ELEFANT JAGDPANZER SD.KFZ.184
INSTALLING LIMIT SWITCHES
07/11/13- A typical elevation mechanism is industrial era-tech: It's just a motor with a worm gear that drives a train of gears to reduce the speed and increase torque at the output. At the output end, there's an arm that rotates slowly, raising or lowering the barrel (depending on which direction the motor is turning) via a linkage attached to the barrel. When your gearbox-actuated barrel elevation (or traverse on a tank destroyer like the Elefant) reaches the end of its travel, it can either stop with the gearbox clutch clicking (to protect the gears), or reverse direction if you've defeated the gearbox limit tabs, or eventually strip gears and stop working. True servo-actuated barrel movements don't have this problem because the limits are built into the electronics of the servo.
The limit switches turn off your gearbox motor when the barrel reaches the end of its travel, so there's no slipping clutch sound, reversal of direction, or stripped gears. Sounds like a great idea, huh?
Limit switches are one of those RC tank modifications that sound like a great idea and are the sort of thing that a tinkerer might do when he starts to run out of ideas for cool things to do to his latest tank project. My motivation was slightly different. Yes, it seemed like a cool thing to do to my latest RC tank, but I was also really distracted and annoyed by the difficulty in getting the barrel elevation to stop after doing nearly anything with the left stick. That's mainly because the transmitter doesn't have vertical self-centering on the left stick. When I traversed the barrel (moving the stick horizontally), if I let go of the stick and returned to horizontal center slightly off on the vertical center, the barrel would raise or lower. Same for starting up the tank, firing the machine gun, or firing the cannon by the left stick.
It doesn't help that the activation range for the ElMod circuit is very close to neutral stick. The elevation control, while in theory proportional, acts like an on-off control since there's so little fine control to give adequate feedback. The stick works fine for RC helicopter throttle because of this continous operational feedback. For something that acts like an on-off control, you have to know exactly where it will kick in or you may unintentionally trigger it, then correct and have it trigger the other direction, then correct, etc.
In my Elefant, the limit tabs on the Tamiya gearbox were folded down, so the elevation reversed direction. However, the way it was mounted caused the gearbox arm to eventually bind during the course of its full 360 degree rotation, so at that point, it started clicking from the clutch slipping.
Even though the slipping clutch is annoying, the 360 degree arm rotation is as annoying, if not worse. Unless you keep track of it, you don't know which stick direction will make the barrel raise or lower. When the gearbox's arm rotates to the top to start the reverse cycle, there's often an unloading effect from gravity and the slop in the gearbox-- the barrel suddenly drops and does nothing for a while until the gears catch up. So the Elefant's elevation setup had both annoyances. It was very confusing too.
Because of this, the limit switch was a useful mod. When I run the tank now, upstick always raises the barrel and downstick always lowers it. If it's at the end of its travel it doesn't try to move any further, and there's no distracting clicking sound. The barrel elevation sound effect still plays (from my understanding, the elevation and traverse were hand-cranked, but I prefer having the sound), but when the engine sound is running, it's drowned out. So in effect, it doesn't matter if the left stick is perfectly centered because the worst that can happen is that the barrel will be either fully raised or fully lowered. Since the elevation range is relatively limited, neither looks odd.
This might not be necessary if my Elefant had a transmitter with a fully centering left stick (one's coming). It's easy to live with an occasional clicking clutch sound, because if you hear it, you release the stick and it stops. You know it can't be raised or lowered any further, so you stop. That's why I didn't bother putting one in the horizontal traverse. It's reasonably controlled; when it reaches the sides, it stops traversing and I release the stick.
That's probably why it's a relatively uncommon mod-- it's a neat trick, but probably not worth the trouble it takes to install it. You also lose a bit of range because the motors turn off when the switch is actuated, which is rarely at the true end of the barrel's travel. It's hard to predict exactly where the switch will "make" at the last bit of travel. Nevertheless, it is a neat trick for the tinkerer and folks who are interested in this sort of thing.
