I’m really having problem with getting my sabertooth 2x5 r/c to work properly with ssc-32.
Symptom: When motors are running, all the servos become very jerky
Port connections are as follow:
Sabertooth forward channel 1 connected to port 0
Sabertooth turn channel 2 connected to port 1
Servo 1 connected to port 28
Servo 2 connected to port 29
I tried 2 configurations for power on ssc-32:
VS1 to VL to VS2 (default)
Disconnect VS1 from VL and VS2
Sabertooth is connected to 9.6V battery.
When VS1 connected to VL connected to VS2, I got the 5V from the saberbooth through port 0 and 1 and it runs the whole ssc-32 without connecting to another battery.
When VS1 is disconnected from VL and from VS2, I connected VL and VS2 to a separate 6V battery.
In all cases, when I sent forward motion:
#0p1360#1p1500
the motors move just fine. However, servo 1 (port 28) and servo 2 (port 29) both become very jerky.
I have change all kind of combination of jumpers on the sabertooth but nothing help.
Hi Jonathan,
I don’t have a sabertooth, but I would try isolating the power from the sabertooth from the SSC. You will still want the ground and signal, but if you pull the power socket out of the plug that goes into the SSC and isolate it, that might help.
So what you’ll have then, is the ground, for refernce, and signal between the two. Each will need to be powered separately.
I don’t know if this will work, but I’d give it a shot. You can always plug it back in.
I did that in one of my mentioned configurations where I isolate vs1 from vl and vs2 and I used a different battery for vl and vs2 (6v) than the battery used for the sabertooth (9.6v). That didn’t help much
I’m a little confused as to the connections you REALLY have. Your descriptions are a little confusing. I really think you have power issues. Power the VL from a 9vdc battery. Remove the VS1 = VL jumper. Power the motor controller from it’s own 9.6vdc battery, but this is the important part! Make sure the Sabertooth is not sending power back to the SSC-32… You can do this by removing the red/orange wire from BOTH of the Sabertooth connectors. There is no other way to dissable the BEC on the Sabertooth. I would be very suprized if this doesn’t fix your problems…
I was thinking the same thing that if the motor and servos have their own power source (separate batteries) then there should be no issue unless the motors are creating some sort of interference with the servo signals.
Yes Jim, I disconnected jumper VS1=VL and also disconnected 2 jumpers for VS1=VS2 and I connected VL and VS2 to a different battery than the sabertooth’s battery and still had the same problem.
My problem is probably what you mentioned: power being sent back from the sabertooth. I didn’t remove the wires so basically the whole VS1 bank was getting power from the sabertooth. I was confused of why it does that but didn’t know what to do with it so I tried to isolate VS1 from logic power source and VS2 power source which the servos are connected to (but that didn’t help).
I will definitely try removing the orange and red cables and update the result but i’m sure it will fix it.
This little beast of a problem just won’t go away . Here is what I did:
Disconnect middle red wire of the sabertooth connectors (I verified that it’s the VCC line that outputs 5V so I cut off this line and left only ground and signal).
Remove VS1=VL jumper
Connect sabertooth to a 9.6V battery
Connect VS1 to a 6V battery
Connect VL to a 9V battery
So this is as much isolate as I could do. However, the problem just won’t go away. If I send something like this to the motors: #0p1370#1p1500 then I can’t see the servo jitter
But if I do: #0p1300#1p1500 then the servos will become jittery (not as bad as before but still does a little bit which will be pretty annoying for camera).
Just across the terminals or one from each terminal to the motor case as well? Ideally you use 3 capacitors, one across the terminals and one from each terminal to the motor case.
Wow a 25mm motor with a stall current of 10 amps! I don’t have any difficulty believing this motor can create some electrical noise. The three cap filtering method would be recommended.
Heh, a 4T or 5T RC car motor perhaps? Still, it seems like if you control your ground connections and use seperate power supplies it should not mess too badly with the servo outputs of the SSC-32.
Which brings me to the question… how are you routing your grounds? Are you daisy chaining the grounds between boards or bringing the SSC-32 and Sabertooth back individually to a battery common distribution point? This is important because the SSC-32 is delivering both the power to the servos AND the signal ground for the PWM signals to the Sabertooth. IF the ground to the SSC-32 is twitching due to the current draw on the motors it could be coupling to the servos as well.
Thinking about this, another question… are you using analog or digital servos?
The ground lines for the batteries are completely separated Eddie. However, the signal ground is linked between the sabertooth and the ssc-32 since I have to connect the ground and signal lines from the sabertooth to the servo ports of ssc-32. I am not sure when I disconnected jumper VS1=VS2 and used a different battery for VS2 (where the servos are connected to), there is still some ground connection between vs1 and vs2 (servos still jittery then). I don’t know of a way to completely separate the signal ground of the sabertooth from ssc-32. All my servos are just standard analog servos.
I had to do something last night so couldn’t get to fixing my capacitors yet. Will do today. Just to clarify Eddie, I will need 3 for each motor right?
Also, is there a way to to attach image directly to the post rather than uploading it to somewhere and link the url? I want to post my diagram or maybe a picture of the setup.
yes there should be 3 caps on each motor, one between the terminals and one from each terminal to (typically a common point on) the motor case. be careful not to damage the case, a lot of heat very fast may be safer than a lower wattage iron taking several seconds to heat up the connection.
there is not yet anyway to post pics inline that I am aware of.
my thought on analog vs digital servos is this. a digital servo is measuring the pwm pulse width in time, resolving it to on and off, typically through a buffered input stage with hystersis to reject noise. an analog servo can be (depends on the servo) actually using the average energy in the pulse in an analog fashion. voltage drops in the ground lead due to current spikes can appear as signal to the servo. this can make analog servos susceptible to forms of electrical interference that the digital servos may not be (conversely digital servos can be susceptible to jitter in the pulse width which get averaged out by analog inputs). If you have even one digital servo available to test with it would be very interesting to know if this problem persists while using it.
I played around with your advice. It helped a little bit. The servos were still a little jerky. I finally got a whole new set of motors and the problem went away.
. Here is what I did:
The three cap filtering method would be recommended.