It's taken me a while to get around to installing my 2 new ServoStiks, and installation itself took a while, too. Here is my review...well, a whole lot of disjointed numbered bullet points is more like it...
1 - Design. Totally clever, easy to take apart & work on, etc. I love how the 4 microswitches are captured by the restrictor plate. Slick!
2 - Throw. Going by subjective feel, the ServoStik feels much much better than my old Mag-Stick Plus. The ServoStik's throw feels far longer. Compared to my "main" Player One stick on my panel (a Happ mechanical rotary stick), the ServoStik throw isn't as satisfying (in terms of distance thrown off of neutral), & the square restrictor feels funnier than smooth "round" motion, but it holds up OK. The square restriction seems like something inherent to a switcher stick, so I'm not counting it as a drawback, just noting that round restriction feels nicer to me in 8-way mode than square, in general.
3 - In my order was a baggie with 5 little set screws. Where do these go? In this same order, I also got a steering wheel for Ultimarc's spinner, but that had no set screw holes in its hub...
4 - I found that 1 of the Fiberglas motor mount boards interfered with a button near the joystick. It was a simple matter of taking a beefy set of angle cutters, snipping the board, & then sanding the rough edges smooth. I wasn't using the board to mount the controller PCB anyway, so I didn't need that Fiberglas real estate, & that portion of the board is not in the load path between the mount fasteners & the motor mount features, so there was no harm in trimming it.
5 - Based on the presence of ejection pin circles, the white restrictor plate looks to be injection molded. However, the 4 "banana" shaped slots to clear the mounting screws appear to have been cut with a mill or router. These features were "hairy" as heck, with tons of dangling material. Not wanting to have little white plastic hairs falling into my control panel over time, I deburred these slots. Upon further inspection, these holes provide generous clearance for the 4 assembly screws that pass through them--why weren't these slots injection molded to save cost? The central square/diamond restriction feature also appears to have been machined but deburred nicely before shipping. All other features look to be as-molded: foreshadowing--see #13 below..
6 - During assembly & handling, I accidentally pulled off 1 wire from 1 motor & damaged the other wire tab on the same motor. Be careful when handling: small wires & small tabs! Consider potting the tabs or adding strain relief. On my 2 ServoStiks,I re-soldered all 4 tabs on both motors & then potted the connections w/ hot glue & then also looped the motor wires through the motor retention band twice for strain relief.
7 - Speaking of the motor retention band, it's totally clever. I've never seen this means of mounting a motor to a gearbox before. I wanted to remove the to inspect it & look inside gearbox, but I feared breaking the elastic band in the process--if I accidentally snapped it or introduced a cut while prying it off, it would be difficult to repair or replace. Wonder how this band lasts long-term. Any heritage for reference?
8 - The compression springs. 1 spring was in the wrong place in the sandwich: between restrictor plate & microswitch, not between screw head & washer. Upon many disassemblies & reassemblies in my troubleshooting (see #13 below), I noticed that the springs often pass through the plastic washer holes when torquing the screws, so they end up not performing their job of keeping the restrictor plate pushed lightly against the stick housing. The washer holes should be slightly smaller or the spring ID should be an little bigger to prevent this. I imagine that vibration & handling over time might lead more springs to weaseling their way into the washer holes.
9 - When reassembling the joystick, make sure that restrictor plate is seated--sometimes requires a decent press to snap it in (see #13 below for related discussion). Seating often requires pushing the plate in w/ your thumbs--the 4 small springs alone are too weak to seat it.
10 - Software. I'm no software guy, so I had a little trouble. I'm using these in "Hardware Mode", where pressing a momentary admin button on my CP switches them to 4-way & pressing another button switches them to 8-way. I ran the Ultimarc program (Windows 7), plugged in the driver board (no motors attached--just PCB), & Windows recognized it. (Going from memory here!) After Windows did its thing, I clicked "4-Way" & believe I saw something like "4 detected" in the little box in the program. I clicked "8-Way" & I believe it said "8 detected." I wasn't sure if this meant that the software was reading some feedback from the PCB & confirming that the command had been sent. I clicked "Upgrade FW" for fun & the USB device umounted. I made these clicks just out of curiosity, as I was going to use Hardware Mode anyway, but I wanted to see what the software buttons did before making this irreversible change. I clicked "Hardware Mode" & got a warning/confirmation box (good!). I confirmed my Hardware Mode decision & that warning box disappeared, returning me to the program's main screen. Did it work? I don't know? No feedback of confirmation? Hmmm? I repeated the process & got the same results. I guess it worked...? Later, when I installed it into my cabinet, I couldn't get it to work at first, so I removed it, took it back to my laptop, repeated the above process, & then it did work. I'm going to lean towards operator error on that last weirdness, but the moral of the story is that I would have had an easier time w/ some sort of software message or PCB blinking LED notification or something to give me some feedback.