The heart of the limit switch is the single-pole double-throw (SPDT) lever switch. As its name suggests, a lever depresses a button that makes electrical connection alternately down 2 different paths depending on whether the button is pushed or not. The lever is for leverage-- it makes it easier for a moving thing to push/trigger the button. You need 2 switches at the 2 ends of the travel.
The "trick" part is the common rectifier diode (1N00(n)--the black ones with a silver band at the end). A rectifier diode has positive and negative polarity and will pass electricity in only one direction (because of this, they're used to convert 2-direction AC power to single-direction DC power in a DC power supply).
When the barrel is in the middle of its range, both switches connect to wires that complete the circuit back to the motor, in both directions. The motor direction and speed is controlled solely by the stick.
When the barrel is at the end of its range, the switch connects the diode in the circuit. The diode at the actuated switch will block DC power in the direction that it was traveling (in the pic above, the green path), so the motor is turned off. Elevation movement stops.
If you move the stick in the other direction (in the pic above, the red path), the polarity changes so the diode can conduct electricity as if it were a wire (almost-- there's a voltage drop, so less power). The motor kicks on and goes the other direction. Once the switch is "untripped", the diode goes out of circuit and the plain wire completes the circuit (the motor speeds up slightly).
The electronics are dirt simple if you can solder: just solder a diode across the two output switch paths, observing the correct polarity. The hard part is figuring out how and where you're going to mount the switches and fine-tuning their positioning so that they operate like you want. Sounds easy, but different setups require different solutions. It can be as challenging as figuring out how you're going to make the elevation mechanism and its linkage work. That's because they really should be designed together so they don't get in the way of each other.
The "brute force" way would be the simple and obvious way-- just put switches on either side of the barrel. Practically speaking though, there's rarely a situation where you have such mounting options in the space that's available, especially if you want to make it reliable and fairly easy to maintain. The are usually better options around the gearbox, triggering the switches by movement of the control arm. One reason is that it's where the motive force originates, so it does a better, more reliable job of pushing the buttons. Unfortunately, the arm and the linkages often get in the way of placing the switches, so you may have to make do with other areas. Like I said, each setup is different, which makes it a challenge.
The barrel up is limited by the switch on the left, activated directly by the gearbox control arm when it rotates upwards and pushes the lever up. The barrel down is limited by the switch on the right, which is activated when the rear of the cannon mechanism rises and pushes the lever up. These pics may have been taken with the switches in the wrong positions-- I don't remember.
Some test "tricks" are useful to help figure this stuff out. This also helps you design how the elevation linkage works. Basically, you remove a gear or the motor from the gearbox, so you can quickly simulate by hand how it works without using your transmitter or a power supply. This helped me discover that the Elefant's elevation gearbox had a gear that was rubbing against the mount-- it's something you can feel when you rotate gears by hand that you might not notice when it's driven by motor. You can also use this to help you place the linkage's anchor point on the thing it's moving: Rotate the gearbox arm and observe how different linkage attachment points might affect way the elevation works. As you do this, think about where you might put your limit switches.
A variable DC power is a useful tool to have around, for this and other things. I used an electroplating power supply to verify the skeletal circuit for operation before attempting to install it. The circuit was designed on paper first to figure out the pathways and the diode polarity, then assembled to the switches and motor. Running it in one direction and then the other while manually actuating the switches verified that it worked and which switch to put at each end. (I marked one of the switches for full up elevation position, then ignored it while doing the actual install and spent too much time "winging" it with wire and diode changes to get it working again. Arrrrgh. So much for careful planning.) The power suppy is also useful for testing elevation mechanism installs since you can run the motor very slowly; you can also use it to break-in motors and gearboxes. (And electroplate.)
I mounted the switches with servo tape and hot glue. (Servo tape by itself can be unreliable over the long term.) I would have preferred to screw mount them, but discovered that what I thought were mounting holes through the switches actually weren't, and were probably used to assemble them-- probably not a good idea to drill through them! A craftsman-like job would probably involve building a mounting to secure them, but that would be a lot of work, and you need room to do it. Servo tape and hot glue provide a strong, secure attachment so they're a practical, flexible solution that seems to work in a lot of tough situations. If you've made an error, they can be removed (although hot glue residue can be tedious to completely remove).
