Flea Fli Build Blog

[Martyn K]

Thank for your comments kc.

I have just been out to take a close look at this and to be honest I don't think it will need (much) further strengthening.

If you consider the load path from one wing through to the other, the loads are transmitted as follows:

Intuitively, under normal flight loads, the loading on the wing increases proportionally (but non linearly) from the tip through to the centre and then reduces at the same rate going outwards.

From the wing tip - down 2 sets of spars to the first major discontinuity, which is the u/c block. However at this point the block isn't carrying any flight loads, so the load path continues through to the centre supported by the 2x ply dihedral braces then back out into the opposite wing.

However there is a second major discontinuity where the braces support the spar at the dihedral break with a potential compression failure point just outside of the braces.

Moving out from the centre, the load path is slightly different and this is compounded because the fuselage sides act as a pivot point and is therefore a major stress area on the wing when under +G bending load. This is aggravated because this is also where the dihedral braces end.

I think that in these circumstances, the load outwards will sill primarily be transmitted through the spars but also through a secondary path through the u/c block as it is currently fitted. There will be a compression load on the lite ply ribs (which coincides with the fuselage sides, but because they are ply (and wont crumble) the bending load will also be transmitted out through the bearers (as well as the spars) and then back into (mainly) the main spar via the balsa ribs and sheeting. Providing the joints at the dihedral braces are well made and properly glued, these should be as strong as the material they support.

It would be possible to add a third ply dihedral brace across the u/c blocks but I don't think this will add a lot of value unless I also add something to stick it to - i.e. extend the u/c blocks as well - to prevent the ply brace twisting under load. Adding a 3rd brace will not prevent the main spars from buckling either - I agree that the wing will be very stiff across the u/c blocks, but the same bending forces will still occur at the LE and TE.

My preference is to add balsa webs to the front and rear spars 2 bays out to prevent the spars from buckling at the area of maximum load (and hope that my assumption about glue and joints is correct).

You also have to bear in mind that this is a small model (39" span) and will never have huge flight loads imposed on the airframe no matter how you fly it - only when you crash it.

Regarding bandage, I agree, I certainly don't see the need for a glass fibre bandage. It does nothing for the strength of a built up wing as it only really works when you have a none compressible solid core.

BW

Martyn

[Martyn K]

Despite having family round last night, I made some good progress on the wing.

Webs were added just outside the centre section as described above. I decided on 2 bays for the front spars

and 1 bay for the rear spar. Note that the purpose of the web is to prevent the spar from buckling. Balsa is strongest in compression down the grain so they are fitted with the grain vertical.

As I was about to start work on the ailerons, I decided to get the aileron leads out of the way. I am trying Alex Whittaker's advice in RCM&E and cutting small slots in the ribs then putting the cap strips on top. I just hope that I don't have to change a servo...

So now it is time to mount the ailerons. Remove the last bit of rib from the end of the wing and the ailerons should now slot in with a nice 1/32" gap at each side..

You will need to insert a piece of 1/2" x 1/16" sheet balsa into the top of each aileron for the hinge to fasten to.

I normally cut the sheet, bevel the front edge to the right angle, lay into place and then mark the ribs where the back edge of the strip goes.. Cut the slots by eye - it's quite easy to do this with a 1/6" deep slot and a sharp blade. Glue them in.

The ailerons are quite thick and horn mounting isn't particularly straightforward. Phil Kraft used home made fibre glass horns sandwiched between 2 balsa ribs. I want to use small commercial horns, so I have glued a couple of 1/2" x 1/8" spruce bearers (again with the front edge planed to a bevel) in the inner bay on each aileron. The horns will be screwed into this using small wood screws.

(You can also see how the lower sheet has been sanded to a slight point in this photo)

Now we mount the ailerons (when the glue has dried. I cut a strip of the Kevlar cloth, (almost) the full length of the aileron and 5/8" wide. This will be sandwiched under a 1/16" balsa strip under the wing.

Glue the strip down using cyano, then add the top balsa strip (again using cyano) and then finally, I add a wick of cyano down the seams. Press down with your fingers to make sure there is no air trapped. Before you add the balsa. sand a small 45 degree bevel on the rear face

 

When dry, use the aileron as a template and cut the 1/16" sheet top cover. Add a 45 degr.e bevel to the top front face. The next bit needs a mixture of glues. I use PVA and apply glue onto all of the ribs, the TE and the rear half ov the 1/16" doubler we just applied, followed by a seam of cyano at the front edge.

Warning - you will not get a second attempt at this....

Place a 1/32" spacer at the end of the wing where the aileron fits, locate the aileron against the spacer and push the hinge onto the cyano, apply a layer of cyano on the top of the Kevlar and then add the top sheet. Push down to make sure there is no air, remove the spacer and pin the sheet where the PVA glue is drying.

Remove the packing and Voila! - a perfectly fitted aileron with no air gap - and I guarantee those hinges will not fail under normal use.

Next step - finish off the sheeting, cap strips and add the aileron horns then I'll make a start on the fuselage and tail.

More to come.

 

Martyn

Edited By Martyn K on 08/11/2013 09:45:01

[Martyn K]

one thing I forgot to mention. On the original, the ailerons were extended out to and incorporated the tip block. This leaves them a bit vulnerable so my ailerons will be inset and the wingtip blocks will be one piece. There is a slight reduction of aileron area doing this, but I don't think it will be too significant and it will help prevent damage to the ailerons that could be transmitted to the servo.

Martyn

[kc]

What is the Kevlar strip you used and where is it sold?

As regards the u/c blocks I feel that the wing is so much stronger along the u/c blocks and then is much weaker at the fuselage sides. So in the event of a slight crash or cartwheel it will surely break just inside the fuselage. If the wing is rubber banded on then it may be less likely than if it's fitted with wing bolts. However as long as it does not fail in flight is all that matters. My comments are just to advise other builders to consider making the blocks meet in the middle.
Presumably it's just a servo lead extension not the actual servo lead that would be built into the wing. Extensions are so cheap now they are expendable but very unlikely to fail except in a crash. Servo could then be replaced easily if the plug & socket are within access. i.e the servo lead iiself is not builtinto the wing only the extension. This may mean a slightly longer extension would be needed.


Very comprehensive building instructions here Martyn and mostly applicable to lots of similar designs, so I am sure this will help many people.

[Martyn K]

Thanks kc - again I really appreciate your comments - always constructive and thought through.

The Kevlar material is actually sail cloth - I have enough to last me a lifetime and probably the next 10 generations of family aeromodellers. It is very tough, much more flexible than mylar and so easy to apply.

Happy to share it, if you want some to try (enough for a model or two) send me an A5 SAE and I'll cut you a length. It shouldn't be folded, it will be returned in a toilet roll centre so the envelope will need a large letter stamp.

Regards

Martyn

[Bob Cotsford]

I know it's a bit late, but with the dihedral braces I like to extend them along the next bay out where practical, and taper them over this last bay to moderate the sudden transition from ply braced to balsa. A horizontal V - > - cut into the end of the brace spread over a rib bay makes it look as though you know what you're doing

[Bearair]

I have been following this thread and enjoying the constuction a great deal. I remember building a Flea Fly many years ago. I cannot remember what mag the plan come from but I do remember it was on yellow paper for some reason. If I remember correctly I had a ST 15 in it and it flew very well with plenty of power. Radio would of been a Century micro rx with fleet mini servo's a 250ma deac and a home made transmitter. I sold the airframe to finance a 1/2a Baby Birdy and a 1/2a FW190. Being able to cycle to the airfield with a couple of models during the school holiday was my top priority!

Roger

[Martyn K]

Hi Roger

The plan was free with RCM&E back in 1968, plan value 7/6 (real old money). I am quite pleased with the way it is going together as well, I just hope that I have got the electric bits sized properly.

Despite only getting less than 4 hours sleep last night, I was back in the shed at 9:00 this morning with the target of getting the wing finished today.

So..

First a shot from underneath the wing showing the aileron and the gap to allow the downward deflection. I have also completed carving and sanding the tips. I have decided to leave them "squarish" rather than fully rounded off.

A view from the top. Note the servo leads are wrapped loosely round the servo to keep them out of the way.

All the cap strips have been added - they are stripped from light 1/16" sheet, 3/16" wide and cut accurately so they hold themselves in place between the LE and TE sheeting. I then flip the wing over and run cyano into the joint. plus a tiny amount at the butt joint on top.

The upper centre sheeting is added but glued using PVA and pinned in place until dry. Small slots to allow the aileron extension leads are cut and they are fed through before the sheet is finally glued into place. When dry, sand out any imperfections.

That is the wing just about built. The 3/16" locating dowels have been fitted and glued with epoxy. Apart from securing the pegs into place, the epoxy adds a certain amount of additional local strength to the LE..

The weight of the wing at this stage is just over 7oz..

So... I have started planning the Fuselage construction. It will need a few mods to accommodate the brushless motor, the ESC and the LiPo.

So a quick assemble. I hadn't actually looked at the motor until today - it has a couple of extra bits that I wasn't expecting including 2 large O rings. Does anybody know what they are for?

Proposed layout of motor and 60A ESC, plus a shot of one of the 2 "O" rings supplied.

First pass at the revised F1 location, I am moving it forward (by 12mm) and making it smaller to maintain the lines of the model. The top hatch will be extended forward to provide access to the motor and LiPo which will be located on a false floor above the ESC. I'll need to get some cooling air in there as well so an intake in the upper hatch and lower floor will (hopefully suffice) and then vented out through the fuselage and exit somewhere near the tail

The 1/32" ply doubler will need extending and cooling slot located in F2 to let the air through. This will be a bit of a challenge as F2 has the mounting holes for the wing pegs.

Finally, a shot with the LiPo located. The power leads for the LiPo will pass through F2 and loop down to the ESC underneath it.. (Well that is the plan so far - in a few days, I should see if any problems appear from that idea.

More to come

Martyn

[Martyn K]

Posted by Bob Cotsford on 08/11/2013 17:29:13:

I know it's a bit late, but with the dihedral braces I like to extend them along the next bay out where practical, and taper them over this last bay to moderate the sudden transition from ply braced to balsa. A horizontal V - > - cut into the end of the brace spread over a rib bay makes it look as though you know what you're doing

Hi Bob

I have to admit that is what I would have done (except the 'V' - nobody can actually see that so I don't bother), but I have followed the plan for this with PK's recommendations.. TBH, it's such a small model, I think it is massively over-engineered, The wing is so stiff, I cant see it breaking in normal use - or even if thoroughly abused.

 

Martyn

Edited By Martyn K on 09/11/2013 22:25:43

[Bob Cotsford]

I can usually find a way unfortunately it's not usually through flying!

[kc]

The O rings are possibly for a prop saver type of prop adaptor.

[kc]

How do you fix the nosewheel if the bulkhead is moved forward? A cooling hole in the sheeting under the ESC might be easier than putting another hole in F2.
I like to use a liteply battery tray with plenty of alternative holes for plugs etc. This reinforces the fuselage despite the old IC engine bulkhead being relocated forward. The ESC always goes undeneath and the Lipo on top in my models. It is surprising how difficult it is to tuck the wires away so I put the extra holes in wherever I think the plugs might need to go and this should also allow the air to circulate a bit. I found it necessary to put the ESC the other way round on my small models i.e. the motor wire come out aft and the Lipo wire forwards. This was because I didnt want to chop the wires off short and they wouldn't tuck in such a tight space. A bit of time spent making the bulkhead and lipo tray first and juggling the wires around saved me chopping the fus after construction on my second electric model ( dont ask what happened on the first! )

Edited By kc on 10/11/2013 15:32:23

[Martyn K]

Hi kc

OK on the O rings, there is another bush in the pack but no instructions! I'll see if I can work it out.

My plan is to build a liteply box as you have suggested, but I need to be able to get access to the ESC which will be underneath it. I'll worry more about it when I get the fuselage together. One thought I had was to loop the main power lead from the LiPo back into the area above the wing then forward again to the Deans linked ESC plug. This means that I will be able to connect the battery then feed the lead back into the fuselage - hopefully this will make the cable routing easier.

As you have spotted, mounting the nose wheel leg will be tricky, I have got an idea for a self steering (rather than servo connected) nose leg that I want to try. It should simplify the mounting, it's something worth trying.

BW

Martyn

Edited By Martyn K on 10/11/2013 21:21:15

[kc]

On a couple of my electric models I made the liteply tray and the ply doubler sides the main strength to hold the motor bulkhead. A sort of I beam instead of a box girder. This allowed a hatch on top for Lipo and a smaller hatch underneath for access to ESC, motor wires etc. If you see my photos in the Swamp Rat section and also Mini E Frantic build you will see what I mean. Both these were quite small models and it would have been rather cramped and messy without tucking the ESC & wires out of the way.

Edited By kc on 11/11/2013 11:38:06

[Martyn K]

Thanks kc

I'll take a look later.

Martyn

[Martyn K]

A bit of progress last night, I started work on the engine and nose wheel mount.

A reduced size former (45mm x 52mm oblong) cut from 6mm ply and drilled for captive nuts (M2.5). Additional holes (10mm) have been added to allow air to circulate. The motor is screwed into place and the excess bolt length cut off with a dremel and carbide cutter.

I am trying to mount the nose leg on a single beech block. The nose leg was bent up using my wire bender but significant additional rake (castor) was added. The idea is that the nose leg can rotate freely, but the castor will keep it tracking wherever the rudder dictates that it should be pointing. There is space to add a steering arm if required, something I may do if I am not happy with the solution. The hardwood block is glued and screwed into place, you can see the pilot holes in the lower photo.

You can also see from this shot the ESC has had the motor leads significantly shortened. I may loose a bit more yet if I have trouble getting the cables to fit neatly.

Finally. I am avoiding using collets on this model. The stops for the leg (and wheel) are simply a few turns of copper wire soldered into place. The pitting is caused by rapid cooling (water spray) as the heat was burning the wood. It doesn't look brilliant but the joint is sound.

Just started cutting out F2 before I started work this morning.

More to come

Martyn

[kc]

Martyn, I am going to criticise your unusual design in the hope it will provoke discussion and others say whether they think it will work OK...
1. Does a castoring front wheel work? Those who have had the front steering arm come loose have often found the plane unmanageable and then they lock them solid.
2. The proposed noseleg has much increased leverage on the front block compared to a normal setup where the wire goes most of the way up the engine bulkhead and is restrained by a clamp. Your setup has perhaps 3 times the leverage in my view. It wont hold in a poor landing or even a bump in take off ( in my opinion.)

Hope I am wrong and twenty real experts give their opinion & say you are right....
I am just playing Devils Advocate to get other opinions for you...

[Martyn K]

He he.

I don't mind criticism and I am always open to ideas.

If you don't caster a wheel sufficiently, then it is likely that the wheel will 'shimmy' and potentially lock as you have described. If the point of maximum drag - ie the ground contact area is significantly behind the pivot point then it should simply trail and follow the nose. Think supermarket trolley - then think what happens when they are not new.

When the wheel vibrates quickly from left to right (around the leg pivot) and back again - this is shimmy. It is usually caused because either the wheel bearings are stiff or the pivot bearing is stiff or it has been damaged in some other way. It is aggravated when the wheel spindle is directly underneath the nose leg pivot point (typical on most models) and has been known to cause significant problems on full size aircraft when (for what ever reason) the hand steering linkage has failed.

One of the downsides of too much castor angle is that the wheel becomes hard to steer, however, I am not trying to do that so I am hoping it will simply follow its nose.

Regarding your second point - that is very valid. However, my mitigating counter claim is that the pivot it in hardwood which is glued and screwed to an extra thick bulkhead which will be well braced when glued to the ply fuselage doublers. I *hope* it will be strong enough, the first landing on our boggy winter field (especially as the nose wheel is only 1.75" diameter - small) will prove the point. I think the landing plan is slow - slow - slower - and keep the nose up until it stops.

I would welcome other thoughts on this though. If I am going to change it, now is the time...

BW

Martyn

Edit - one point I forgot to mention. The wheel centre line must be EXACTLY under the nose leg centre line and all angles must be either vertical or horizontal as appropriate. Any deviation from this and it wont work...

Edited By Martyn K on 12/11/2013 12:08:12

[Bob Cotsford]

I'm with KC, I think it will last at most a couple of landings before it becomes too slack as the wood compresses.

What would I do? Use conventional brackets on the bulkhead and move the battery back the required 8-10mm. Fit the brackets with captive nuts on the motor side, cutting the bolts flush as you have with the motor mounting ones.

My KFC has a fixed noseleg, it steers ok for take-off and landing but not for taxying on grass.

Or just put the wheels aside for another project and chuck it up.

Edited By Bob Cotsford on 12/11/2013 12:25:55

[Martyn K]

Looks like I am out-voted...

The original plan did show hardwood blocks for the nose leg - there were 2 of them and the upper one was just below the centre line. If you look at the lower of the 3 shots above, you can see the blocks and the symbols for the wood screws.

Remember, this is only a 2lb model we are dealing with. I am trying not to move the battery back if I can help it as it (at the moment) will be located nicely between the 2 front formers.

I'll take another look.

Martyn

[kc]

Aeromodelling progresses by trying something new and seeing if it works....we may be wrong and Martyn could be right. It's a matter of whether you want to play safe or trust your own judgement and try it....
Personally I would hedge my bets....drill the bulkhead for u/c clamps in the conventional way - 2 vertical and 1 horiz- in case the block breaks off.

Either way I would make sure that the soldered substitute for collets did not chew into the wood then get nearer the wires & chew them causing a short! A proper collet would have a larger bearing surface.

Edited By kc on 12/11/2013 13:48:07

[Martyn K]

Posted by kc on 12/11/2013 13:38:11:


Either way I would make sure that the soldered substitute for collets did not get near the motor wires and chew into them & cause a short!

Ah yes - a very good point. At least there wont be fuel on board to catch fire when the sparks happen.

Back to the nose leg design, I may (probably) place a pair of ply gussets each side of the hardwood block each side of the point where the cables come through. That will transmit any loads up the firewall in much the same way that second block would do. If you think wear on the pivot will be a problem, then I can add a brass bush in the hardwood to provide some support, I am not convinced that will be necessary though. The wheel bearing takes far more abuse than this and I suspect that the block will last far longer than the wheel. Again - time will tell.

One thing that I did consider was to make the leg out of 12g wire rather than 10g. It was far more springy and probably handle the shocks better that the stiffer 10g wire. I can't imagine a 10g wire bending at all under normal landing loads. I may have a rethink on that.

Martyn

[kc]

A brass bush would be metal to metal and could cause radio problems.

A plastic U/C clamp instead of the block and another clamp much higher up would be so much stronger and more reliable! Then only a smooth bit of u/c passes the motor wires. It would mean making a longer wire leg but less trouble in the long run. The Lipo need only move back the thickness of the plastic clamp- about 4mm.
But it's your model!

[kc]

Just had another thought about the castoring nosewheel. Supposing you are flying along straight and level then decide to bank left....the nosewheel would presumably fall left too, a bit more weight on the left side and it's a bit further forward too, changing CG slightly, maybe the wheel side on causes more drag on that side? Does the model turn more sharply or just behave normally? Would the airflow keep the wheel and u/c back in line? Would stops to limit movement be necessary?
Although castoring tailwheels are common they are small & light. I cannot recall any model having a castoring nosewheel - can you?

[Martyn K]

Hi kc

Good points as usual. I have added an additional heat shrink sleeve over the motor wires where they exit from the bulkhead and added a stop limiting the wheel movement to about 30 degrees in either direction. I don't think that there will be any shift in CG, certainly it will be far less that the movement of fuel in a tank or due to the fuel diminishing.

Regarding the strength. I think it will be ample strong enough. I am using the same bearers that I use for the main u/c legs, but this one is glued and screwed into a 6mm ply bulkhead which will be braced into position with balsa blocks along the fuselage sides. As the main wheels(!) should be taking the landing impact, I think that this will suffice.

As you can see, I have also added ply gussets to help support the block, although I don't think they are needed.

Martyn

Edited By Martyn K on 18/11/2013 09:47:34