11 - Microswitches are super loud. This comes from somebody who's neurotic about silent button & joystick actuation, who spent hours & hours on various folded paper shims in microswitches ( http://forum.arcadecontrols.com/index.php/topic,72400.0/all.html
) & GGG Micro-Leaf switches. (I ultimately settled on Happy Cherry K-Series w/ paper shims for all buttons & Paradise Arcade Zippy "silent" microswitches for all sticks.) Flashback: because of the inherent short throw of the Mag-Stik Plus in 8-way mode, I'd found that these microswitches actuated a little too soon & make it difficult to hit the 4 cardinal directions without going diagonal (i.e., if you break up the stick's throw into arcs, the diagonals' arcs felt much longer than the non-diagonals' arcs). My solution at the time was to file the outside ends of the microswitch plungers to make the plungers .290" long, which resulted in a little more stick throw to actuate, which made it a little harder to hit the diagonals accidentally. Because the restrictor is a square & because you're pushing against a flat square side when going N, S, E, or W, if your direction of push is not quite normal to that square edge, the stick will tend to want to slide to a corner & go diagonal on you. So, I transferred my Zippy microswitches into the ServoStik to replace the stock Sanwa switches included w/ the ServoStik (swapping the lever arms & the plungers so that the ServoStik had full-height plungers). Same issue as with the Mag-Stik Plus. I could barely go N/S/E/W--diagonals were far too plentiful. (Smash T.V.'s diagnostic mode is a great utility for testing straight-vs-diagonal tendencies--I use it when fussing w/ my leaf switches on my "Robotron" sticks.) So, I swapped in my filed-down .290" high plungers into the ServoStik & one worked fabulously, while the other was now adjusted too far & I couldn't hit 1 of the diagonals in 8-way mode. So, another plunger swap back to the stock plungers (.305" high or so, I think?) from the ServoStik's Sanwa microswitches & that second ServoStik was now behaving nicely & quietly. (I used to be a machinist, BTW!) Moral of this story: you might want to fuss w/ the switches...if you're a fussy guy...
12 - Musing on restrictor feel. I wonder if these 4-Way/8-Way adjustable sticks in general could have their straight square sides bowed out slightly. I sketched this up quickly (see attached), not paying too much attention to proportions. The dotted lines represent the existing designs & the solid lines represent my idea. In 8-Way mode, pushing in one of the 4 cardinal directions might result in fewer undesired diagonal actions if the user is pushing into a shallow "pocket" instead of against a flat surface. Diagonal actuation in 8-Way mode should not feel any different. In 4-Way mode, when sliding from one direction to the next, bypassing a restricted diagonal, there would be a slightly "rounder" & less straight travel, but the travel couldn't be *so* off-straight that the diagonal would be triggered when going from one direction to another. Just some food for thought that might improve the feel of switchers--this would probably require lots of prototypes & testing & would be quite tedious! :\
13 - The actuation needed a lot of modification to work. Out of the box, the motors would stall when about halfway through their travel. I uninstalled & reinstalled the sticks several times over, removed them from my panel & worked on them "loose" on my workbench, etc. etc. See this other guy's video for reference--I'm not the only one!
There were a combination of interference/friction/binding issues that ganged up & prevented the restrictor plates from moving smoothly: a - The servo horn exiting the gearbox was too high relative to the restrictor plate & the bottom of the restrictor plate rubbed against the top face of the servo horn. b - The OD of the restrictor plate was too large relative to the 4 walls of the stick housing.
Solutions: a - Sand the OD of the white restrictor plate gently bit by bit until it spun more freely. b - Ditch the compression springs altogether & instead use another set of nylon washers to shim the restrictor plate up slightly (so it won't rub on the servo horn) & to remove the compression springs from the equation, as they tended to work their way into the restrictor plate. c - I didn't want to do this, but it seemed like a good idea & adds insurance--I put some light grease on the OD of the restrictor plate & in the restrictor plate's slot in which the servo horn boss rides. All these adjustments combined make the ServoStik servo quite nicely now. Perhaps just 1 or 2 of these 3 solutions would be enough, but I'd spent so much time on these that I wanted to really kill this problem & not have to revisit it later.
Recommendation: instead of screws with full thread, how about shoulder bolts that will bottom out at a known height when torqued into the joystick housing's threaded inserts, & then a precise stack of slippery shims/washers can be used to take out the restrictor plate's axial play, rather than use those compression springs.