Random thoughts on building

[wdyasq]

John,

I would recommend two or three 'temperature sensors on the part. One of the crazy things one can do is mix epoxy, let it 'cook' in the cup for a bit then, apply to the laminate. This will vary the cure time about double/half ... according to how one looks at the equation.

I propose we build the tool, make up some numbers. Remember, these numbers will vary according to resin hardener and other factors. We will then find out the the speed we guessed is over/under/on the number we guessed. I'm still thinking about the equation, time and temperature. The time by temperature will need to half every 10C the temperature rises, or double on the decrease. Once some cure time to guess numbers can be confirmed, an equation can be developed for THAT resin/hardener combination. I suggest each resin/hardener get a designation after a curing curve is established. These would be as simple as WF(WEST Fast);S3M(System3 Medium); MS (MASS Slow) and such.

Because the Arduino is so easy to reconfigure, Each could modify these numbers to fit his/her/its' particular situation and way they work.

Once someone starts to build one of these things, I'll relate more thoughts on how it might be programmed. I think it may be just be 'take temperature'; factor cure time above/below 20C; multiply; add to base number, when number reaches X sound alarm.

Ron

[wdyasq]

Hey Guys,


Ron, very cool about the spar. Looking forward to seeing some info. Do you have any pix of the test piece that you could post up?

John

John,

I'll try to remember to take my camera out of the car and take a picture or three.

NOW! Even more FUN!... I bought something to destroy things .... no, I did not purchase an active politician ......

I just bought a 500kg load cell. That means I'll be able to load something like a wing spar to near 1G w/o any pulleys. Now, with a few pulleys, I should be able to load a spar (half an airplane) to several G just to look at things. Of course one of the laws of Physics is I now will need to build things to resist these forces.

I'll see if I can take the time to draw up a 'plan' to test such a spar. If I get some idea of a realistic constructed box beam and hos strong this 'Pultruded' fiberglass rod is, a real spar can be constructed from commercially produced sections, tested and 'proved' before building an actual wing.

Another interesting thing to build might be a CNC machine using pultruded angles as the gantry corners. This would allow many feet of span and maximum stiffness. As there is usually a charge for orders below a several thousand dollar minimum, it might be worthwhile to purchase the materials to build one 'hoss' of a CNC machine at the same time.

My next project is to find a wiring fault in my abode. I've spent several hours on it already and may have wire that has been broken and is 'open'. It is the only diagnosis that fits ALL of the symptoms. I enjoy electrical problems as much as I like getting coated in epoxy.

I think the CPM (Composite Process Monitor) will be a lot less complicated than one thinks. Epoxy chemistry and cure rates are pretty well established and predictable. We will need to develop a technique and some initial ideas. From that point, a 'new' resin/hardener combination will probably just need a couple of 'time numbers' to establish a trim time. There should be enough latitude in the cure where those numbers will work well. Of course, there will still be the problem of the well informed fool that fails to follow or understand directions.

Waiting to tear stuff up,

Ron

[jeff54il]

Hi Ron,

With your permission I'd like to use your experimental composite spar and load cell to calibrate my calculations - hope this is OK with you?

To that end I modeled a spar according to (what I imagined to be) your dimensions in your post. It looks like this:

I then ran an FEA analyses assuming a simply supported beam of length 120 cm with a 500 N (~50Kg) load at the center. The displacement plot came out like this:

The displacement at the center came out as 9.5 mm (about 3/8"). Here is the stress plot:

The stress in the pultruded rod showed up as 120 N/sq.mm (under the above 50Kg load) and the stress in the plywood web is (as expected) much lower - about 10 N/sq.mm.
Out of curiosity I tried doubling the height of the beam (100mm instead of 50mm). Also as expected the stresses and displacements are greatly reduced: Stress in the pultruded rod is 30 N/sq.mm, in the ply 5 N/sq.mm, and the displacement drops to 1.7 mm.

Be interesting to know if your experimental results correlate to the above.

John, I tried adding a fourth picture to the post but it didn't get accepted. If I remember correctly there was such a limitation (to 3). Could the limit be increased to say 5? (or 500 or 5000 )

[John Nicol]

Hi Jeffrey,

Increased to 50!

