Of course I did the smoke test today. Mounted the motor, connected a 3S battery and hit the throttle.
I ran the Emax 2204 motor for 10 secs before I could fry an egg on it. Hmmm, maybe I should have looked beyond ”it fits”?
An 8*4 prop is not quite the same as a 5*4 drone bit of plastic. No problem, what else do we have in the BigBox of Stuff? Ah, look at that, sort of ancient history here. A HET-RC Typhoon Micro 6/3D. More or less the first motor I got just before the foamy craze started.
Weight : 35 grams
KV : 1460
Max Current : 12 A
Power Nominal : 90 Watts
Recommended Props: APC E 9 X 4.7
The 9*4.7 is what is relevant here. Ok, lets see what E-Calc says:
50 Watt on a 2S Lipo with almost 4.8 m/s vertical performance at 370 gr RTF, and a 5 mins motor run on a 450 mAh battery. Ok, no problem. Lets go!
Getting a bit nostalgic there while looking for the plane with that Typhoon. At the top is my Blaster in good ol’ N-Ireland. It had a variety of radio gear in it, this is I think the Spektrum-Mpx version.
Can’t resist looking back: Yellow Road. With the right northerly you could hang for an hour in the breeze, untill your fingers turned blue. Further back was a similar bowl facing south-ish. That was the better spot. My first flight with the Birdy lasted 30 mins or so.
And I found it! First ever CNC-ed plane. Not a real serious design, just made it, because I could. This was powered by that HET-RC Typhoon. Did fly well enough in the backyard!
While waiting for some parts, I decided to have a go at another project that’s been waiting to be started. I still have my original Blaster 1, RTF is approx 350 gram, and while it was fun flying it from a slope, it always a struggle to throw it higher then 35 meters on a good day. In the meantime I have something black that weighs 220 gram, and does go a lot higher!
That says much about my technique! Anyways, I’m electrifying it. There’s some quad engines lying around, some small Lipo cells that appear to be good enough for a few more flights. All that’s left to do is some radical surgery.
The first cut is the hardest, after that it’s a matter of making enough room for some servo’s and Rx, and somehow getting the CG sort at the right location. As always, I’m using the TLAR technique to get the dimensions right. There’s still a bunch of D47’s from Indoor days gone by, I’m sure I can find a receiver that will do the job, so I’m almost finished, just a matter of assembling a few parts.
I could not find the right size aluminum in my stash, which is a good excuse for doing a quick printed version of a prophub. I have a few 8*4 foldable prop blades, not really fit for this purpose, but I’m not using them for anything else. The first straight center part looks good enough, but then I wanted to have the blades tighter to the fuse. Some headscratchings later I have a fully printed foldable propeller including spinner part. Printing is is just rough and quick proof of concept, not the final one yet.
In the meantime I started preparing the fuse. Those were the days that Vladimir used kevlar for the front end. Real fun when sanding!
Will a printed part be strong enough to hold on to those blades? The answer to the question will have to wait till I assemble the battery. If not, I can always fall back on a proper 5 euro part.
That looks about right! so far 19 grams more then when I started, I need to add some glass around the front end a something to hold the servo’s. Also a small lid to stop stuff escaping. CG used to be at 78.7 mm, 3 mm to go!
Getting the cabane structure built was always going to be one of the challenging bits of the build. So, off we go. With the aid of my trusty 0.01 degree level, I’ve now lined up the fuse and a flat plate that will serve as the build-reference surface. As we speak, I just finished setting up the main center spar. Last week I bought myself a bandsaw from the interwebs. It was advertised as B-Ware, but since the price was right, I figured I probably could fix whatever minor ailment it suffered from. As it turned out, the power switch wiring had come lose inside the frame. I could fix that in about 5 mins. From the same lot I got a disc/beltsander. Also B-ware. The table in front of the disc had some boltholes drilled out of alignment. Yup, big job again. Clearing space for the toys took more time then the repair!
Which all leads up to: I need to cut some metal. I have asked for prices for laser-cutting the parts, but likely I can cut them just as well with the bandsaw. For which I of course need to locate the not so standard size saw. Luckily they can be made to order and do not cost the earth. (as in 15 Euro/pce, made in Germany.)
AcrodusterComments Off on TLAR (that looks about right)
That’s the way we used to (and still mostly do) design our stuff. If it looks right, it will fly right. Works since the days of the Wright brothers (and all those that went before and came after them.)
However, fun has to be had! I need to sort of start thinking about the wings in order to build the center section so that I can finish the pcovering of the front section. Given that the plans show a nice 1:1 rib, and given it is often stated to be a ‘Modified M6 profile’, I want to find out what Lou Stolp was thinking.
The short story: Take a picture, turn it into a black and white image, remove all noise, fiddle some more and import the result in Profili. Then say you want to find similar airfoils, job done.
It does not take long to come up with a list. We’ve got a NACA0013, a RAF30, a Goetingen 459 and an Ultimate profile, that all are very similar. Mix NACA0013 with RAF30 and you are 99% close to what Lou made. Running the results through XFoil shows characteristics that are somewhere between the 2.
Before anybody asks: I haven’t got a clue what I am doing, but the curves look pretty! But I did have to go and find mr. Reynolds to figure out what I was looking at. (He says his number is somewhere between 300 and 500k for the speeds I guess this will fly.)
Given that the Ultimate is a known well flying Biplane, and the RAF30 profile is:
The basic symmetrical section ( R.A.F. 30 ) was calculated by the method described in R. & M. 911, using the constants k = 1.08, n = 1.95, B = 0. The aerofoil shape so obtained ends in a sharp angle, and so the last 1 percent of the chord was cut off in order to avoid a thin trailing edge. The form of the aerofoil was also adjusted slightly towards the trailing edge in order to remove a slight reflex curvature. The aerofoil has a maximum thickness of 0.13 of the chord at a distance of one third of the chord from the leading edge, and its shape approximates closely to the symmetrical Gottingen section 459 which was known to possess good aerodynamic characteristics.
Yup, that’s good enough for me..
In my younger days I did not have all this fancy stuff, I drew profiles that looked nice, and all those contraptions flew, so actually, this is just a fun exercise, that probably does not add anything to the overall results. But it looks good when you say you figured out what Lou Stolp did!
However, how we go from a M6 profile to a symmetric one remains a mystery for the moment. Searching some more I find this remark on kitplanes.com: The new airfoil is the RAF 34, which is similar to the M6 airfoil used on the Pitts. Searching for RAF34 on Airfoiltools.com indeed shows a profile similar to an M-6. The original Pitts did not have a symmetrical airfoil! Ok, happy with that. I’ll stick to what I have 😉
I could figure this out if I want to, but at the moment I’m not interested. I want a symmetrical profile, that is what I have, and how they got there is not really important.
AcrodusterComments Off on A day figuring out the center section.
Making the top wing center-section is a bit of a challenge. The drawings are not very clear. I started by making some fake parts, and see what makes sense. There is also no need for the wing-tank, which saves a bit of hardware. Given that the original construction was usually very clever, it pays to make things as the designer intended.
I also spend quite a bit of time digitizing the wing-profile, then trying to find an equivalent that is easy enough to make. Profile is very close to a standard NACA0013, so I will go with that. If I change my mind for a fatter profile, I need to adjust those plates below. A profile that came very close too is the one from an Ultimate. Either way, I’m close.
There are lots of notes on the drawings, several pages of part descriptions, but references are not always consistent, so lots of fun to figure this out!
But I think the method above should work. It’s drawn upside down. The main spar dimensions are 107 * 20 mm (yes folks, that is all!) scales down to 35 * 7 mm ( ’cause I found a stash of 7 mm rectangular pine sticks). Just needed to glue them together, and job’s done.
After this is all done, I can use this as a jig for building the cabane structure. Below is e real one. As you see the front attachment point is slightly higher then the rear one. As in it is a straight line offset from the center of the profile. It could also simply be to make life easy for the builders, because you need to have the attachment points parallel to the main frame top.
