Clive Hall

I do not know how to calculate fin offset, but I tried it on a Flair Hannibal. I took a SWAG and by sheer luck it worked. [SWAG —scientific wild-ass guess.] There is some mythology about offset fins, to the effect that they will cause endless trouble on low power. Apparently they don’t, probably because the offset is calculated to work at high speeds with a strong slipstream and has little effect, if any, on the glide. Not that WWII fighter pilots tended to stay on board if their aircraft were disabled to that extent, so maybe it didn’t matter to them. As for changing fin offset, you can’t, so it’s safer to leave it alone unless building a scale model where the amount is documented. Incidentally I hope to join in my first Mass Build: how does one go about starting a build blog?

Seconding Peter Miller’s comments on the lack of offset thrust on full size aircraft, consider what would happen to aircraft with Merlin engines with a few degrees of side thrust; they would have to be much fatter in plan view to get it all in. I know that the Fairey Firefly had a strong swing to one side on take off and that lots of rudder had to be applied in good time or it could barely hold it, so that would have been a case for sidethrust. Instead they depended on pilot skill. Note, however, that downthrust was used on some WWII fighters, notably with Griffon engines and, of course, the P-51 Mustangs.

May I chip in here with a comment about spar webs? The prime function of webs is to prevent the upper and lower spar booms from shuffling relatively in a spanwise direction. Without a web the wing under a lift load can bend, moving the upper beam outwards relative to the bottom one and cracking the joins to the ribs as they try to assume a different angle to the boom, sharper on the top, wider on the bottom, and most evident towards the wing tips. Therefore to resist this movement the webs are in shear rather than compression. The ideal material for wooden wings is plywood with the grains set at 45° to the spar booms, though the standard practice of vertical grain sheet is fully adequate for models. Some full size aircraft do not use sheet webs but fabricated box section joiners between the booms from the bottom of one rib and the top of the next one: this was done on the Boeing B-17 along the whole wing. Another point concerning sheet webs is whether they should be fitted to the front or rear of the spar booms. The only reference I can find here relates to a decision in August 1936 to make Spitfire wings torsionally stiffer. This was done by moving the webs from the front face of the spar booms to the rear face.

1. Oodalally 2. Ballerina 3. Werewolf

Erfolg’s thoughts on model spar structure remind me of a conversation years ago on the topic of wing spar design. The first instinct was to use ¼ square hard balsa for the beams, sheet balsa webs, because that worked well on similar sized models. Then a magazine article explained how the flying loads are distributed on a spar and we progressed further into thinking that some weight could be saved by using progressively less material in a wing as the construction moved out towards the tip, reducing the beams at the appropriate point to ¼ x 1/8. We nearly made the mistake of applying this admirable, and correct, theory until a wiser voice said ‘What happens to the wing when you cartwheel it on landing?’ Memories of some undignified landings put us back on the sound path of relying on experience, and the advantages in the real world of a bit of over-engineering. On another topic entirely: about the 3.5-root or cube-root basis for calculating a change in model size. It matters little if the range of engine change is small, but it begins to turn silly for a wider range. Take a Spitfire I, 1030 hp Merlin, wingspan 442 inches. Convert it for a model using an OS GT15cc petrol engine. The result of a 3.5-root calculation is a model size of 74 inches span, a reasonable prospect with hopes of a scale performance, although the structure will be that of a typical model, not duralumin sheet. The cube-root version of the calculation indicates a model of 55 inches span, something of a hot-rod with that OS engine, better suited to a glow 40 or 46.

Additional thought; it doesn't matter what units are used for span or engine capacity so long as they are consistent. The formula, strictly speaking, should use engine power, but model engine makers tell such whoppers about the power of their products that it is much safer to use capacity. Within the range of typical model engine sizes it works well .

There is a formula for scaling up a model size. It works well though at first glance it is not quite what you might expect. Use these values:- Original model: engine capacity is EO, span is SO New model: engine is EN, span is SN. The formula is:- SN = SO x 3.5root of (EN ÷ EO) The natural instinct is to expect it to be a cube root function, but because power is involved (and power is a rate) there is also a time dependent factor and it makes it the 3.5 root instead. Try it, it’s reliable. Obviously it works directly only for engines of the same type, i.e. if both engines are four-strokes or both are two-strokes. For other comparisons one has to make the usual allowances for the different types.

Great news! I have in mind a Li'l Mustang at 58 inches for an Evolution 10cc gasser. If anyone knows where I can get the new carburettor to convert it to the pumped version I would love to hear.

All my engines are too big for the published PM designs. Are there any rules on enlarging a subject for the mass build?

The difficulties being experienced here are probably compatibility problems. I suggest installing firmware 2.0.17 but then be absolutely sure to install OpenTx Companion version 2.0.17. If you prefer to use 2.0.15 then again make sure the loaded firmware and the Companion have the same version number. The old 9x will l not work with a Taranis Plus. Also when selecting firmware downloads be careful to tick the box for Taranis Plus and NOT just plain Taranis.

If we are to curtail the activities of terrorists I think laws are essential to limit the possession of dangerous materials such as glow fuel above 30% nitro. Once that stuff is available only to licence holders surely they will not try to buy it, after all they would not want to do something illegal. ??

Quoting Alan Roberts . . . “ . . . it’s what people do with the tech that should be regulated and not the equipment.” My point remains that, given the kind of equipment that the majority of people are using, it should not be necessary to regulate them, but there is a danger that this can happen if we get a blanket of ill considered new rules.

