Oi, it’s too heavy!

 DG800  Comments Off on Oi, it’s too heavy!
Apr 112020

Yeah, sort of expected that to happen 😉

With the added mudguard and wheel bearings and other small bits, the Hitec 645MG is struggling to get the gear up. There is a point halfway between up and down, where max force is needed, it’s just how the design of the gear works. In theory I could redesign everything. (Yup, maybe sometime soon). Or I could add another voltage regulator to give it it’s own supply, 6V would work (most of the glider runs on old 5V gear), or solve it for now with slightly more powerful servo.

I found a SAVÖX 1256, which has probably twice the torque of the Hitec, so will strap that in for now.

Of course with the way I wired things up, that’s easier said then done. Since I installed a 9 channel Rx in the plane, and needed an extra channel for the Flashy Lights, I came up with the cunning plan to use a single channel for multiple things.

what a mess…

In short, gear up or down is only 2 positions. Light on/off same thing. So why not do a bit of mixing and send the resulting servo signal to an Arduino-Pro-Mini, and let it drive the LG servo and Lights. I cooked that software a year ago, knowing it was a quick fix for a self imposed problem. (That’s what this hobby is all about, if you don’t have a problem, make one!)

Anyway, long story short, the pulse that the servo needs for up or down is hard coded in the software. No biggy, but it means it’s not very convenient when you change servo’s.

Oh, it’s sooo much fun. With the aid of my little magic ToolkitRC.com M6 I measured the output pulses to the servo. 1000-2000mSec. Perfect, but the servo only rotates 90 degrees and a bit? (By the way, there is an software update for the M6, V1.31 dated 23 March 2020)

Oh dear. Now what. Time for a rethink, and do remember in future that you really want a 180 degree servo!

After some reflection and relaxing in the sun, I remembered that I used to convert servo’s to 180 degrees by adding a few resistors to the feedback pot. That triggered of course the thought ”did I maybe modify that servo?”.

2 * 1.5k resistors make it a 180 degree servo. (you call that soldering?? Sigh.. One gets older and the hands less steady…)

And the answer is yes, I did! Funny enough I posted that same fact on 3rd Dec 2016 on my Facebook account. Oh where would I be without my facebook memory. Scary..

Well that sure was a fun exercise for a locked-down Saturday. So next question is: Am I going to modify the Savox? The answer is below. Same trick, 2 * 1.5kOhm resisters either side of the potentiometer will do the trick. Funny this servo is rated to produce 4096 steps. It never stops to amaze me how you can get that resolution from a cheap 5k pot. But they work, so I’m not counting! by the way, there are no wires in this servo, which should make it more reliable (cheaper). But it does mean some sort of skill is required to remove the pcb!

Luckily there’s enough space for 2 resistors, same 1.5kOhm as on the Hitec. Rotation is now just a tad under 180 degrees. That will do nicely!

And while at it, might as well try to sort out the wiring mess a bit. I’ve got some space at the top of the fuselage, out of sight.

Less clutter, now see if it works!
that’s better…

By the way, that stepdown converter acts as a current limiter as well. Max current the chip will produce is 4 Amps, stall current for the servo is 5 A. Should be safe. And since it is only ever used for up and down, it is not likely to get warm. And if it does, it will disconnect, saving the battery. In theory this is what is supposed to happen. Real life might work differently.

So, that’s enough for Corona-Easter-2020. Left’s hope and pray the world will not have to repeat this ever again!

Christmas is near, so we slow down a bit.

 3D print  Comments Off on Christmas is near, so we slow down a bit.
Dec 152019

This time of year other things need my attention, so I’m a bit slow at the moment. I’m still working on the tail side plating near the stabilizer of the SA750.

However, something as mundane as ‘a bit of plain 2mm iron strip’ proves hard to come by at the moment. Everywhere I look I can get stainless steel, but I need the lowest, softest iron I can get. I need it to make those small pesky tags to attaching the plating. This means drilling the holes at the right size for a 2 mm tap, without braking everything. As these things go, the first side went without a hitch, repeating the other side has wipes out my stash of properly sized drills. I am now waiting for Santa to deliver more.

So, in the meantime, I got a request for some multi cylinder gadgets. No problem, he says optimistically, after all the printer prints nice and smooth, at a reasonable high speed, what could go wrong?

Nothing, just that my standard slicer (the one I cursed before) did not want to know. So again back to Cura. At least we are on talking terms, it’s just a matter of dialing in all these variables.

And so you discover that PLA does shrink. Most parts I have printed never had any issues, but this one was challenging. No matter what I tried, it kept warping and popping loose from the bed. At 140 mm diameter, a 1 % shrinkage is 1.4mm. And it does shrink more then that.

Anyway, he who wastes a lot of plastic, will succeed. One of the ”secrets” I discovered was a special magic substance for coating my super duper sticky bed. (as in, I never got to stick large parts to it). There are stories around about using pva based hairspray and such. Then one dark night the lightbulb went on: I’ve got loads of PVA, it’s called wood-glue. Squirted a bit in a container, added a bit of water, mix and apply. (I’ve seen suggestions for a 10-15% ratio water/glue.) And to my astonishment, my very unwilling parts stick. No more warping, no more popping loose.

Printer doing it’s job.

If that was all there is to the story, I would have been finished long ago. But Murphy had a good time!

  • My extruder stopped at random times, mostly when I was not watching.
  • The feed tube at the top of the filament heater popped out at midnight.
  • The filament escaped from the extruder
  • Of course a blocked nozzle to make it more fun.
  • My printer can print quite speedily, but that proves its (my) downfall.
  • Printed parts started cracking after a day.

The extruder problem was caused by a broke wire at the connector motor. Why now? Why not now? It took a bit of searching to locate that problem.

Feed tube is held by an Festo connector. It just got tired from being moved around continuously. Fixed that.

Filament guide had to be remade, including better alignment of feed rollers and tubes.

spaghetti, but not as like it!

Blocking nozzle, obvious some cr@p in there.

Printing too fast: oops, actually, that is fine for small parts, where shrinkage is nu issue. Bigger parts, contrary to gut feeling, need to be printed much slower, so that the hot plastic does not cover more than 2-3cm in length.

So after all this was overcome, I have now printed multiple parts where the print-times are in excess of 18 hours. As always, the end result is much experience gained, a much better printer, and plans for a much larger volume version are being hatched!

The result is a nicely printed 9 cylinder dummy engine, printed at 0.1mm layer (with variable layer thickness at the top) speed of most sections ca 60 mm/sec. Bed is 45 C, nozzle at 215 C. GCODE file is approx 75Meg.

in the end, all is well 😉

Printing some streamlining bits..

 Acroduster  Comments Off on Printing some streamlining bits..
Aug 102019

Given that it is difficult to find the right size streamline tubing, I better go ahead an make some. With the open-scad script I found I made a handful test pieces. The idea at the moment is to attach the plastic with a layer or 2 of 25 gram glass cloth. That should fill the little grooves in the printing and provide enough strength. Obviously this has to be done after the brazing.

I also tested the deflection in the motor mount once it is attached to the frame. Can’t move it. This is a comforting thought.

not moving..

Streamline tubing

 3D print  Comments Off on Streamline tubing
Aug 072019

I did say undercarriage? A whole new can of worms just opened. I had hoped to be able to use a normal streamline tube as offered by various vendors. However, can’t find anything close enough to what I want.

I could make my own from round tube, just like friend Achim , but that’s a major effort. Many say that deforming a round tube introduces cracks, and therefore failure waiting to happen. Don’t need that.

How about adding some plastic and cover with some glass, a bit of sanding and no one will see? Time for a 3D print job.

A quick search for airfoil scad scripts turn up something useful. This is the first idea. Ideally there should not be a round nose, but given the rather short length of tubing and the ample power of the engine, I think I can forget about the added drag of not having a perfect airfoil shape.

Most streamline tubing has a more rounded TE, we need to fix that!

So this would be glued to the back of a round tube. I’m sure it will look right 😉