Had some issues with the 3D-printer. Sometimes it would tell me the bed was not heating at the required rate and shutdown unceremoniously. Not Good. After the usual suspects were proven to be innocent it was the ntc resistor under the bed. Tiny as it is, it is a vital bit. Why oh why it would work for hours on end when I watch it, but throw a fizzy when I was not, is something only known to it’s designer. Anyway, after 4 years or so, it’s been retired. Since then, no more failed prints. So, to sensibly waste some plastic I quickly drew some parts for testing. Below you see a false former to support the front plating. Later the plating will be supported by the firewall. After my ABS adventures, I am now using ASA from FormFutura. (The light grey stuff). It has all the good habits from ABS, none of the bad. And indeed, shrinkage seems to be absent. This long part just fits the print-bed. With ABS I would not be able to print it without popping loose. (yes, this give me ideas for a future project 😉
And next.. Before i do the front paneling I need the cabane struts on the frame. Before that I need the hardware to mount it, so let’s see what I can cook up. This part is drawn full size on the plans. It is basically a bit of tubing (11mm diameter) with a 1.6mm sheet wrapped around it. This is really all that holds the top wings on the fuselage. For convenience I make mine from 4 mm tube (so I can use M3 bolts) and 0,8 mm sheet. I did some calculations, and if I am not too far of, that little thing can hold over 300 kg before it breaks apart. Actually, the weakest part is the soldered joint. Silver solder (worst case) has 300 N/mm2 tensile strength, the soldered area is 10 mm2. I think I’m safe enough with this.
The temptation is still there to make the tabs per full size, just need to find a method to keep things light enough and easy enough to fabricate. In fact, I don’t need to do the whole fuse, just this cockpit area… The method used at the tail end is way too difficult, mostly because I can’t source plain simple 2 mm strip..
A nice in between job will be the throttle quadrant, and instrument panel.
Anyway, I guess I’ll have to build at a minimum a mockup of the center section, which means I need to think about the wing mounting.
All the images above are downloaded from BiplaneForum.com, I hope they don’t mind, and if they do, I am sure they will let me know 😉
3D printComments Off on When you are determined to get things right..
I was still not happy with my ABS printing efforts, and determined the only way forward is to chop up the large pieces. Mind you, that 9 cylinder engine is definitely a printer killer. If you can print that, you can print anything!
So, chop chop. Yep, there is an app for that. Anyways, results are improving. Also found that thinner layers do not necessarily print better. Makes sense I think, 0.1 mm layer cools down fast and prevents the plastic sticking to the previous layer. Best results are with 0.15 at the moment, and variable layer setting (in Cura), to get the top nice and smooth. Most stress appears to be in the bottom sections, even at 95 Celsius build-plate temperature, but hey, it works. Speeds are still up in the 70 mm/sec, (yup, I candothat!), because of the thin layers, prints still take many hours. (note: I don’t have enough power in the build-plate heater to reliably go over 95C. ) I can do 105C if I keep the shed warm, but not when the temp drops to 15C at night. For now, I have to live with 95C. Higher temps would relieve some of the stress, but alas, can’t have everything. As I said before, since using the magic glue potion on the build plate, I have no longer issues with parts popping loose.
Since the winter appear to be absent in my shed, it is not too cold to make some good progress. Progress the last few days has been getting to grips with Shapr3d. I Mentioned before it is one of these bits of software that will change the way casual users are going to be designing their parts. Anyway, I need something to hold my skins on the frame. For the tails end I did not mind soldering, but preparing those bits is very time consuming. I also have it in my mind that I need to make some progress to get things flying soonish. Anyway, I figured I should try to make some clips to speeds things up.
Now that I have the hang of it, it will take me 10 mins or so to ‘design’ these parts. Since this stuff is relatively small, I do need to keep the limits of the printer in mind. Still printing in 0.1 mm layers, It takes forever, but it looks nice!
Of course the Big Question is: will this survive in real life. I guess only time will tell. There are other, more stronger filaments out there, so I am not too worried at the moment.
This support is for the 6 mm carbon tube. The red tube is a 3 mm cross piece to mount it on. The complete contraption consists of 3 parts. Why so difficult? I told you, it’s all practice!
Daniel (#2 son), is doing some free-lance development work for game-figures. Once you have the design it is a small matter to see if your figure can be printed in Real Life. Since the printer is loaded up with ABS plastic, just press ”go”.
I printed this one flat, with 0.2 mm layers, and not enough thickness). I wanted to try out the effects of adding an aceton slurry to the part. I did not do anything to it, just brushed on abs-slurry. (put some scrap abs in a container with aceton, shake and stir. Add more plastic to make the slurry thicker.)
The result: promising. This is more or less the same idea as the vapor treatment, where you hang the part in a closed container with aceton vapor. Using a brush seems to produce similar effects.