NTN – 600 Hull Rigging – Part VI

When I went down to the workshop last night, the servo bracket was stable and the epoxy was cured. So, I shot a little video of the moving mechanism and converted into a GIF animation.

Rudder Servo

Testing the rudder mechanics with a rudder servo.

After that, I started looking for parts for an antenna mount. There are nice looking aluminum ones on the net, but I don’t want to wait another two to three weeks for the package. I ended up with a couple of metal parts in my hand, which looked like suitable candidates.

Options

The parts I found in my nuts and bolts box. From left to right: needle of a ball pump, unknown part with a hole and outer threads, LED panel mount, unknown part with internal threads, a small drill chuck, and a SMA connector ferrule in the middle.

The SMA ferrule fits perfectly in the LED panel mount, yet it looks a bit short to my eyes.

Choice

The LED panel mount with a SMA connector ferrule tightly fit in it looks ok.

I also tried a 4 mm brass tube (I should have simply used my calipers), and it fit snugly in the LED panel mount, as well. I also added an o-ring below its flange for a softer hold on the hull and waterproofing.

Brass Tube

4 mm brass tube also fits nicely in the LED panel mount.

The plastic antenna tube fits nicely in both of the options (ferrule and brass tube), So I quickly joined them all together, and insulated the whole thing using heat-shrink tubing (red at the tip).

Final RX

After 2 heatshrunk tubes, it looks much better.

I will install it on the hull tonight, along with the cooling water inlet from the rudder.

NTN – 600 Hull Rigging – Part V

Yesterday, I finished the rudder mechanics. First, I coated the wooden base of the bracket with 20-minute epoxy (I didn’t get a new one, I used to cloudy old one, hoping it would not fail me) for more strength, and waterproofing. Then, I countersunk the holes for the brass screws on the aluminum part, before mounting it on the wooden base. Finally, I stabilized the servo on the bracket with the 3 mm stainless steel bolts. It was time to epoxy the whole thing at the bottom of the hull, close to the transom.

Bracket Epoxy Mount

Plywood base was epoxied, and the whole assembly was clamped to hull bottom.

I epoxied the bottom of the wooden base, placed in where I wanted to be, and clamped it with a c-clamp for curing. The bottom of the servo rests against the port side of the hull. When the epoxy was cured enough to hold its place, I attached the push-rod to the servo horn and the rudder links. As I was going through the RC boat parts in a big plastic box in my shop, I found the original push-rod link for the rudder. Since the one I used as a substitute was smaller and loose in its hole in the rudder tiller, I replaced it with the original for smoother motion.

Pushrod Links

Replaced the tiny link with the bigger original one that I recently found among the other RC parts.

Big Link

Pushrod link on rudder tiller changed with the original one for smoother action.

I attached the plastic ring link to the rod by heating the push-rod and burying it in the link from a very short push-rod (almost the same size, but the threads did not match), and lightly hammering it afterwards. Feels strong, but I don’t know how long it will last.

Servo Pushrod Close

Epoxy almost cured, plastic pushrod link attached to servo horn using a 3 mm bolt, a nut, and two washers.

Rudder System Top

Completed rudder mechanics.

The epoxy is probably cured by now. I’ll work on the antenna mount and the water inlet from the rudder tonight, as I still don’t have a motor or an ESC to work on.

NTN – 600 Hull Rigging – Part IV

Today, I made a bracket for the rudder servo. I have seen (needlessly complicated) examples using L-profiles around the web, and since I have lots of aluminum L-profiles laying around, I decided to make use of one. I cut a 53 mm wide piece from a 20 x 20 mm L-profile, then I sawed and milled the gap for the servo. Finally I drilled 3 mm holes for the servo bolts, and 2.5 mm holes for the 6 x 2.2 mm brass screws for mounting the aluminum part to the plywood base.

Servo Bracket Bare

Cut and milled from 2 mm thick 20 x 20 mm L-profile, the width is 53 mm.

Servo Rubbers

Rubber gaskets that came with the servo.

I used 9 mm thick birch plywood for the base of the bracket (which can be easily epoxied to the hull later). Since the hull is a deep-V mono, I would have to shape the wooden base to match the angle of the V, in order to have a square servo position. So, I placed the incomplete part in the hull and measured the clearance from the edge of the base facing the keel (and in parallel with it) to the bottom of the V-hull, as I held its base parallel to the imaginary and flat water line.

Servo Bracket Wood

Aluminum bracket mounted on 9 mm (birch) plywood with 2.2 mm brass screws. I may have to countersink their holes.

The clearance was about 3 mm. So, I marked it on the wood, put the part in the vise, and sawed off the excess with my GOP-250 CE. When I put the part in the vise, the edge facing the keel was flush with (and in parallel with) the jaw of the vise, and the starboard side edge was sticking out 3 mm. I tested the part in the hull, and the angle was not enough. I sawed and sanded off 2 more millimeters this way. The result is below.

Wood Chopped Cross

Servo bracket after the plywood parts is cut to angle.

Wood Chopped Side

The cut is more obvious from the right.

This time it looked good and square in its place. As I was doing these tests, I held the part in place using a drill bit and a neodymium magnet.

Angle Testing Cross

Testing the angle of the cut on wood. The drill bit holds the bracket in place thanks to a neodymium magnet under the hull.

The 20-minute epoxy I got in hand looks cloudy (moisture?), so I don’t really trust it. I’ll get some more tomorrow, and use it to stabilize and waterproof the base of the servo bracket.

NTN – 600 Hull Rigging – Part III

Today, I drilled the hole (from 2 mm to 4 mm, 6 mm, and finally 7 mm) for the push-rod bellow (or boot), and epoxied the bellow into the hole. The new push-rod is an old bike wheel spoke, heated and buried into a plastic push-rod link. I later on sanded the crappy old rod for a shinier look. Here are some pictures from the workbench…

Rudder Complete

Final look of the rudder mechanics, with pushrod and rubber bellow.

Servo Pushrod

Pushrod and bellow from top.

Servo Pushrod Hull

Servo in the hull with pushrod.

Cooling Inlet

Bellow from below, with the cooling hose attached to rudder inlet.

I’ll probably start building plywood stands / brackets for servo and ESC tomorrow.

NTN – 600 Hull Rigging – Part II

I got done with cleaning and rearranging the workshop yesterday, and finally started working on the hull. As you can see in the picture below, the 7.4 V 5800 mAh Li-Po (I will switch to a 3S if necessary) pack fits nicely in the hull, and the shaft can easily run under the pack.

Hull Open Top

The battery fits nicely on the sides of the V. There’s enough clearance for the shaft to easily run under it.

I started with marking and drilling the holes for the rudder bracket. I drilled the center hole first, then marked the holes for the holes in the mounting ears. Finally, I drilled the holes with a 2 mm bit first, then enlarged them using a 3 mm bit. I also roughly sanded the area where the bracket would be mounted,  so the liquid gasket would stick better.

Transom Holes

The holes are ok, I know they look crooked from this angle.

After applying a generous amount of liquid gasket to the mounting bracket, I installed it on the transom extension using the 3 bolts. The bracket was stabilized square with the top of the hull (the only part of which flatness I was sure of), since the transom extension is not all that square, thanks to fiber hull warping.

Bracket Mounted

Bracket mounted on transom extension using liquid gasket.

Bracket Square

Rudder bracket is square to the top of the hull.

Having big hands can be a problem when dealing with the narrow parts of the hull, such as the transom extension. I used pliers to hold the nuts in place as I tightened the bolts with a hex key. I could only use one washer, since the the other two could not fit where they were supposed to be.

Transom Inside

3 bolts to hold the bracket, only one with a washer due to narrow space.

After I was sure that the bracket was mounted firm and square, I installed the rudder for visual conformation. I think it looks a bit slanted towards the bow in the second picture below, I may have to raise it a bit from the center hole of the bracket.

Mounted Bracket Rudder

Transom from the bottom.

Rudder Bracket Side

Rudder attached to bracket on transom.

I attached a 2mm push-rod (salvaged from a weird cloth hanger) to the rudder to find the entrance hole for it. I stabilized the push-rod link nut with thread locker. I’ll probably do the servo / rod / rudder installation part today or tomorrow, since I still have no motor or ESC to install. I may also test the water inlet from the rudder.

Servo Pushrod

Push-rod attached to rudder.

We got a 1.5 day break starting from Tuesday Noon. More work in the workshop…

NTN – 600 Hull Rigging – Part I

I’m taking a break from the CNC Project, since I keep losing steps, and I’ll probably need to open, adjust, and clean the whole thing. I may even have to mill the gibs.

So, I started working on the NTN – 600 fiberglass (deep v) hull I got a while ago. I ordered a water cooled brushless motor (Turnigy AquaStar 2842-2800KV) and an ESC (HobbyKing 50A Boat ESC 4A UBEC), an adjustable stinger drive (55 mm – Black), an aluminum water cooling outlet, and a pair of turn fins last week from Hobby King. I’m waiting for their arrival. I already have various shafts and props, a good quality aluminum rudder with water intake, some servos, and some LiPo batteries (7.4V – 5800 mAh) in stock.

Props

CNC props (38 mm and 35 mm) that require no sharpening or balancing. Their shaft holes (3 mm and 3.25 mm) from left to right) need to be enlarged to 4 mm.

Prop and Shaft 2

Brass prop attached to 4 mm flexshaft (in brass grease tube and Teflon bushing), with a drive dog and a prop nut.

NTN-600 Front

675 mm hull.

NTN-600 Back

Long transom extension.

As the two-part rudder bracket was holding the rudder too far from the transom, I decided to replace one of the parts (the mounting part) with my own (leaving the hinge part as it was). The width of the new part would be 30 mm as in the original, but since the rudder mounting extension on the hull’s transom is 17 mm thick, the mounting section of the part would have to be 7 mm shorter than the original piece. I would still have to make the extruded and slotted part 24 mm long, to be able to hold the rudder hinge part firmly.

I designed the part on SkecthUp since I still need to learn a real solid modeling CAD program.

SkethUp Sketch

The result was a bit different than what I actually designed.

Here’s what I managed to do on FreeCAD. It keeps crashing and messing up on me :|

FreeCAD Weak

A mixture of what I had in mind (no center hole in the slot) and what I ended up with (no double holes in mounting ears).

I had a 15 mm thick machinable aluminum piece laying around, and it took me more than half an hour to cut out a suitable piece with a jigsaw. After that, I squared the piece to 30 x 24 x 15 mm (width x height x depth) on the mill.

Raw AL

Cutting this piece took me about 30 minutes with a jigsaw.

Square AL

Block Squared to 30x24x15 mm

Later on, I milled off the sides of the slot section (9 mm wide, for a 3 mm slot with 3 mm walls) down to 5 mm using 8 mm and 6 mm end-mills.

Pockets

Milling the pockets with an 8 mm end mill.

Rudder & Bracket

Rudder with the incomplete bracket.

I decided to drill the holes while the work piece was still thick and solid. However, I messed up on the mounting holes, since I was supposed to reduce the height of the mounting section of the part (5 mm thick) from 24 mm to 17mm first, and then decide on the location of the holes.

Wrong Holes

I drilled the wholes prematurely (too wide apart), before milling off the 7mm from the mounting sides.

4 mm Reaming

Reamed the wrong holes.

Eventually, I had to mill off 10 mm instead of the actual 7 mm to get rid of the useless 2 holes (actually the larger countersunk area), leaving me with a 14mm tall area and only 2 holes for mounting. You can see the old and the incomplete new bracket in the picture below.

Long & Short Brackets

Still incomplete bracket (middle) with the original bracket parts (right).

Later on, I drilled a third hole in the middle of the slot for a better hold, and enlarged that part of the slot to a 5.5 mm hole, to be able to fit the 5.2 mm bolt head there, also countersunk the bottom hole to avoid blocking the slot.

I also messed up the holes on one side of the slot. I drilled 2 x 2.5 mm holes on the side of the 9mm wide extruded part which would later form the slot, so I could cut threads in them with a M3 x 0.5 mm tap. After milling off the 3 mm slot in the middle of the 9 mm extrusion, I tapped the threads for the holes on one side, and enlarged the holes on the other side to 4 mm instead of the actual 3mm. Since threaded holes were ok, the hinge piece has 3 mm holes, and the countersunk holes hold the 3 mm bolts fine, that was no big deal, really.

Countersunk

Still incomplete, but the holes were countersunk. Hole in the middle needed a cylindrical guide in the slot.

Holes Back

Three 3 mm mounting holes seen from the back.

I gave the part its final shape with my belt sander and files, and polished it with my Dremel Tool to a mirror finish. The chuck came off twice due to vibration (I’m no machinist, and I don’t make parts everyday) and marred some small regions of the part. I think I need less fluted end-mills for machining aluminum. I don’t really mind, I’ll do better next time.

Bracket Cross Bracket Front Bracket Bolts Top

Here’s a table of the weights of the parts and some notes about those parts.

ITEM WEIGHT  NOTES
Stinger Drive 35 Adjustable and nice looking.
Servo 60 Add 45g weight to balance with ESC on the starboard side.
Battery 280 To be centered parallel to transom on adjustable slide.
Motor + Coupler 150 Motor mount centered in hatch opening.
Shaft + Prop + Tube 53 These easily go below battery.
Cables + Hoses +Pipes 35 Silicone hoses, brass tubes, aluminum tubes.
ESC 105 50A 4A UBEC
Rudder 57 New bracket milled to get the blade 20 mm closer to the prop.
Trim Tabs 30 I need to make some from L profiles and 1 mm stainless steel sheets.
Turn Fins 20 They are also on their way, 40 mm length is probably too small.
825 gr

I also finally bought myself a workbench, a tool cabin, and a tool shelf rack. I’ve been busy with rearranging the stuff and cleaning up. I’ll finish it all today and start working on the boat on the new workbench soon.

Bench Clean

Final status of the workbench side.

Proxxon MF 70 and LinuxCNC – Part VIII

This last Thursday, I went ahead and got a second hand HP DC7700 (with KB and mouse, but without a monitor) for about $100, along with a cheap and suitable external video card (since LinuxCNC is supposed to fail with on-board video cards) with VGA, DVI, and HDMI outputs.

I installed LinuxCNC from a flash drive (the other one was too old to boot from USB!), and tested the on-board and the external video cards for latency. They were pretty similar, around 15000ns. But since my LCD monitor only has VGA, and I don’t want to take no chances, I use the external one. I use soft OpenGL drivers, and all is well. Good to use a 2GHz Core II Duo with 2GB RAM. No more flickering AXIS and late opening windows! I did a test cut, and all was fine.

True Holes

Aimed for 6 mm holes, not bad at all.

The next day Pamir came, and we worked on some cuts, and a dust collector system from noon ’til night. I also gave up on the rig for the lighting and the dust collector hose, and resorted to Redneck methods (for now). The light fit nicely in one of the spare holes of the Y-axis stepper motor. I also changed the Z-axis motor for it was heating up extremely (full coil bipolar driving a unipolar motor is not worth it), and I was losing steps. Luckily I found the exact same motor I used for the x and y-axes at my friend’s shop.

Pamir’s CAD/CAM work was well done. Yet, as the program was about to cut around the part, (I dunno what caused it, I remember accidentally hitting the keyboard with my elbow) it started from a wrong spot and hit the clamps, breaking the bit. This was red acrylic, clear acrylic I got melts like a candle, and I still don’t know what makes them different. Our second cut was a failure, because the bit was not a perfect cylinder, but a slightly tapered one. So the holes were ok (6 mm) at one side, but not ok on the other (about 5.5 mm). Our final cut was on red acrylic again. This time all was good until the program started cutting the frame, the final line went through the holes, and the last three rows were offset irregularly, due to some reason still unknown to us.

Fail Cuts

3 similar cuts, 3 different failures.

And here’s a video of the failing cuts on acrylic. 6 x 1 mm + 1 x 0.5mm scalar feeds on 2 x 3 mm red acrylic stacked together.

As the cuts took about an hour each, we worked on a dust collector system in the mean time. We had to improvise as always. I had some 1.5 mm PVC hose, and managed to make an adapter from a garlic crusher (easy screw on lid for portability) and a syringe. All fit snuggly, and I sealed them with black hot glue. An old vacuum cleaner was also laying around to be utilized for this purpose. We mounted the hose around the spindle with cable ties and sticky cable tie holders.

Hose Adapter

Hose adapter improvised from a garlic crusher and a syringe.

Hose Adapter

Final form of the hose adapter.

The dust collector works nicely, yet I still need to build a good jig for it, and add a mouth part for better and wider angle dust collection.

Today, I worked on the sound absorbing part of the box I was planning for.  Since most of the sound was coming from the hole (for the spindle) on the top of the box, I decided to cover it with a sound absorbing box. I made the box out of 4 mm plywood, using PVA glue and small nails. Later on, I covered the inside with sound absorbing sponge (foam?).  Finally I covered the side walls and the ceiling of the box with more sponge. The result is much better than before, but not as good as I expected it to be. I know nothing about this subject anyway. The distance is probably too short.

Clamp Box

Box glued with PVA.

Glue Box

Box after gluing and clamping. The Bosch GOP 250 CE did it all.

Sound Absorber

Box filled with sound absorbing sponges for the top of the main box, where the spindle motor screams through.

Box Mounted

Sound absorbing box mounted on the main box.

Sound Sponge

Sound absorbing sponges mounted on some of the the inside walls with 3M Super 77.

I still need to install some sort of clamp to hold the acrylic door closed (worst sound barrier). I’ll probably work on that tomorrow. Here’s what the setup looks like right now.

Setup

Latest look of the setup.

 

Proxxon MF 70 and LinuxCNC – Part VII

I added a power switch and an emergency button on the electric box, and wrapped the motor cables with plastic cable spirals (or whatever they are called) last week.

Electric Box

Electric box with power switch and emergency button.

Inside of the electric box at its final  stage.

Inside of the electric box at its final stage.

This last Friday and Saturday Pamir and I worked on some CAD/CAM designs for a future project, and  tried to cut what we actually wanted to cut, using OpenSCAD, OpenSCAM, DXF2GCODE, and PyCAM. We also started working on a closed box for the system, which I worked on further on Sunday. Here’s his final work, which we didn’t get to cut since the new HDD died right before, and he fixed his design after he got home.

Part designed by Pamir on OpenSCAD, QCAD, and DXF2GCODE. Shot on my actual PC with LinuxCNC installed along with other OS.

Part designed by Pamir on OpenSCAD, QCAD, and DXF2GCODE. Shot on my actual PC with LinuxCNC installed along with other OS.

AXIS

From another angle, the simulation continues. The cut is for two pieces stacked together, to be joined (we need to design and cut dovetails) side by side later.

We still need to figure out the details of Tool Compensation in the various open source software we use, to be able to decently cut what we actually want to cut. Modifying the tool table in AXIS helped a lot, by the way.

DXF2GCODE

Pamir working with DXF2GCODE. My ever handy Bosch GOP 250 CE Professional is laying in the back.

Here’s a cut we did on a small piece of 3mm red acrylic before learning about tool compensation. The circles were 1.5 mm larger than what we (0.75mm radius tool compensation was needed) actually wanted, so the material between them is also 1.5 mm less than the 2 mm we had aimed for.

When the CNC PC’s HDD failed, we started working on a box for the system. A small coffee table looked suitable for our needs, and it saved us from building the box from scratch. It wasn’t tall enough to cover the Z-axis assembly, though.

Box Front

The first stages of the box. 3mm acrylic lid with brass hinges. The back is made of 4mm plywood screwed on the sides and bottom.

Box Rear

The back of the box with the cable holes cut out. One will also be used for the dust-collector hose.

I was going to spray paint the plywood parts to black, then I got lazy, and decided black duct tape (living in Alabama for 6.5 years finally pays off) would be tougher against wear and easier to apply. I later on added a small back door for easy access to the power switch and the speed pot of the MF 70.

HalfDuct

Half duct taped box with more plywood added.

I cut a small piece of plywood for the back door, and used duct tape as its hinge, as I duct taped the rest of the box including the edges for insulation. I also added rubber hose windows insulators to the frame for the acrylic door.

Box Progress

Box after the duct tape attack. With the lid limiter attached.

Box Back

Same from the back.

Limiter Close

Limiter screwed  to a short piece of  wooden cylinder of which sanded flat side is glued to the lid.

This week I’ll be working on a lighting system and fasteners for the acrylic door to hold it tight against the insulators. I have a 5 day break coming this Friday, we’ll be working hard in the workshop ;)

Proxxon MF 70 and LinuxCNC – Part VI

I’ve been playing around with FreeCAD (learning with video tutorials on YouTube), PyCAM, LinuxCNC Image to Gcode, OpenSCAM,  etc. We were planning on routing a 2D shape, and I thought Image to G Code would be helpful. I was wrong. It’s actually for milling contours (constant stepover?), but not for routing a stencil.

img2gcode

Image to G Code on LinuxCNC 2.6

Since it was taking ages to process an image on the CNC PC (800 MHz P III + 512 MB RAM + Trident Blade 3D 8MB), I did all this on my actual desktop PC (i5 2.8GHz overclocked to 3.8 GHz + 1600 Ghz 8 GB RAM + NVIDIA GTX 660 Ti) running LinuxCNC 2.6 from a USB disk. So I could try the image with different parameters quickly. Well, Max Baser Jittter was so much better (around 5000 ns) than it is on the CNC PC (around 50000 ns), while everything was running so smoothly as opposed to the flickering windows and late responses of the CNC PC. So, I’m not really sure about those “a faster PC doesn’t necessarily mean you’ll get faster RTAI response” comments.

OpenSCAM in action. The result is not really what I need.

OpenSCAM in action. The result is not really what I need.

I’m thinking about installing LinuxCNC 2.6 on a different partition on my actual desktop PC just for the kicks.

Still no real work on the CNC. I’ll either shallow carve this shape above in brass for a test, or create a DXF to route it in 2D.