MG996R 55g Metal Gear Torque Digital Servo (4 Pack)

Price was awesome for 4 fast Metal Gear 20KG Servos! worked perfect on two of my 1/10 crawlers both with rear steer definitely an upgrade over the 6KG stock,

Only used 2 of the 4 so far but they are working great in my Tamiya TT02 and M05, so far. I'll update if they stop working like some others had experienced.

I got these servos to use for a small electronics project (completing a 100 jump challenge in Mario RPG, haha). It was a fun motivation to gain a little experience working with electronics hardware.My first attempt used a little 5 volt 2 amp power supply I had laying around, and wired directly to a Raspberry pi 3b. But it turned out this was not nearly good enough, as the servo would move erratically or not at all. I upgraded to a much stronger variable voltage desktop power supply and put a proper servo controller between the Pi and the servo, and then it worked great. There are wiring diagrams on adafruit and tips on reddit. The best way to learn is to try!

These aren't the best servos you can buy, but they might be the best you can get in this price range.I bought 20 of these to build a hexbot; overall they work OK. They're a little noisy and loose and sloppy in their control. I had 2 fail (lock up/burn up) and probably 2 more with positioning issues that cause them to oscillate when they should be still.All that being said they're "good enough" - I'll forgive a lot of these sins for the price point. With the better alternate being 4x the cost, they're cheap enough I can buy extras to swap out the ones that have issues.

Good servos tested well at 5V

Crazy torque for a cheap servo

Thus far, the single servo I've put to use has only been used in one test flight, so I don't know if there are any longevity issues with these servos. What I can tell you is that the ones I received are branded "Deegoo FPV" and all the gears are metal. I've seen other reviews mention plastic gears, which I was okay with when purchasing, as I needed cheap servos for my project. The photos of the servos with plastic gears have a different name on the stickers than mine do, so if you order these and they don't say "Deegoo FPV", they most likely have different gears, and I'd return them for replacements.The only con for me, is the grease seeps out of the casing, and can leave grease marks in a toolbox or plane. But at least they're lubricated!

I never even powered the servos on. while installing the horn on them I noticed a lot of "wiggle" to the horn. This is usually caused by improper fitment of the horn it's self. I tested with the horns it came with as well several others. I got the screw tightened up as much as I could get it to go without stripping anything. and the wiggle did not change at all.Upon using a magnifying glass to look at the spline shaft and horn while I wiggled it I found that the horn was not moving on the shaft but instead the entire shaft is moving.I pulled the 4 screws on the bottom and lifted the lid off and as advertised all metal gears, not much in the way of lubrication inside the thing. I put my finger on the spline shaft and it is wiggling. I remove the spline shaft and gear and low and behold what do I find.. a nylon bearing/bushing on the bottom side of the gear. The bearing/bushing on the top of the shaft is metal and is pressed into the top of the servo.This is not a good thing. There should have been a pressed in metal bearing/bushing on the bottom as well. I pulled the metal bearing/bushing out of the top and slid it down the spline shaft and this thing has got a MASSIVE amount of play in it. No wonder the spline shaft moves about. I can move the metal bearing/bushing fo it sits crooked on the shaft my about 1.5mm. That 1.5mm of play translates into 6 mm of movement by time you get to the end of the horn. That's 1/4" of wobble in the horn.Now when the entire spline shaft is wiggling that much that means the meshing of the gear on the shaft is not going to be ideal and will cause failure extremely early in use. probably right after the return period has elapsed.I purchased 8 of these things and all but one has this kind of play in it. I will be returning all 8 of them.The meager use of grease inside this thing means planned failure. There is maybe 2 drops worth if it could be applied from an eye dropper.***UPDATE***OK so I did a little bit of a modification to them. In my specific use case I am using a round metal servo horn that has 4 threaded holes for attaching something to the servo. I found that using a small piece of 1" x 1" x 1/8" aluminum L channel stopped the wiggle. I drilled a hole in the L channel so I could put it over the top of the servo so that the "L" would go down the side of the servo. This keeps the aluminum from rotating when moving the servo. The hole has to be large enough so the servo horn could fit into it.Doing this actually solved 2 problems, the first was the lateral wiggle of the spline, the second was servo drift when no power is supplied. To solve the drift issue the hole drilled in the L channel needs to be a really snug fit with the servo horn. This provides enough friction so that when powered off it doesn't move easily.I also want to provide a useful tidbit of information for people using these with micro controllers. I had a need to know where the starting point is when powered on, if the servo is not in a position where you need it to be if you send it a signal to go to where you need it to be it's going to move there as fast as possible. That movement might be too fast, it was in my case.If you open up the bottom cover you will see the 3 leads soldered down to a circuit board inside. If you look at the board so the orientation of it is so that the leads exit the case away from you just above where they are attached right at the edge of the circuit board you will see 3 solder points. If you solder down to the middle one you can use the voltage reading to calculate where the servo's position is.I do not know if this voltage is going to be different depending on what the supplied voltage to the servo is. it might be so be careful and test before connecting it to a micro controller. In my use case I am powering the servo with 5 volts and I am using an ESP32 which has a maximum input voltage of 3.6 volts on the GPIO pins. The voltage I would be reading surpassed the 3.6 volts so I had to use 2 resistors to make a voltage divider to trim it down so it would be lower than the 3.6 volts.The voltage reading can be temper mental It is best to take 20 samples with 0.5 millisecond delay between each sample, drop the 5 highest and the 5 lowest readings and average the remaining 10.I have not looked at deviation due to temperature yet. I will be doing that at some point.