There are a lot of things to keep in mind when you're figuring out this stuff, and the Elefant is especially full of challenges. In the final assembly phase of doing this mod, I also discovered that the servo on the back of the gun limits the barrel's upwards elevation. It makes the end of the gun extend far down in the hull, where it's blocked by the motors and the tubing from the smoker; a side-mounted or smaller servo would have helped with this and is another factor to consider when designing this stuff. I discovered this after putting the assembled turret back on the tank. The turret had worked fine when tested before assembly, but assembled on the hull, clicked on upward elevation before the motor shut off. To fix this, I did 2 things-- glued the clutch gear so it wasn't a clutch (bad for gearboxes without a limit switch) and bent the limit switch lever just a little bit so it cut out earlier. I'm not sure which fixed the problem, but I didn't want to have to go in twice to fix the problem.
Like I said, this mod isn't something most people would need. It's a lot of work and you do sacrifice some of the travel range. It's kind of a kludge since the switches don't talk to the software, so the elevation sound continues after the motors have stopped. I think servo control would be the better solution since they're usually smaller than the gearboxes, operate at tighter tolerances (so no slop) and the endpoint limits are usually programmable and fine-tuneable. The speed should be too, but it all depends on the programming of the controlling hardware. (Kinda takes the fun out of tinkering for the old-schoolers.)
This lame video shows that the motor turns off (the whirring sound), but the sound sample keeps playing until the stick is centered. With the louder engine sample sounds playing it's not really very noticeable.
LIGHTING THE ELEFANT (AND MORE)
07/19/13- The ElMod board has built-in lights control and it seems like a shame not to use it. On the Elefant model there's only one obvious place to put lighting: At the rear left there's a box-ish mounting with two framed squares-- no doubt a convoy light. Neither of the two surviving Ferdinand/Elefants have intact lighting there, and I haven't run across wartime photos showing that detail to verify what they're supposed to look like, one way or another. There's also the unresolved issue of what color convoy lights were. For the most part, modelers have traditionally rendered them as blue but there's evidence that they might have been green. No matter. Nobody's making this stuff for the 3rd Reich so there's no penalty for getting things wrong, despite the rivetcounter nazis!
Openings for the convoy lights were drilled out at the corners with a small drill bit, but even with a needle file it was difficult to create a smooth-walled square channel of the length needed. Coincidentally, a brass square tube fit the hole perfectly with a little help from a soldering pencil. The tubes do narrow the aperture slightly, but it's more than adequate for the LEDs to shine through. I initially thought I was going to need to use tiny SMD LEDS, but since the channels go all the way through to the chassis, I used regular 3mm LEDs, 1 for each channel in green. They snuggled right up against the ends of the tubes with some heat-shrink tubing.
07/29/13- I found a rear-end photo of a Ferdinand and deduced that the actual light was a more conventional design. I added a 3rd LED in the center to light the 2 inner lights, which was a quickie salvage of the work I'd already done. Looks like a squared electrical conduit runs from below to the light.
I've seen plenty of photos of the Ferdinand with headlights, but few (if any) of the Elefant with headlights. It's hard to tell with some of those old grainy B&W wartime photos. At any rate, since I'm putting in lighting for the convoy lights, I might as well put in headlights too, historical accuracy be damned. Why? Because it's more funner.
I spent waaay too much for the Schumo nickel-silver King Tiger lights but they included all the parts needed to mount them similar to the way I'd seen in some pictures. I didn't realize (or care) that they were nickel-silver when I ordered them, but once I tried to drill out the conduit at the rear, I realized how much harder nickel-silver is than the usual materials. I was able to do one conduit (almost), but broke a thin carbide drill bit or two doing it and had to keep dunking the part in water to cool it down. I settled for routing the wiring the way they had intended; through the bottom plate and back to the wall. I drilled wiring and mounting holes in the back wall; fortunately, the wall was thinner and easier to drill.
I used SMD LEDs and very thin single strand wiring because there wasn't much space inside the lights and the wiring had to be threaded through narrow channels. Soldering the wires to the SMD LEDs is dicey business with only one good eye, but that seemed like the easy part-- I had exactly 2 SMD LEDs left, and I was paranoid that they would jump into my carpet, or get flung across the room while extracting them from their packaging. I've learned never to handle stuff like that with tweezers and use adhesive putty instead. They were given a mix of red & yellow Tamiya clear tint to make them look more like incandescents.
I wanted to cut away the Zimmerit behind them but was too impatient; there wasn't a lot of room to work there either. Looking at this pic, I'm more bothered by the soft detail on the fender-- there should be hinges there and a spring!
(07/29/13- So of course, I added the springs. The hinge detail was too low profile to get a tiny drill bit in there and I didn't want to cut off the fenders and have to make working hinges, so I settled for scoring the separation line more deeply. Larger periscopes were added to the driver's hatch interior side since the original tiny nubs looked like they would have been unusable by a hypothetical 1/16th scale driver. Plexiglass blocks were inserted in the exterior structure to show glare if light happened to strike them.)
The inside detail of the Hooben kit's hatches is pretty minimal, but it's hard to find photos that show otherwise. It really only matters if you intend to show open hatches; it's kind of odd to show microscopically detailed exterior weathering with an open hatch that's perfectly clean and undetailed.
The infamous IMAI Elefant kit of the '70s helps fill in the information gap (via its assembly instructions-- I don't own one!) since it has details not present in the Hooben kit (like the fender springs). The IMAI kit looks to have the split sliding panels for the gun sight at the top left front-- on the Hooben, the detail is sculpted on the turret as a single panel. Also, the IMAI cupola hatch has a hinged panel for its periscope binoculars-- on the Hooben, the panel and hinge are sculpted on the outside, but not on the inside. (It's fun trying to figure out how this stuff might have worked in 1:1 scale.)
Cutting out and articulating the crescent sliding gun sight panels is more work than I'm up for at present, but it was easy to scribe the lines that suggest it's actually 2 panels. (It seems that there should be wear marks on the surfaces that the panels slide over and where the retainers make contact, unless they're supported on interior tracks...?) Adding the hinged periscope panel to the commander's hatch was another fairly easy job. (Ignore that the figure's wearing an SS tunic and a Wehrmacht cap... Ooops!)
I've sunk to doing the nerdish level of working detail now... Haven't decided what to put in there yet. The useful tool (an allen wrench) is behind, stuck to a magnet. The lid was hard to remove since it was thoroughly glued and the back had to be cut out to give the box some depth. The hardest part was attaching the cut up dollhouse hinges since they wouldn't stay put with glues or even epoxy putty. Finally, I drilled some holes and stuck in some pins along with the superglue (left some extra length on the pins to make sure that they stayed put). That seems to work. I couldn't fabricate the tiny top latches in brass so I took the easy way out and used wire which inserts into hoops. It ain't pretty but it's sufficient to hold the lid closed.
Speaking of Zimmerit: Although I didn't notice in person, some of my photos alerted me of the defect in Zimmerit application in one section on the front port side. In the photos, a few areas pooch out with air underneath and if you push them, they're soft and flatten out, feeling like a thick layer of paint that hasn't bonded with the surface. I've never applied the ATAK Zimmerit so I don't know what it's like in its virgin form, but it must be awfully thin. I think most hobbyists use superglue to bond it, and if that was used on this kit, it seems to work well overall. However, I imagine that it's difficult to make sure that it remains consistently "wet" over the entire surface if you're doing coverage of a large area. If coverage isn't complete and spots have dried underneath before bonding with the Zimmerit, you may end up with islands that aren't bonded, that pooch out as the thin resin panels relax or react to temperature or humidity. That's my theory, at least. If I were to try this, I might use contact cement since it's designed to remain wet until both coated surfaces are pressed together. Whatever.
The fix for this is to slit open the areas with an Exacto knife and use a hypodermic needle to inject superglue underneath. Have a lint-free towel ready to sop up the excess that oozes out when you push the pooched-out area down. Have a container of Acetone ready to run through your hypo when you're done so you can re-use it.
I was able to get most of the pooched-out areas to flatten, but there are still a few lumps here & there where it feels like there's something underneath or the resin panel has some variations in thickness and hardness. Good 'nuff though.
The ATAK Zimmerit looks great-- very professional and to-scale, reminiscent of the stuff that the model companies sculpt onto their kits. One might argue that it's too perfect, and lacks the random funkiness of hand-applied Zimmerit. That may be true, but it sure seems like it would be a heckuva lot faster and easier to apply! Keeping your sanity matters.
MACHINE GUN ACTION
08/01/13- Since the Turnigy 9x receiver has plenty of spare channels, I felt obliged to use 'em. The ElMod board does a lot in 4 channels, and with the better control that the 9x transmitter gives, I haven't felt the need to run it in 5 or 6 channel mode (I'm really not interested in playing "The Eyes of Texas" on command.) So it's sort of a challenge to come up with something interesting to do with all this wonderful potential.
I once saw an animatronics machine gun that someone had added to their RC tank (it may have been the same guy who came up with the machine gun switching mod?). I've always thought it was a really cool mod, but one that required a spare channel that I didn't have... until now. For the Elefant, the mod was especially attractive since the laser machine gun isn't mounted on a moveable turret, so it just shoots where the tank is pointed. Wouldn't it be cool to manipulate it separately from the cannon? I thought so.
It was a very simple job as you can see. The sculpted machine gun ball mount was drilled out and the machine gun was re-mounted with a leather bib glued to it and the interior surface, holding it in place and providing full movement (it was easier than making rigid axes). Everything else happened to be right there-- the support for mounting the servo and the receiver, so no long wire runs were necessary. I used a very thin arm of piano wire since the servo was pushing the equivalent of a feather. (The volume pot was move a little further back than shown in the pic.)
The servo was connected to channel 5. On the transmitter, channel 5 was configured to be controlled by the left-hand upper pot (P3- pitch trim) with a slow up/down setting of 2, to smooth out twitchiness that you'd get from direct control of the servo. That's pretty much all there is to it. Except... I've got to gush some more about the Turnigy 9x/er9x firmware: It's amazing, and I doubt I'll ever use or figure out a fraction of its potential. It seems like it can do pretty much whatever you can think of, and you can spend hours tinkering and tweaking! Life is good.
Even though extra channels let you control a bunch of things independently and simultaneously, your brain and control appendages have to be able to manage it effectively. I need at least one more brain and probably another couple pair of limbs to take full advantage of the remaining 3 channels. But that's just me.
Having extra channels lets you "splurge" and not worry about using them for less-essential amenities. In this case, it's a servo-controlled volume control on channel 6, mapped to P1 (hover pitch). All my other tanks have the volume control mounted on the exterior, disguised under a surface detail that can be twisted. I've found that to be very useful since the tanks run indoors and outdoors and firing them up indoors can sometimes be a jarring experience; a too-loud tank is something that I want to remedy quickly. Unfortunately, the Elefant doesn't have any surface detail that readily lends itself to this solution, so the pot was placed under a hatch. It took a little longer to access it and since it was recessed, it was a little harder to operate. The solution: Use one of the transmitter's control pots to turn the knob remotely.
ElMod sells an optional electronic version that plugs into their unit (I think), but it's an easy thing to do the old-fashioned way, with a servo and volume pot. For greater rotational range, I joined the shafts and bodies together, independently, instead of using a servo linkage. It's the sort of self-contained gizmo that you can just throw into the hull without mounting, like a Pico switch. The servo's travel is about 180 degrees without increasing its limits in the transmitter so this doesn't cover the pot's entire rotational range. However, it covers its entire useful range: I can't imagine a situation where I would need the pot to cut the sound level to zero.
RC TANK INDEX