Ramps 1.4 servo issues

Hi folks,

I'm trying to control a 9g servo with my ramps board. It works great until the printer is printing something, then the servo will meander aimlessly the entire time. Obviously I can't have my servo wandering like this.

Could my ramps board be bad? Is there any sort of stabilizing circuitry I can add to my servo so this doesn't happen? Maybe a cap?

Any other ideas?

Thanks!

Have you wired in a dedicated 5v power source for the servo, or are you running it from the Arduino regulator? If from the Arduino, I think you just discovered that that regulator isn't up to the task, and you would benefit greatly from adding one.

If you have, then shielding (which I am not sure will help, since PPM to the servo is so low frequency) . . . other than that, not many ideas . . . I have speculated that on the deltas, since the processors get so busy that it might send some oddly timed PPM, thus twitching the servo, but it's just a theory - ai have no evidence to back that theory.

- Tim

Nope! No 5V regulator, just pins off the board going straight to the servo. Ah, so you are suggesting using a whole other 5V supply, just for the supply pin on the servo. Interesting idea. I would probably use a wall wart going into a 7805. Does that sound like a sane idea?

Thanks!

That's what I did - 7805 and a small cap dropping the 12v to 5v to feed the servos. If you really want to go low-road, tie the cap ontomthe regulator, attach and shring tube the output lead, and stuff the input and ground into the 12v input connector to the RAMPS . . . works quite well for me, since I discarded the crap Chinese knockoff Molex connectors and went to screw down Euroblock . . . The 7805 isn't the most efficient solution, but considering that we are starting with a 30A supply, it's not like the small loss is going to be missed either . . .

- Tim

Sounds good! I never knew about this issue.

Do high quality RAMPS boards not have this issue, or is it a design flaw that should be addressed in RAMPS 1.5? Perhaps I should file a bug report.

Cheers!

It´s an issue from the mega board. RAMPS is only providing the servo pin headers.
If you want higher efficiency than the 7805 solution, buy a RC-UBEC circuit instead.
It is plug and play instead of solder and pray
-Olaf

The 7805 is plug and play as well, and the efficiency difference largely irrlelvant . . . I'm not a big fan of over complicating solutions to simple problems . . . and there is a lot less to fail in a 7805 than some cheap offshore RC product . . .

And the regulator on the Mega is *NOT* the actual problem, but rather an artifact of a bad RAMPS design. It was never designed to run any and everything that folks hung off the board, but is perfectly adequate for the Mega. Why a 5v regulator was not included on RAMPS boggles my mind, considering the loads of servos, LCDs, etc. which are absurd to thing the Mega should power . . . . so yes, a 5v source in RAMPS1.5 only makes sense . . . be it a simple regulator like a 7805 to keep the cost down, or whatever . . .

- Tim

Or...if you're using a PC ATX power supply just connect a 5V line to the 5V servo pin. No extra need for a regulator. In either case, an inline fuse isn't a bad idea Just In Case™.

Quote
cdru
Or...if you're using a PC ATX power supply just connect a 5V line to the 5V servo pin. No extra need for a regulator. In either case, an inline fuse isn't a bad idea Just In Case™.

In that case, RAMPS should still have a legitimate 5v input connection, since servos aren't the only thing that loads the 5v rail.

Idea: Add a connector with 5v in, ground, and 12v out. Those with an ATX, tie on the 5v line. Those using a Chin-Fat-Fong el-cheapo RC BEC, tie on 5v and ground. Those choosing a 7805, put it in the connector . . . . Seems to cover all bases at trivial cost (1 3 pole Euroblock . . . ).

Thoughts?

From #reprap IRC, it sounds like the servos should be able to accept a wide range of supply voltages. Thus, the suggestion was that my "meandering issue" isn't related to a sagging supply voltage.

I'm unsure either way. Thoughts?

I just built an external 5V supply for my servo. It works, but the servo meandering issue is still there.

Thus, this is _not_ a problem with the servo Vcc.

I opened this bug with Marlin:
[github.com]

Anyone else have this problem?

Duh, there is a Marlin option for this. Sorry to waste everyone's time.

The option in question is
#define PROBE_SERVO_DEACTIVATION_DELAY 300

I also had the very same problem with my ramps 1.4 board. The 5V pin is not connected to the arduino's 5V pin. But carefully following the traces of copper on the board I realized the Vcc pin on the Ramps 1.4 board is connected to the 5V pin on the arduino board.

The easy solution I found was to connect the Vcc pin in the Ramps 1.4 board to the adjacent 5V pin so that the servo pins get the proper alimentation using one of those jumpers. Just conect Vcc to 5V (one is already next to the other) and you're good to go!

Works like a charm and no soldering needed.

Hope this helps!

Edited 1 time(s). Last edit at 02/21/2018 09:16AM by rgalate.