but not commanded to move):
avg Voltage = 6.3V
avg Current = 0.8A
When all the servos are commanded off with “ALL=0” command (quiescent current of servos + SSC-32 + BotBoard w/Atom Pro 28):
avg Voltage = 6.8V
avg Current = 0.040A
Something to consider when selecting how large your bypass cap (probably should be 100uF or higher) should be for a similar setup. Oh word of caution when caps are fully charged. You can “load” the bypass cap with a 10K-Ohm resistor to “bleed” the charge out of the capacitor if you would like. A 10K resistor wouldn’t load the power down much but will aide in eventually bleeding the cap when the unit is shut off for a while… Also, I use the minimum “x2” rule of thumb when selecting the “WVDC” (working voltage, i.e. breakdown). Preferrably “x3” For a 7V source such as your battery, I’d select minimum of 14V so transient spikes doesn’t destroy the cap. Here is a cheap brute from Jameco:
You’re getting 5.5vdc with the robot up on a bucket?
Remember that the capacity of the battery pack (2500mA) is not a reflection of the maximum current the pack can deliver. In other words your pack should be able to deliver 2.5 amps over a one hour period. However there is no guarantee the pack can deliver 10 amps for 15 minutes even though the math may suggest it will.
Other things that can adversly effect power transfer is small guage wire, twisting wires instead of soldering, cheap plastic battery holders, bad switches, etc.
I would expect a good 6.0 to 6.5vdc with the robot up on a stand.
The SSC-32 (1.06XE) has a 4.0vdc reset. The 2.01XE version dropps out at 4.3vdc. The voltage regulator needs only 5.5vdc on the input in order to provide the 5vdc output. It will follow the input voltage less .5vdc for input voltages less then 5.5vdc. So for you, the battery voltage is actually dropping down to 4.5vdc for the SSC-32 1.06EX to be resetting.
The hexapod should draw around 7 to 8 amps when walking. Hope this helps…
Yeah, I understand. The 2500mAH is just capacity and is independent of the discharge rate, which is the real current sourcing capability of a battery.
The discharge rate depends on few things like the type of battery(nimh, nicad, lipo, etc…), the way it is constructed, and its internal resistance.
I’m leaning towards just separating the power supply for the servos and the ssc-32/botboard. Probably going to use the weakest pack out of my six for the ssc-32/botboard and use the stronger ones for servos…
All my packs are soldered together with beefy battery bars made of copper. I’m using my RC’s 16 AWG stranded (dean’s wet noodle, possibly the best stranded wire you can buy for hobby use). I’m also using duratrax power poles, which has been my favorite connector for about 15 years…
I’m currently using the power switch carried over from my biped scout, going to use this to switch the battery pack for the servos. The switch you guys pack in with the round hexapod kit is going to be used to switch the electronics. I’m sure glad you guys packed in that switch!
The fix is to:
