1/16/2024 0 Comments Tower pro 9g datasheetThe unit comes complete with 25cm wire and 3 pin 'S' type female header connector that fits most receivers, including Futaba, JR, GWS, Cirrus, Blue Bird, Blue Arrow, Corona, Berg, Spektrum and Hitec. This is the digital version! The unit features a stalling torque of 1.8kg in a tiny 9g micro package. "Not accounting for any possible backlash in gearing." I started the incremental moves at (nominally) 88 degrees.The SG90 from TowerPro is the most popular 9g servo in the world. Might even dream up a way of automating it - a precision pot or rotary encoder coupled to the servo shaft comes to mind. ![]() If I got at all serious about performing such calibrations, I'd write a bit of VB.NET code to do all the calculations and plotting for me, but it was just a quick'n'dirty one-off to characterise the SG90. If I had not used this calculation method, the results would have shown an (approximately) linear trend away from the true angles, not the random scatter that I actually got. AutoCAD calculated the angles for me (effectively using the inverse tan() ), and I then copied them into an Excel spreadsheet to plot the deviations. I marked all the spot positions on the paper, then measured them with a vernier caliper and drew the results in AutoCAD. Many hours in the physics lab, maths class and machine shop taught me that (among other things ). I have enough knowledge and experience to realise that the distance away from the perpendicular would affect the calculated angle. Its basic trigonometry you can also plot out with any protractor. So given the semi random set up you have I would say your servo performs quite well. However I cannot imagine how the program for a simple test like this would exceed 9000 characters. If it is a program then you can add it as an attachment. System won't let me paste it - says it's over 9000 characters :(. It would be instructive to get a regular potentiometer and, with a chart like in Reply #5, move the potentiometer to different angles and measure the resistance at each step, or, the Arduino ADC value at each step - is it any better than the servo? I imagine that the accuracy of a hobby servo depends almost entirely on the resolving power of the internal potentiometer that identifies the position of the servo arm. If each point is an average then what is the range of values from which the average was calculated.Īnd what was the servo position BEFORE it moved to each of the specified angles Your plot only shows one value for each angle - what does that value represent? Is it a single test or is it the average of several tests. The deviation varies, up to 1.6 o, apparently randomly. The plot in the 2nd attachment tells the story. If anyone would like to donate me a better quality servo, I'd be happy to test it for them! Unless it's the Arduino output that's at fault! No 'scope here to test that theory. I don't know whether more expensive, branded servos would do any better but these cheap ones are not worth the effort of trying to use when accuracy or higher precision are needed. I gave up after a few increments the servo moved in steps of about 0.14 o, sometimes it didn't move at all, and other times it moved after a few seconds then again after another few seconds. ![]() I tried it with the laser 2m from the paper. ![]() Using the Servo.writeMicroseconds() method, given a range of 1000 to 2000 usec corresponding to 0 to 180 degrees, 1500 should drive the servo to 90 o, 1501 should be 90.18 o etc. I was originally going to test the precision. I then calculated all the angles (or rather, AutoCAD calculated the angles for me ).Īs you can see, the actual angle can be over 1.6 o away from the requested angle ! I marked the position of the spot with the servo set to nominally 90 o, 91 o. I got some laser diodes and glued one to a servo horn, and wrote a sketch to control the servo by typing in angles, increments etc.ġst attachment is a photo of the rig, on the settee in my pokey little flat. Think cardboard, Stanley knife, hot glue, sticky-backed velcro. I'm working from home now, so gone are the days of nipping out to the materials store for some bits of ally, then round to the Bridgeport to do some machining and thence to the fasteners racks for some M3 x 12s or whatever. I got a couple of cheapo (unbranded) SG90's to play with, and wondered how accurate they were, and if it was worth trying to control them more precisely than the 1 degree increments offered by Servo.write().
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |