I’m building a tracked bot using the tritrack system. I want to make a pan and tilt drive for a wireless camera. The drive
will be controlled from a ssc-32.
I hooked up the servos of the p/t system to channels 0 & 1 of the ssc-32.
The servo shown (pan) in the video is on channel 0 and is driven by the command Print “#0 P800 T3000” and Print “#0 P1200 T3000” (using a Atmega 128 micro controller)
I want a slow movement and see that the movement is not smooth throughout the movement range. Is this how it should be or am I doing something wrong?
The commands you are sending is not enough. What’s sending the commands, and how often are they being sent. The video doesn’t make any sense. It should be moving a longer distance and the moves should take 3 seconds… Can you also post images to show the wiring?
VL=5.5V fro main pcb power supply
VS1=5V from BEC circuit (I’m using a 12V battery to power all the system)
Baudrate is set to 9600
I’m sending each command only once and the servo changes position (the commands are sent by the hardware UART of an ATmega128 micro ).
The led on the SSC32 light ON when power is first applied and then blinks each time a command is set.
The 3 sec. movement is just about right.
What am I doing wrong ?
Your power supply is suspect. You are feeding a 5vdc regulator with 5.5vdc. I would like to see 6vdc minimum on that line. The video shows several small movements. Are the movements happening each time you send the command?
Well since you are using a Low drop regulator on the SSC32 I think that the voltage is OK for powering the board. In any case I’ll try a higher voltage the see what happens but I doubt that it is the problems since I do not have resets on the SSC32…
The small movements start when a command is sent and finish when the command is finally finished buy the ssc32 after 3 seconds.
I’m sorry, I don’t understand your reply. I still would like to see the big picture here. What’s sending commands to the SSC-32 and when does it happen. Also are these the only two commands you are sending?
Print “#0 P800 T3000” and Print “#0 P1200 T3000”
And what is that clicking sound when the servo is moving?
Do you have another servo to try in its place?
Don’t tell me you have a broken gear in the servo and that’s the only problem…
As I mentioned earlier, the commands are sent to the ssc32 from the hardware uart of an ATmega128 micro controller. The controller has 2 button. Each button corresponds to a command. When I press button “A” then the first command is sent and when I press button “B” then the second command is sent.
I tried the servo by controlling the ssc32 from lynx term and it seems OK.
If you look at the video, the servo makes about three short movements each time a command is sent to the ssc32. In each short movement there is a clicking sound.
Let’s say the servo is in position 800. If I send the command so it will go to position 1200 then the servo will make 3 short moves and stop in the 1200 position. the same goes the other way around.
If I decrease the time value in the commands from 3000 to 1000 then the movement is fast and smooth.
Try connecting the SSC-32 to LynxTerm. Add some friction to the servo horn, just put your thumb on it. Slowly move the slider for this channel. Watch to see if the servo output goes limp at a certain point. If it is a gear this will show it.
A servo with a few teeth missing on an intermediate gear will act normal at some speeds and not normal at other speeds.
Sounds like the problem may be with programming on your ATmega128 micro controller. You may want to use the speed instead of timed movement to keep the rate of movement constant between points.
You are essentially saying because the T commands appears to not be working for you try the S command instead… If the T command is not working then why do you think the S command would do any better.
This is clearly hardware. There is no way to foul up the SSC-32 commands in an Atmel microcontroller that would result in the generation of a clicking sound from inside the servo.
I had a chance to test a new servo today an sure enough the new servo was OK. I took apart the faulty servo waiting to find a broken tooth in the gear but all the gears were OK. So the fault was in the electronics of the old servo. Anyway, thanks Jim for your help.
What!.. It does not matter if you use S or T, the result is exactly the same. Using a T of 3000 is a 3 second move. I think it qualifies as “slow”.
The video doesn’t look anything like slamming abrupt movements. You are right in as much as I did make an assumption that the code was sending the proper commands. He said the commands are sent when he presses the buttons, and the SSC-32’s LED blinks once per button press… If the servo was getting spurious commands “slamming” the servo, wouldn’t the LED be blinking when this happens? No LynxTerm does not have the ability to heal broken hardware. But he never said he tested it using the same commands with LynxTerm did he? No, he said he connected the servo to LynxTerm and it appears to work correctly. I suspect he just used the slider and moved it back and forth. This would not show the gear problem as I mentioned before. Yes, I still think it’s a gear, but it doesn’t really matter. It was the servo anyway.
Hey guys, the problem is solved. Jim was correct, it was a servo problem which was related to the electronics of the servo and not the gears. The gears are in perfect condition. I have two servos that have electronic problems. I used lynxterm to test the servo using the slider and saw that it reacted OK. In fast speeds the servo reacts Ok, the problems shows when the servo movement is slow.