John

[wdyasq]

Geeze Jeffery,

I sent the work out to the prototype shop and THAT fool thought the calculations would be easier if he built it like THIS:

My Load Cell is taking a leisurely trip from its' previous home in New Jersey and is on the way to the one man shop in Texas. I'm guessing this is the test, with deflection numbers, you would like done:

The green blocks designate both the support points and the load point. Of course the load will go opposite support. I hope I can get the shop guiez to understand this procedure I'll get the numbers.

Another test I will do to destroy the part ... Is to build a stand to secure the part through 6mm holes 50mm and 150mm from one end and then load it 1m from the outermost of those two holes. It may be so much fun the shop will forget to record the results!

Just to keep John happy:

I'm going to have to talk to the Human Resources about the help.

Ron

[wdyasq]

Jeffery,

I have a bunch of the 3/16" X 4' pultruded rods, probably enough to build an entire wing. If you want me to build a particular size and/or construction beam and test it, please ask. I can get almost any thickness birch ply and can easily slice it to make 45 degree plywood. If I had to, I'd build a foam core box beam using the proper glass and direction, multiple 'pultruded rods' and vacuum laminate as necessary. If necessary, pultruded FG could be used as posts and diagonals in the 'beam'.

I do believe it will be a good idea to build a full size spar and 'test' it before building a full size wind and loading with bags.

I've been thinking about the wing loading and might use feed bags as they are individually weighed and accurate. I also know some farmers and ranchers who would probably allow me to 'haul a load' or two to the farm/ranch and understand they would spend a day or so in test mode.

Ron

[wdyasq]

The parts and pieces finally came in .....

The 'rig':

Final Load:

What it looked like.... I measured 10mm total deflection at 50kg ... the load cell weighs 1094 grams and the cell has a resolution of ~.2kg. I think with the more narrow beam, this will probably be close to correct of Jeffery's calculations. If one takes a close look, there is an ink line ~ 10mm below the scale that represents 'zero deflection'.

Other loads and deflections:

10kg - 1mm
20.4kg - 3mm
30kg - 6mm
40.4.4kg - 8mm
50kg - 10mm

Working in metric wasn't all that bad ... all I did was buy metric rules and count the little lines. It sure beats counting barley corns and carrying rocks!

Now, to build a rig to destroy the evidence!

Ron

[John Nicol]

Hey Ron,

For some reason, I completely missed this post. Shame on me. This is wonderful work and a fantastic addition to the body of knowledge. This looks like a great way to build a spar. What do you think the next step is with this?

John

[wdyasq]

Hey Ron,

For some reason, I completely missed this post. Shame on me. This is wonderful work and a fantastic addition to the body of knowledge. This looks like a great way to build a spar. What do you think the next step is with this?

John

There will be several steps. I'll load this one in a cantilever and destroy it. I'll see if my camera will do long enough 'video' to actually record load and destruction. How is 'blows' might give us a clue to build a better one.

I'm guessing Jeffery can SWAG (Sophisticated Wild Arse Guess) what size cap and web/box(including web/box material) will be needed for a full size airplane. It may be we need to build a few more test spars before a full size one is built. If my Guess-da-muck thinking is correct, a 1/2 scale spar would carry 1/4 the load of a full size spar. It is a lot more economical to build in scale. If, it will scale up. It will not be difficult to build a loading device that will proportionally load a cantilever beam similar to what it will be loaded in actual use. Once one builds a full size spar that passes a multi-point load test, it will be time to build the full scale wing and load it.

Well, at least that is my thoughts on how to practically and economically build a wing. I'm working on a small, light weight device that will measure the flex of the wing under load. The idea of this is we get a 'trend line' of an "X" deflection under "Y" G force when the spar is new. This will be automatically calculated and compared every time the wing is loaded. If the spar starts flexing more, the device will indicate something not so kosher is happening. Hopefully, we never have a catastrophic spar failure. A flex under load measuring device will at least indicate the structure isn't as it was new.

Ron

[andrew_frost]

The article also explains why some men may prefer enclomiphene after TRT because it works through the body's existing hormonal feedback system. Unlike external testosterone, enclomiphene stimulates LH and FSH rather than suppressing them. However, the article makes clear that not everyone will recover enough natural testosterone production to maintain adequate levels on enclomiphene alone. Individual response is therefore an important part of the decision.

https://valhallavitality.com/blog/switc ... at-to-know