Techone

10 December 2018

Test Session 4

Port wheel component superglued.

On arrival, the EPP foam pant parted from the port wheel component.

I used 1 pc of 2s800 followed by 2s500.

Was it me that it exhibits port side heavy now?

The model was more neutral with 500 but the 800 seems to provide better power to the motor.

Out of range seems to be a possibility.

It recovered in time and did not crash.

1. Consider using 'proper' Flysky receiver.
2. Hotglue the port wheel pants.

No rush, I am not too into this hobby now. Currently, interest in golfing.

7 December 2018

Test Session 3

I used 2 nos. of 2s500 and 1 of 2s800.

I discovered 3 units of 2s800, and had only charged 1 last night.

Straight flights seems more positive using the 2s800.

Inverted did not need input after 1-2 clicks elevator on level flight.

The CG must be in the zone.

Changing back to 2s500 is no problem.

Most landings displaced the port wheel component.

I have to snap it back on and/or adjust.

I think I will superglue it on.

I swapped the battery to the port side.

The wings became level and the low starboard wing symptom is past.

Power on, power off is much better now.

Straight ups, rolls etc are better behaved.

I do not know if it is worth it, given the nature of EPP, will it be better if I move the esc to the starboard side where the receiver is?

Tidying up, I could glue the esc and receiver.

Or, should I try a 'proper' full-range Flysky receiver?

There was a glitch this morning.

6 December 2018

Test Session 2

Normal rate is now 30% aileron, rudder. Elevator, I forgot.

The model became docile at Normal rate.

The previous battery that caused a deadstick caused a deadstick again, no power and straight down from 10-20 meters, zero control. Cell checker says it is 7.6v though. Anyway, this battery is marked.

The port wheel component broke off from the landing leg. Next flight was one wheeled, later on, at the end of the flight session, I discovered I could just snap it back on. 

This crash caused the elevator pushrod length to be lengthened. I had to use the adjuster several times, this time at the servo horn's end.

Hovering and knife edge seem doable in the morning air. Snatches of 3-5 seconds of hovering and 10-20meters of knife edge were done. Tail-slide is possible, 4 points roll, hammerheads/stall turns and Cuban eights were also executed. All that was lacking is familiarity and my skill level. Just need more practice.

I think the CG is about alright. I could put a drop of superglue to lock in the pushrod lengths, and maybe I will do them at a later stage. I want more and better batteries to do future trim sessions. That, or simply have more trim sessions with the current batteries, 1 marked and 1 good.

5 December 2018

I glued the servos and adjusted the pushrods in the manner I wrote of previously. I charged up a battery and left the model alone for a few days. I was surprised how fast that battery charged, but discovered this morning that I had charged the un-used battery. I had only one battery to fly this morning.

The elevator looked flush with the stabilizer, but the whole horizontal tail appears to be at a slight positive angle to the wings. That ought to be impossible, but that was how it looked like, maybe it is the EPP foam.

I hover-launched and pulled elevator to maintain attitude. Landed shortly and  shorten the length of the pushrod a bit by a bit. After a few more hover-launches, landings and adjusting. I got to the stage when the model is almost neutral rightside up and upside down, and I would describe the model as "floaty", not glider-floaty but the imprecise wishy-washy type. I just did not get consistent behaviour. Should I move my CG forward just a bit?

The model continued to exhibit a starboard low attitude. I tried a combination of left aileron and rudder. I was just playing around, to see if they would help. The wings did not stay level, and knife-edge appears to pull towards canopy.

The elevator throw is enough for loops and bunts but I think the aileron throw is still too much. Maybe the model will be easier to manage after I tone down the aileron throw. Thinking about it, I think I should reduce the rudder throw as well because currently it is too much for knife edge, just a bit lesser might be good.

Flight ended with a deadstick. 1st cell was about 2.86v, 2nd cell 3.7something v, must have got a bad battery.

For now:

1. Charge the 2 batteries.
2. Reduce at the transmitter, the aileron and rudder throws by 1/3 (at normal rate), sticking to the same exponential setting.
3. More test flights with elevator adjustment and roll/yaw adjustment.

29 November 2018

Last night, I sighted the guides and superglued them. I didn't re-insert the guides with superglue, I relied on applying droplets to the junctures of guides and foam. Maybe I should have removed the guides, coated the guides with superglue and then Insert and align the guides once again. Thinking about it now, the cure time of the superglue is quite long, and it ought to be enough for aligning. I was perhaps a bit obsessed with getting the guides aligned and placed my faith of the ability of superglue to seep to the contact areas. That is what superglue ought to behave, no?

I didn't use kicker to accelerate the curing of the superglue. The idea being that I would like to give the superglue time to seep into the foam. I am thinking that if the medium viscosity superglue cured too fast, it will not seep in properly. So, I dipped the junctures of guides and foam and left it undisturbed until morning. 

This morning I checked on the guides and discovered one loose, a success rate of 5 out of 6 isn't too bad. So I applied a bit of superglue and kicker onto the loose guide, found it cured hard and secured. Does that mean I should not have bothered with my overnight curing idea?

I test flew it when the sky started to lighten, and observed:

1. A roll tendency to the right. Would that be the ailerons or the thrustline?
2. A tail-low attitude. Would that be the CG or the drooping ailerons?

For the apparent right rolling, I tried to shorten the left pushrod. Seems to work, then it seems not to be effective. I tried again, but it was still not conclusive. I also added some left aileron trim.

For the tail-low attitude, when the model was brought vertical, it sort of hovered. Just saying that it was interesting to watch. When the plane was inverted, it needed a lot of elevator. I moved the battery forward slightly, it seems to be better, then it seems to be the same, the cause-effect seemed non-conclusive again. I tried substantial amount of elevator trim, it seemed better, but when the model was inverted, it needed even more elevator. Why was I perspiring before the sun came out? Perhaps it was the humid air, I forecast rain for today.

Strange enough, the amount of trims do not change the positions of the surfaces much. Is this because this is an EPP model and slack is inevitable? Perhaps I should try to shorten both ailerons' pushrods, that ought to relieve the elevator trim totally. I still want my tail to be straight and flat. CG/incidence/thrustline should all start with the tail straight and flat. Taking the analogy of golf, I liken this to a 'neutral' grip. This is what I believe to be the 'neutral' setup which will result in less drag and less correction issues.

And then as I wagged the sticks in the morning light, it dawned that the servo horn moved strangely because I had not glued the servo to the fuselage and it was twisting! I had forgotten that all the 3 servos needed gluing. I did not write them as to-dos to remind myself last week.

That ended this morning's test flight. I shall glue the 3 servos and re-set the transmitter trims. This morning test flight is a waste. I think my procedure for setting the ailerons did not account for the weight of the ailerons which then resulted in a "flap-down" wing and therefore positive incidence and which require substantial elevator trim.

I shall set the ailerons in this manner:

1. rest the model vertically on a wingtip, this is to relieve the effect of the aileron weight;
2. on the upper half, loosen the pushrod's set screw so that the pushrod can slide in and out;
3. clamp that aileron of the upper half to its wing so that that aileron is flush and straight;
4. tighten the pushrod's set screw; and
5. repeat for the other half by flipping over the model so that it rests vertically on the wingtip of the adjusted half of the wing.

My other observations and further thoughts:

• My WLToys F949 flies faster.
• I did not balance the propeller; it was not "purrly smooth" with the motor running, it is definitely imbalanced, but hey...there's already plenty to do.
• I got to quickly progress to hover catch, nose dunking is just not cool.
• The bottom of the rudder is the lowest point at the rear. Should I glue something like an ice cream stick to the side of the fuselage or just release-launch and hover-catch throughout? 
• The roll to the right... could it be because my battery was strapped on the right side? I recall the roll sort-of stabilized once the model is at a certain shallow angle. This is one more to be take into account as probable cause. Really important to have wings level.

21 November 2018

 I started by soldering the 90 degrees header connector to the receiver board. It was a fiddly job, I pushed the header connector all the way into the board. It fitted snugly and will probably work even if I omit the soldering. But of course I have to give it a go, my soldering tip does not reach into the midst of the bottom end of the header connector. The board heats up (I must have placed the tip against the header too long) but the solder doesn't flow. Worse still, I noticed that some bits of solder formed a bridge with adjacent pins. I tried removing it with a needle nose plier, then a cutter and that didn't work. I heated the solder blobs again and tapped the receiver against the table and this worked. The excess solder broke loose and fell away. It doesn't look good, I don't see shiny silver at the base from the underside. The connecting points were blackened and the solder contact probably exists on faith alone.

The receiver, servos, esc & motor were all connected and I was happy that I managed to bind the receiver and have it working the servos and esc. As can be seen, the 4 servo plugs occupied the width of the receiver (15mm) and looks long.The receiver board is 21mm x 15mm.

Binding sequence (I thought YouTube would help, but was disappointed by the first search result,  that video didn't show any binding procedure. In the end, I googled a web page.):

1. Press receiver binding button, plug battery in to esc.
2. Press transmitter binding button, switch on.
3. (ESC start up music sequence)
4. Unplug battery from esc.
5. Switch transmitter off.

Setting up my Flysky transmitter:
Hmm... I used a previous AFHDS slot, and the "model-reset" didn't work, so I have to run through everything manually. :

1. Model name changed to "Extra800"
2. Receiver setup to "AFHDS2A" (obviously done before testing of the receiver)
3. "Reverse" for Channels 2 and 4 for direction correction (Channel 1 was fine)
4. Manual trims of all 4 channels were centered.
5. "Sub-trim" for Channels 2 and 4 for centering of servo horns (Channel 1 was fine)
6. "MIX" all set to off.
7. Didn't bother with the range, I think they look alright, so I left them all at 100%.
8. "Dual Rate" on Channels 1, 2 and 4 on left toggle. "Normal" at 50% range, -50% expo,  "Sport" at 100% range, -76% expo.

The horn connectors were adjusted to give approximate centers. Then I tried to get the pushrods to be in straight lines by sighting the pushrod guides along from the rear. But when the stick is moved, the pushrod ends moved closer to the surface/servo. Should I have the guides and pushrods straight when at neutral, at extreme or somewhere in the middle?

I didn't decide, so I left the guides unglued.

19 November 2018

The EPP Extra 800 looks so easy to assemble and since I have bought the kit including the motor, esc and servos, I believed the manual when it wrote that it takes 30 minutes to assemble. I was very surprised when I took many times longer, quite a few hours.

EPP is sort of porous, the superglue drips, and I messed up the parquet floor. It cured readily on my fingers and thumbs but not on the assembled pieces. The EPP pieces are cut accurately for assembly, but not that accurate for gluing. If on one side you see that the pieces match almost perfectly, then the other side you see gapping joint seams. I had to remind myself constantly that the EPP concept is flexibility and it is pointless to have full seam strength, so just dabs would do. So I spot glued with superglue, flip over, applied stationery tape on both sides of the seam and swiped hot glued over the seam. Basically, when the fit is good, I used superglue, and I use hotglue when I am confident I can bring parts quickly together and hold them in place.

1. The hinges do not seem to be long lasting, but I told myself it is enough and if they tear, I can just do a hotglue hinge later.
2. The EPP pieces for the fuselage didn't match at the nose, the vertical pieces were shorter. I pared the right side of the horizontal fuselage for a bit of right thrust angle and used hot glue to secure the fiberglass motor mount. I figured that hotglue is forgiving in terms of flexibility and gap filling and I don't know which way to go regarding down thrust.
3. For the LG, I inserted the CF rods to the T-cross pieces but did not glue them together because they fit very well. The wheels were inserted with the rims away from the T-cross, to keep the rims away from the wheel pants. The ankle pieces were glued to the CF rods because the fit was a bit loose. The ankle pieces were inserted deeper, exposing ends of the CF rods, then a dab of superglue on the exposed ends and then pushing on the ends until they are flushed with the ankle pieces. The legs were inserted through the slot, their ends inserted at 45 degrees to the 2 FG pieces with elongated holes, then superglued to the horizontal fuselage after making sure that the wheels are parallel to the fuselage. Then the inverted U pieces were superglued at the slot area, snugly against the LG.
4. The aileron servo and the elevator servo were placed with their horns towards the nose, but the rudder servo's was placed towards the tail because it seems the pushrod length is a bit short.
5. I used the short servo horn on the elevator but the long horn on the rudder to bring the pushrods straight.
6. The pushrod supports has to be inserted to varying heights when pushrods are kept straight. I will superglue them, and the servos, after I have my receiver installed and servos and horns centred. 
7. The bullet connectors for the Motor/ESC were too stiff to insert by bare fingers. I had to use a plier when my grip failed. I didn't have this kind of problem before.
8. I superglued the spacer ring that best fit the provided 8040 prop to the motor. The spacer ring doesn't fit the motor properly, I had to eyeball it to ensure some concentricity.

I am using this receiver because it works with my Flysky FSi6 transmitter (which works for WLToys' F949). I need to solder the 90 degrees header connector.

15 November 2018

I thought I wouldn't but I did. I bought a Techone Extra 800 (comes with motor, esc, servos, prop) and then another, a Techone MXS-HCF that was on offer. That meant also I bought 2 pieces of 2s500mah and a bottle of foam safe (for the depron MXS-HCF, but I figure it is ok for the EPP Extra 800 too. Hotglue was in my mind but nah...).

I suppose the Extra 800 will be a good plane to have in the boot. Now where is the charger? I need to find a receiver too.

Foam Cutter Frame

12 November 2018

Here I go again, another thought written out with no sight of when I might actually do it.

A dual usage adjustable bow construction frame for foam cutting; as a regular bow or as a gantry slicer. Adjustable collets allow easy securing, adjustment and replacement of the heating element. The heating element may be placed right on the tip of the cutting bow; allows thinner foam to be used and in the case where the airfoil has a flat bottom, eliminating the need to cut the bottom of the airfoil. Also, it can be a thin foam slicer.

