Is it possible that a new “specialized” bracket can be made that would have a “mount anywhere” SES application? Does it have to piggy back on the motor? A specialized bracket would allow the flexibility of mouting the board anywhere with the exception of wire lengths. An example is the low profile SES bracket. It can be mounted anywhere with the standard SES bolt pattern.
The point is there is an electrical desire to keep the control PCBA in close proximity to the motor terminals. Introducing “cabling” of any sort between the h-bridge output and motor degrades the efficiency of the control. Somehow that needs to be balanced with the whole mechanical SES fit goal.
I understand.
I wonder how much performance would be sacrificed if cables were used.
I will have to whip up some conceptual designs when I get home tonight.
SN96,
That’s the spirit, thanks!
On long cables, inductance and capacitance in the motor cables will make the FETs work harder, generating more heat. The bigger problem is long cables for the position feedback. The longer the cable, the more noise it will pick-up and beyond a certain point, the motor will end up chasing the noise because the servo thinks the position has changed. The pot cable noise will also cause more hunting at the end of a move to a new position.
Cliff
Has any actual field testing been done with various cable lengths between the pots and the control boards, and if so, at what length do issues arise?
That’s what I was thinking. At what point do the elements begin to influence the performance of the design. I used soft leads for my Parallax Ping sensor, around 6" worth, and it worked with the same accuracy. Test before and after the softleads showed no signs of any performance problems. I understand this might be a different ball of wax but it’s just used as an analogy
In the case of the motor controller, heat seems to be the main issue. It would be interesting to see what happens with cables vs. no cables. I think ideally you want to have the board mounted as close to the motor, but when considering flexibility with other SES parts, some compromises might be needed for some sort of balance. Tonight I will play around with some bracket ideas for the purpose of discussion and hopefully spark new ideas.
zoomkat,
The answer to your question is that it depends on how the cable is routed and what electrical fields it encounters. If the cable is routed away from other cables and away from servos, it can be a few feet long and not be a problem. If its routed in a bundle with power cables and other servo cables, it can pick-up enough noise to be a problem in a few inches.
In the case of the new servo, the temptation will be to route the motor and pot cables together and since the motor cables will be carrying a lot of current, they will induce current in the pot cables (think of it as a single turn transformer). Since the pot is typically 10K Ohms, it doesn’t take much induced current to generate a noise voltage large enough to be a problem. There are solutions to this problem, but none of them are as cheap as keeping the cables short.
Cliff
With the use of the gearhead motor setup, I’d think a priority would be the locating of the control boards remote from the gearhead motors themselves. It would make for a much cleaner mechanical setup, and possibly for group cooling of the FETs being used on the control boards.
I’d think an appropriate value capacitor on the pot line would add some noise filtering. If so, it might be good to also incorporate it on the board itself.
Cliff said that this could be done but it’s cheaper to keep the wires short if possible. If a satisfactory bracket design cannot be produced due to design limitations, then a cap might be something to consider.
zoomkat,
As is, the servo firmware filters the high frequency noise, but if a capacitor is added and is large enough to take care of low frequency noise, it will affect the position signal (slowing changes) and the PID control loop will overshoot on position changes. The cure could be worse than the disease.
Cliff
LOL! 
I have a few bracket designs in my head. Work is dragging, I want to get home and get crackin…
I have been designing some motor mounting brackets also, and have at least one design that might be relevant to this project I could share if there is interest. It mounts the motor under the bracket with SES mounting holes on two bracket surfaces. I designed it for a GHM-04 gear motor, but I am sure it could be easily adapted for any motor. I already have the model for the design of the bracket and a model of the GHM-04 motor in 3D CAD. 
I also have in mind a slight modification that would mount an Open Servo board on the bracket just above the motor or at least as close as practical to it.
8-Dale
The cap is already in place on the PCB for some minimal filtering. Any further filtering will be taken care of by software.
Barry
Post what you have so everyone can review it. I’ll ad my concepts as well later.
I believe the whole PC board mounting issue could be resolved with an additional bracket, just a plate actually, that would mount to the base of the assembly and project under the motor. I will have to check clearance but I think it would work. I am a little concerned about the cost of another bracket, as I am trying to keep it as inexpensive as possible. But lets not stop all of the progress due to an issue of mounting the PC board. I’m sure I will resolve it. I just need some time to consider the options. There are some sheet rubber products that could be added to the bottom of the board and allow the flexibility to mount the board to the motor without the added cost of another aluminum bracket. Anyway even after the board is found to work effectively and reliably, I can manage to make some minor changes to the board to accomodate mounting, such as adding mounting holes, etc.
OK, here is what I have so far, and of course it is subject to modification to handle specifics required by the new servo. I can easily make a new version of this specific to the new gear head servo and can create 2D CAD drawings from my 3D models.
SES Motor Mount + GHM-04 Motor
This is a 3D PDF, so you can rotate it to show a everything at different angles.
8-Dale
What size board are we talking about? This would be helpful.
How about a new lexan bracket that is SES compatable?
While things are still in development, I’d suggest that the motor/pot combo wiring be terminated in a small multipin connector, as well as the controller board output. This way the controller board could be mounted on the motor, or could easily be remotely mounted if desired. If the board is going to require an attached heatsink, then hanging all that gear on the motor will add some extra suspended weight, as well as a significant “ugly” factor to the looks of an arm setup.
lol, um a little trust here. I’m not into ugly… 