Asked — Edited
Has anyone used the mg995 servos for heavy duty work? (moving omnibot 200 arms/head) Would you recommend them? I've read that they tend to overshoot when positioning. If so, is it highly noticable?
Also, anyone know what make the ones are forsale in the ez store?
Upgrade to ARC Pro
Become a Synthiam ARC Pro subscriber to unleash the power of easy and powerful robot programming
As another question then, can anyone recommend a reasonable priced one available in the uk?
Looking to control 2 omnibot 2000 arms, and pan tilt for the head.
on 11-22-2005 XJet posted the following on the RCUniverse RC Radio Forum:
"I've been checking these servos out and can report the following:
1. they are very torquey -- I haven't measured the torque yet but it may well be the amount claimed.
2. they are quite fast, perhaps as fast as claimed -- again, I've yet to actually measure them.
3. they are all metal geared -- even the gear driven from the motor shaft appears to be metal, unlike most other MG servos which use a nylon gear at this point in the train.
4. they do appear to operate with a higher motor drive frequency than a regular non-digital servo so may well be digital.
5. the servo-amp is not well matched to the motor/geartrain/feedback pot. In essence, this mismatch manifests itself as significant overshoot when the servo is repositioning. This is to say that if you command the servo to move clockwise to a position, it will actually move past that position (by 3 degrees or so) then move back to the correct position.
Even more strangely, if the servo is at rest and you try to move the arm, it actually moves *towards* the pressure being applied.
6. resolution is "okay" but not as good as most other brands of digital servos and there are even some good non-digitals that are better.
7. without a servo -arm, the weight is 2oz exactly, not the 1.78oz advertised.
8. these servos adhere to Futaba standards (spline and direction) but come with a JR/Hitec connector.
9. physically, the assembly leaves a little to be desired. The PCB on all the units I inspected was incorrectly inserted into the case so that it was sitting on an angle rather than square to the bottom (not a biggie but perhaps indicative of a lower QC standards?). The feedback pot and motor are connected to the PCB by wires and dabs of glue support these wires only where they mount on the pot. The wires to the motor are unsupported and will almost certainly break through fatigue over time if the servo is exposed to moderate or high levels of vibration.
So is this servo a bargain or a bust?
I guess it depends on your expectations.
If you're looking at it as a sub $20 budget servo then it's not bad at all. You get what would appear to be a bullet-proof metal gearset, a supposedly coreless motor, a digital amp and plenty of torque at a faster than average speed.
However, if you're planning to compare this servo to something like the Hitec HS5945 (digital, coreless, metal-gears which sells for around $90) then you'll be disappointed. Its' when put alongside a "quality" coreless digital that the MG995 shows that it really is a low-cost servo. The quality servos have a markedly better resolution, are rock-steady when repositioning and don't even budge when subjected to load -- all areas where the MG995 suffers.
The ultimate test however, is how do they fly?
I threw one on the rudder of my Katana 3D plane (replacing an HS5925) and I did immediately notice the slower speed. The plane also felt less "solid" in rudder response during knife-edge and in a hover -- but the Hitec certainly wasn't over four times better (as the price might suggest it should be).
If you fly iMAC or turbine-powered models (where precision and reliability is of paramount importance) then I strongly suggest you stick to the more expensive options. However, if you just want a bullet-proof servo that has a snot-load of torque and costs little more than a standard servo then these are worth a look.
I would strongly recommend however, that people take a moment to drop a dab of hot-glue or acid-free silicon glue on the motor leads -- just to pre-empt the inevitable."
Hope is is some help.
I have both 995 and 996R servos but I havent truly put them through any tests.
I've gone with the lot of 4 MG996R's I found on ebay, so will see how they go. I'm sure they will be perfect for the job.