Components:

1. A length of nichrome wire or stainless steel guitar wire for the heating element. Length = length of foam to be cut + 1 inch clearance to both sides of foam + 1 inch pulling grip to both ends of heating element = length of foam to be cut + 4 inches. Obviously, the power supply is a limiting factor of the length of the frame. 
2. Two 4" to 6" lengths of 1.5mm to 3mm springy wire prongs/legs. The lengths and diameters depends on how large the frame is to be and how thick is the foam.
3. Two wheel collets. More may be used and be useful, but two will be used to crimp on the wire legs and nichrome wire. The internal diameter of the collet hole should just clear the wire prongs/legs and the heating element as described above and securely hold the heating element against the wire prongs/legs.
4. Two pieces of wooden slats. Their lengths can be identical to the heating element. Drill holes that matches the chosen wire prongs/legs. The furthest holes are at 1 inch from the ends of the slats. The top piece will have reinforcement cross grain since it will be under tension. The top piece may be substituted with strong cord, then only one piece of wooden slat is needed. The bottom slat may be thicker since it will be under compression and it should not be bowed due to the eccentric compressive force. If using a strong cord for the top piece, the bottom piece become the sole spreader and needs to be thicker to prevent slippage.

Assembly:

1. Thread an end of the heating element from one end of a collet, insert a wire prong/leg from the other end of the collet.
2. Tighten the collet screw. Ensure the heating element and collet do not slip.  
3. Slide the free end of the prong/leg through the bottom wooden slat first and then through the upper slat. Maintain a slight gap between the two slats.
4. Slide the second prong/leg through the two slats.
5. Insert the second collet to the second prong/leg.
6. Thread the free end of the heating element through the second collet.
7. Pull the heating element taut and tighten the collet screw. Ensure the heating element and collet do not slip.

Uses:

1. As a loose bow. E.g. cutting airfoil over templates. 
2. As a gantry. E.g. cutting even thickness foam sheets. Making tapered foam sheets is possible through the adjustment of collets and additional runner.

Further adaptation:

With the frame secured against another surface, it can be used when  

1. Held vertically. E.g. squaring off or cutting multiple identical components from as many sheets.
2. Held at an inclined angle. E,g. cutting tapered pieces especially with additional runner. 

WLTOYS F949 Review

25 October 2018

This is not a new kit, it has been available on Banggood for quite some time, before I started my interest in WLTOYS' products. I bought it recently because it is cheap to have in the boot of a car for small field flying. I have a Flysky FS-i6 that works on the older WLTOYS protocol so I bought the bind-and-fly kit. It was less than SGD 60 for a bind-and-fly, about SGD10 cheaper then the RTF kit which included the toy-like transmitter. Regarding that transmitter, I have a similar looking one from buying WLTOYS V911 pro. That was about 2 years ago, I used it for other projects including those using the receiver boards of F929/939. I remember that the transmitter range is ok, but the gimbals are stiff and it does not have any mixes, reversing and other programming. I have flown F949 a few times and felt I know enough of the model's characteristics to have this to say about it.

When I unwrapped and opened up the kit box, I observed that the horizontal stabilizer was one-side-high. Over time, this became worse. In neutral trims and hands-off, it would fly in left circles. The high amount of dihedral stopped the circle tightening. Twisting the stabilizer didn't work, so it's either a heat twist or maybe a cut out and re-glue. However, it is good enough to fly so I didn't bother.

From the 2nd flight onwards, I noticed that the propeller seems to be moving away from the nose. This is confirmed when the propeller flew off later. There is a nylon piece on the geared motors to insert into the propeller. Pushing the propeller back on didn't help, because on power, the propeller would fly off the nylon shaft. I used a bit of UHU Por on the shaft and this solves the issue, at least for the last few flights that I had.

The power doesn't seem linear. The model only needs slight throttle to fly level turns, way below the 50% mark of the throttle stick position, but pushing the stick beyond the first 1/3 does not seem to get the propeller to spin much faster. Perhaps the propeller's diameter and pitch is set to slow cruising speed, i.e. slightly undersized for full range throttle performance. This trade off resulted in long flight time. It flies very long on the supplied 500mah.

The model has a tricycle undercarriage with tiny wheels. Despite the tiny wheels and the gentle power, ROG is ok on paved surface. Its take offs feels long, the roll reminds me of my Pilot QB 20H that many years ago. This is definitely not a yank and off-you-go rocket ship. You have to coax the speed up in order to perform rotation and take offs comes from having rotation at flying speed. This morning, I extended many touch-and-go to rolling on the ground and then rotating and take off again. It was pleasant and I must have performed dozens of this type of touch-and-go this morning, on the single one-cell battery.

Hand launching is easy. All it needs is a gentle level toss. Full-throttle or half-throttle doesn't make much difference. Just release the model at its flying speed, which is really quite slow.

The slow flying nature of the model makes every flight relaxing. You can make tight rudder turns, fly close by, perform touch-and-go in confined spaces. The model responds well. At glide, the slope ration is perhaps seven to one and at something like jogging speed. F949 will stall, but recovery is quick and flying it smoothly will make those occurrence rare, especially if a bit of height is traded off or when the flying speed is increased by a bit more throttle in the first place.

I have been flying in the calm mornings. Even so, I have seen the model's wings rocking which I think is its reaction to slight air turbulence. F949 has a narrow and low speed range. It flies slowly but cannot accelerate to high speed flights. I suppose I could get the model to outfly medium gusts, but that couldn't be its design intent.

All considered, this is a good beginner's plane for calm weather because it is cheap, flies slow and long, and is forgiving in nature. It practically flies hands-off!

2S Li-Po Hot Wire Polystyrene Foam Cutter

2 October 2018

My 'scroll saw'/'jig saw' type of hotwire cutter was simply the frame hotglued onto  a carton box. Yeah, well, it is easier and faster.

Then after playing for a bit, the guitar string broke and things went on hold until this arrives from Banggood yesterday:

Wire replaced, and plugged to the same 2s battery, the Kanthal wire sagged in the frame.
I took up the sag with the winged nut, and when cool, it was strung very tightly.
I don't know if this wire will be even shorter lived.
It seems this Kanthal wire runs hotter (that means I can lower the voltage? No point since I'm on 2S).

Physically, the wire is somewhere between the soft steel wire and the guitar steel wire. I think it is good enough to make pushrod wire ends. 

Next update will be for some control on the heat using these:

The female plug fits the adapter.

Don't like the smell of polystyrene. It seems this time it burns more polystyrene than the last time. Maybe that's because the cutting wire is now thicker.

12 September 2018

This setup limit my cut to about 10", but it should be enough for my purpose most of the time.
I am just glad I didn't buy a variable voltage DC power supply unit (about $100, 0-30V, 0-10A) because I have no alternate use for that and that will become clutter.

Here is the working combination. I tried it yesterday and all that matters is it worked:

Assembled from:

• Frame: Hacksaw frame (the tension of the wire is easy to adjust)
• Power: 2s lipo battery with balancer plug
• Wire: Fine Guitar string
• Miscellaneous:
• A matching battery plug (I don't like 'wriggly-iffy' contacts)
• Battery/Cell Checker that is plugged to the 2s lipo battery (to check that the lipo will not be discharged too low while under use)
• Bit of wire and soldering (for got electrical conductance)
• Velcro strap to secure battery to frame.

Apart from the guitar string and perhaps various bits which may not be re-usable or recycleable, the rest will revert to their original state and function on disassembly.

Transform it into a 'scroll saw'/'jig saw' type

Adapting the 'free-hand' hot wire foam cutter will need additional material:

1. Secure frame to edge of table at desired angle.
2. Small metal piece, perhaps a short length of wire or a useless key, can be fixed to the point where the wire touches the table will prevent the wire from burning the table's edge.

The table will double up as the foam work platform. The foam sheet rests on the table top as it is fed towards the hotwire. Although there is no platform extension to support the 'free' end, it should meet most of my needs. If I have to, I can go around this limitation by cutting half way and recommence cutting from the other end. To make 45 degrees cut etc, just secure the frame at the correct angle to the table.

This mode allows:

• consistent angle to cut edges
• using only 'top/single' template
• identical copies by cutting multiple sheets in one pass
• cutting in a straight line without template (use guides for foam sheets)
• making parallel cuts
• cutting cones and discs (think spinners and wheels)

Why do I consider carving foam planes?

1. Rounded or compound curves
2. Lesser glue is lighter
3. Lesser mass for small fuselages
4. Lesser work for small fuselages
5. Cheaper if foam is purchased (even though Popular selling polystyrene sheets at $10 which I think is strange)
6. Free if foam is sourced from throwaways
7. Strength and hardpoints:
1. tissue/paper/cardboard
2. wood/plastic rods/strips/pieces
3. glue/epoxy/tape
4. carbon fibre rods/strips
5. music wire
8. Surface Finishing: 
1. Bare (or with some sanding)
2. with markers or paint
3. Cover with tissue/paper/cardboard, markers and paint

SAFETY HAZARD

For safety's sake, it may be a good idea to do all foam cutting and sanding outdoors.

• heat can result in burns or fire
• toxic fumes

15 April 2016

(Was: Supplying power to Hot Wire Foam Cutter using a battery charger; Micro receivers and ESC)

My battery charger can power my hot wire foam cutter. This is the original IMAX, the one without the hot wire supply feature. I used it to charge my lipo batteries, thinking that NiCd, NiMh, Pb are legacy features and useless for electric RC. Glad I found a use for it, because now I charge my 1S lipo with a USB charger.

The charger was set to NiCd charge mode and between 3A to 3.5A. If I set any lower, my hot wire cutter doen't cut foam, whereas if I set it to 4A, I find the hot wire cutter goes through foam too quickly. When I set to low 1-2 A, the voltage displayed was about 2 V; 3A:8V; 4A:10V.

Whilst in use, the charger would sometimes go down to 0V and I have to wait with the cutter in the foam but not moving anywhere. I can continue only when the charger goes back to charging. Charging was terminated once when the charger thought charging is complete. The alligator clips, the pins and connecting wires were warm at 3.5A.

My hot wire cutter is a hacksaw frame, it is not all-metal, the handle being plastic. I like the winged nut very much, because I can attach/detach and tension the wire easily. The two connection points are at the handle so I don't need to cut my leads.

I used the thinnest guitar wire because it is suggested on the internet and is available here. I strung the wire tautly, until it gives a high pitched sound when plucked. I like to think that doing this prevents the wire lagging behind when cutting foam causing scalloped cuts. The cutting length of my hot wire cutter is short, but it is convenient to use, and when not using, to store out of sight.


My hot wire cutter can cut through denser blue foam and the kerf is small as can be seen in this intricate jigsaw test piece in compressed foam/depron.

14 April 2016

I was de-cluttering my room the night before last ("Need Space") and found some small lipos. I have had them for some time now and wondered if they are any good. So I connected them to my blue IMAX copy charger, the one with hot wire output, and after a few lipos, smoke came out of the charger. Too bad but I have to threw that charger away.

My two other chargers do not have this hot wire output feature. The last time I tried, using Pb and Lipo didn't work, but this got me thinking that there must be many others like me attempting to supply power to their hot wire cutter using their chargers for that purpose.

I did an internet search and it seems that you can do so in NiCd-Charge mode. An article by Flite Test wrote that the author did just that by setting up the Amp at 2A and it was about 10V.

I think I will try that with my remaining IMAX charger.



Tito brought this small 0.6gm Flysky receiver to my attention. I don't know if I should choose the 8 channel or the 7 Channel PPM version. I know my Flysky transmitter is 9 channels, but I don't need so many channels, however the board is cheap, about $15.

Using this board (instead of the all-in-one WLToys receiver board) means I need an external esc for motor control.

I am ambivalent to the choice of brushed or brushless motors. My criteria for choice is availability and affordability. So what if the 8.5mm coreless motor needs replacing? So far I haven't needed once. My main consideration is whether I can find small ESCs for my purpose.




From Banggood, this is the smallest and lightest brushed ESC that I can find. It is only 2 grams. If I were to use this, say to power multiple brushed motors since it is 5A rated, I still have a big problem here, I can't imagine using this car/boat type of ESC on a plane because I don't want REVERSE!  I asked around my friends for a workaround. It is possible if you have a programmable transmitter, nobody has a simple solution to modify the ESC.

The smallest and lightest brushless ESC I have is 3A from Hobbyking and weighs 1-2 grams, but I am wary because I have already burnt out one such ESC. And it wasn't cheap.







In the quest for cheap and good brushed esc I did an internet search and found an article posted in 2002, the author modified a servo board and used mosfets and resistors to make an ESC: http://trebor69.tripod.com/speedcnt.html
Now this seems to be ideal because it is for forward use only as highlighted by the author. I hope it is proportional though and not a switch type.

I found my Walkera V120 helicopter has brushless main and tail motors, it has two ESCs to power the motors. I have not 'touched' the Walkera helicopter for years, because my helicopter flying skill then was quite bad and stopped trying when I got my first WLToys V911. Maybe I should try again? I remember I paid about $200 RTF, this price included a fancy transmitter.

For convenience though, I would use WLToys' components.

Heat Shrink Sleeves for Canopies

1 October 2018

I'm excited by the prospect after seeing some of Maxflier's youtube videos in which he formed canopies by shrinking those transparent heat shrink plastic sleeves (those for lipo batteries) over a male mould.

Wow, no one-off 'die' plates, no vacuum box and frame, no heat gloves! It seems all it took him was  just a single male mould stuffed into a heatshrink with some packing pieces. Simplicity and effective for small and light canopies!

His free flight balsa planes are good lookers and performs well. He used trussed spars for the wings and printed tissue for the covering. I enjoyed his videos and will watch more

Volantex Ranger 600, reversing servo throws

24 September 2018

It's gone now, so I write.

I bought one from Rotor Hobby. I didn't get used to the self stabilizing modes, it seems more like a hindrance to trimming so I used the 'expert' mode.