I asked for comment on a model grading system: I got it, and I agree with everything that has been said, but although it is all sound and sensible some of it misses the point I was trying to make. The reason for my grade grouping was that I can see the possibility of forthcoming legislation which could have a too wide effect on what we can fly. I suggested it may be draconian: that means legislation that is oppressive, which is what we could easily get from our politicians. There is an election coming up and they often produce knee-jerk reactions seeking popularity, egged on by our sensation seeking media which will delight in damning all modellers regardless of common sense. Alas, I think the quads are going to have trouble anyway, but there is no reason why restrictions on them should spread to all model flying; it easily could, so I was just looking for a distinction that might be applied to save the simpler models from inclusion in harsh restrictions. I included models without any kind of stabilisation (Grade 1) because it covers the large majority of models as flown at present. I don’t see how anyone could quibble with that idea, but it would be brilliant if someone could come up with an idea that the BMFA could put to the CAA to save us from the worst that might happen.

Hi Alan Another thought which simplifies my grading: the essential difference between the natures of groups 1 & 2 and groups 3 & 4 is the self-levelling feature. That alone could be used as a criterion for separating the sheep from the goats in the model world. But you are right about the desirability of auto-level and come-back-home panic switches. I am looking already at sources for such a system which can be linked to failsafe at the same time. I too want to fly my models only close to the club strip, never out of sight, only occasionally with cameras aboard, but, as you say, that would drop me into the public enemy bracket too. Silly really as it would be a safety feature. It is a knotty problem trying to find any features of the on-board systems that would make a clear distinction between club models intended only for local flying, and the far roaming quads. Edited By Chris Bott - Moderator on 09/12/2014 22:10:55

In my view models, whether fixed wing or multicopters, can be grouped into four categories depending on the complexity of any electronics fitted to stabilise them :— 1 Simple models that are flown manually only, with no on-board aids. 2. Models with 3-axis gyro-only systems that remove the bumps caused by wind or turbulence but do not in any sense themselves have any ability to fly the model. [Example: Orange RX3S] 3. Models with 3-axis gyros and accelerometers that can level themselves and thus become capable of flying long distances after being lost from sight.[Example: Eagle Tree Guardian] 4. Models with full FPV and GPS controls with the possibility of long out-of-sight flights, and waypoint operation. The first two categories do not present any more hazard than models have for many years, whether fitted with cameras or not. These should be easy to exclude from any additional controlling laws. The real danger which will attract (draconian) legislation comes from categories 3 and 4, which offer the potential problems already well aired. Any comments on this grading system?   Edited By Pete B - Moderator on 09/12/2014 17:32:26

I had similar troubles with short Tygon tubes supporting a clunk, and not reaching the bottom of the tank. Then I left the fuel in a tank for a few days and the Tygon softened. Now it flops around the tank as it needs to. I now use this as standard practice. Only a small amount of petrol is needed to keep the clunk tube flexible. However if it dries out it becomes very stiff again.

I don't see how the gyro could hold the spin if the normal spin recovery was applied, as the full stick deflections of a flashed RX3S should override the gyro. If you were using one as yet not flashed, then I do not know how it would behave. It's a good idea to be ready at any time to flick the switch to turn the gyro off if anything unusual happens.

Hi David A picture of the drawing file and a .pdf file for the Algebra F3B wing and tailplane are now available on the following link : - www.cbhmodels.co.uk/algebra.htm There is also a .dwg file prepared but it might be simpler if I can email it to you. PM me with your address if you want the file.

Hi David I guess that what you need is a wing and tailplane plan with the airfoil (Eppler 193) rather than the full drawing, which is big, and could be scanned only with difficulty and even then only in lots of small bits, and would take a long time. Maybe the best move will be for me to re-draw the wing and tail in a CAD program and present the result as a 1:1 .PDF file which I can email and you can print. This would not take as long as the scan. Or I could send you a .DWG or a .DGF file if you have software that will read one. Does that come near what you need? Clive

Yes, Ian, the chip in the receiver is the same one as in the RX3S, and it is flashed in exactly the same way with the same results. There is one limitation—the receiver is only six channel and Ch-6 is used to control gain, so it's not really available for anything else, making the receiver effectively 5-channel in practical terms I have just flashed a 6-channel receiver and it works with all the improvements as listed elsewhere. Excellent range test (60 metres) with a JR DSX9 transmitter this morning.

That's right, the V2.1 firmware is the one which fixes the servo chatter and production is now upgraded to V2.1. The other little inconveniences remain. The firmware flashing described on RCGrps fixes the old snags and goes on to do even better.

Sorry, Phil, I did not directly answer your question. Yes, the new firmware does stop the servo chattering. It makes the dual aileron output pins work correctly. It improves the way the pots work so that they are not so sensitive to adjust. It gives proportional control of gain from the transmitter. It also introduces a second operating mode - hold mode which is a bit tricky to handle. The latest version of the RX3S is V2.1 and is reported as having also solved the chatter, but the rest remains the same as before.

Phil, I have written a fuller description of the RX3s both as sold and as modified. You can find it on : - www.cbhmodels.co.uk/rx3s.htm Maybe that is what you are looking for, hope it helps.

The way most models are set up the dual aileron outputs of the Orange RX3S will not work; they move the ailerons up or down together. There are other weaknesses, such as the pots being over sensitive through only a small part of their movement. Also any control reversal on the RX3S reverses both the gyro correction and the servo output so it then has to be reviewed on the transmitter. There is a method for flashing the board with new firmware which transforms the performance. The pots become reasonable to use, the aileron outputs drive ailerons as they should, and the gain control from the transmitter becomes proportional. The detail for this is covered at considerable length by following this link . . . http://www.rcgroups.com/forums/showthread.php?t=1794672&page=1 . . . but there is plenty to read, 190 pages. Once mastered, the board flashing is easy and quick to do and transforms the orange device into a really useful piece of kit. I've done all mine and it was well worth the effort, including that of finding out how to do it.