One adventure it had was hanging on a low branch of a very tall tree and going through strong wind and rain. I went back in the late afternoon and found that the plane nesting at the foot of the same tree.

It's final adventure was when it was in good lift. I was flying it high and the wind  was apparently very strong up there. It went out of range vertically. The batteries in the transmitter were newly changed, so it is again my fault for forgetting I'm holding on to a 'toy' transmitter and not a regular one.

I had to change all 4 numbers of AA batteries in the transmitter because they were flat. And they got flat because they were accidentally bumped on (I had the plane and transmitter in my car). As the batteries were on low power, it gave out warning beeps which puzzled me when I was driving, until I found the source of the beeps.

On the day of the final adventure, I crashed it immediately on launch because the rudder and elevator throws were reversed. It was puzzling, there being no switches or other buttons on the transmitter or on the receiver. Adrian's 1st suggestion was to remove the back sticker, his 2nd was to remove the front sticker, his 3rd was to keep pressing the trim buttons. I didn't find anything on the manual or the internet.

Finally I discovered that all that was needed to reverse the throws was to move the affected joystick of the transmitter to its extreme position and holding it there until a beep emits. Direction of throws sorted out and trims reset, I found that the elevator servo has very limited range of throw. It was only about half the angle throw of the rudder servo. I don't think it came like that, but regardless, I flew it and it felt fine enough. It was on the 2nd battery that the plane went on its last journey.

The Volantex Ranger 600 was a good sub-$100 RTF with concave airfoils. I know I was in thermal because the plane went up without power. I didn't care about the self-stabilizing switches even though I tried them and they do make the plane fly more smoothly. The disadvantage of having this 6G stabilizing mode is I feel I couldn't see the plane reacting when it crosses any air activities. I want to be alerted especially when the plane come across some thermal activity. It is not important to me whether or not the wings dip this way or that way, because my only reactive tactic is to fly straight towards the wind. I sure have much to learn about thermal gliding.

Pull-Pull for weak structure and hinges, but it will be heavier

27 Aug 2018

Here is an idea of a form of indirect Pull-Pull control system for weak structures and hinges. It is more complicated and it will be heavier, but it minimizes the stress forces to structures and hinges. It uses pushrods from the servo arms to operate pulleys (or pulley strings) and at the rear part of the pulley system, it can use pull strings or pushrods to move the control horns. It is always a balancing act, depending on what I need and what I am willing to forego. Afterall, a simple wire supported in a few spots with tubes might already be the control solution if the slight weight gain is acceptable.

I have done a couple of unsatisfactory pull-pull for 1s RC (WLToys' receiver servos board). Pull-Spring is not optimal because there was a constant drain due to the servos having to resist the spring. Pull-Pull was tried on that large winged stick motor glider and I had to brace it with additional tie strings because the pull strings exerted enough force to bend the long and slender fuselage boom stick. There was also that idea of converting a single horn into a pull-pull by gluing on a pulley to the front of the horns which seems pretty neat and light, except that it could still bend the fuselage and it requires better hinges than tape as all the forces are exerted on the hingelines.

At the core of the proposed indirect Pull-Pull system is a central core which supports the load of a series of pulleys. In this way, the stress is isolated and the aircraft structure's slenderness is reduced. Servo arms' pushrod could be heat-shrunk to the pulley lines, and the output could either be control pushrods similarly heat shrunk to pulley lines, or pairs of pull strings tied to pulley lines.

Small Gliders: HLG; Whitehead 21; F15SG

10 August 2018: F15SG in RSAF 50

Print out the F15SG, paste on foam board.

Cut out the pieces, insert icecream stick to fuselage, cut and cover the other sides.

Glue the fuselage, plan form, vertical tails together.

Far end is one without foam. All has notches on ice-cream stick.

Prior to this I made some HLG fuselages from ice-cream sticks and bamboo dowel. I did a 3rd HLG and found that the nose is much lighter than the tongue-depressor' type and had to add blob of hot-glue at the nose of the ice-cream stick for balance. This 3rd one was given away before test flights.

For the two previous HLGs, I managed hard launches and catapult launches and they behaved. I was a bit concerned about the flexing bamboo boom (i.e. the one with the curved dihedral) , but doubted if I can detect a difference between that and the stiffer boom. Both gliders' left wings tore up on launching; the first was the stiffer boom straight dihedral model on hand launching, the second followed when I catapulted it. Both wing panels have been hot-glued back. Ready for more flying. Flight pattern is rolling left on launch and left glide at top.

25 July 2018

A few weeks ago, I was thinking of making simple the wings for a Whitehead No. 21 inspired model.

The top wing is compound curved by slitting and the bottom wing is by arching.

Then I played with the idea of using 1.8mm bamboo dowel, bit of foam and plastic bag for the bottom wing. Many thoughts were entertained, on the possible failure and fixes that may be necessary. Then I thought that perhaps I should just make the wing from 2mm foam sheet as that should exclude a lot of vagaries. Forget about compound curve or slitting.

Then the shape reminds me of a chuck glider and I set about doing HLG.

Considering my current physical status, This is what I came up with.

I folded a piece of A4 paper (to get the centre line without measuring) and drew free-hand the wing and tail elements.

I traced the cut patterns on a 2mm foam sheet and cut out the wings and tail elements.

I found that the 2mm foam sheet has 'grain' due to the manufacturing process. The first wing was cut with the grain running chord-wise. This resulted in a floppy wing. Never mind though, since I am going to make a light glider, I just curled the wing in my hands for a curved dihedral. The second wing was cut with the grain running span-wise and this resulted in a stiffer wing.

For the straight dihedral wing, I used a piece of 9mm x 3mm wooden section (from Daiso), snapped it approximately in half, tapered the rear end and left a long nose because the CG of the fuselage is not forward-centric, not enough anyway. I will need some noseweight. 

For the curved dihedral wing, I hotglued a whole length of 1.8mm bamboo onto a tongue depressor that I got from my doctor. The resulting CG is much more forward. In fact, it seems to require no additional nose weight as the model glides well. A shark tooth was cut so that it can be a catapult glider. I figured I don't need a finger piece because I can grip/rest/push on and against the tongue depressor.

Maybe the straight dihedral wing's fuselage is a failure, it is however, conventional and stiffer. As for the curved dihedral wing's tongue depressor bamboo combination fuselage, it is lighter, easier and faster to make, the CG is there so it will result in a lighter model because nil or less nose weight will be needed. but, it seems flexible. Tossing it against the bed's head board demonstrated a very slow floaty glide with good glide gradient. Perhaps flutter wouldn't be an issue, given my physical status.

Looking forward to their test flights.

Construction notes:

1. Foam pieces cut with a NT cutter, no sanding, leading edges are 'rounded' with thumbnail both sides and trailing edges are 'tapered' with curved end of steel ruler.
2. Straight dihedral is cut from bottom partway, cracked, and then glued to fuselage.
3. Wing is glued to fuselage first, then the horizontal tail is glued slanted to encourage left turn, and vertical tail is glued straight.
4. Hotglue used throughout. The nose blob of the straight dihedral wing's model is an attempt to use hotglue as safety noseweight; not effective because hotglue is too light, maybe I can glue a paper clip for catapult launching.

Temporary Writeup on model for Whitehead number 21

11 July 2018

On wing construction using 1.8mm leading edge spars and polyethylene film:

1. Prepare paper template by first printing out the planform of the pair of wings. Fold it in half and fit it into the plastic bag. Tape to temporary secure the paper template.
2. Follow the lines on the template and draw on both sides of plastic bag using indelible pen or hot glue gun, the radial spars.
3. Glue the 2 root pieces and 2 leading edge spars to the bag.
4. Cut the bag as indicated on the template. The leading edge shall be rolled over the bamboo spars.
5. By curving the root pieces, the plastic film will be curved into an airfoil.

On fuselage construction using 2mm foam sheet:

1. Cut long rectangle piece of foam, put tape midpoint and fold in half.
2. Mark and cut out the required fuselage side length. The 2 remaining ends can be used for internal fuselage formers.
3. Cut one such piece to the width required and glue the top mast to it. The top mast is a length of 1.8mm bamboo dowel. This former assembly is used to space out the folded piece. Secure free end of fuselage with glue and tape.

Joining wing to fuselage:

1. Apply UHU glue to the areas where the root pieces of the wings will come into contact to the fuselage.
2. The free ends of the 2 leading edges are trimmed to length so that they can be turned up (wing turn down) along the fuselage former, against the top mast and forming the required dihedral angle.

Bowsprit, tail, wheels, rigging

1. Fabricate and install the 2 wheel pairs to formers and glue to fuselage.
2. Bowsprit is a length of 1.8mm bamboo dowel, glued to tip of fuselage and top mast (through the first former).
3. Glue remaining formers, clearing bowsprit.
4. Construct the tail using 2 pieces of 1.8mm bamboo dowel leading edges, spacers and plastic sheet. Set the decalage and glue to fuselage.
5. Glue motor mount to fuselage and then glue the twin motors to motor mount. Glue the control unit so that the CG balances at the desired place (test glide).

Flying

1. Test fly. Change CG location, decalage, thrust angles, vertical surfaces etc., wherever necessary, for powerless glide, power on trim, and turning trim.
2. Rig with cotton thread and add details and painting.

Temporary post on Taiwan driving trip.

Taiwan's driver's position is on the left side and Singapore's on the right. I had experienced driving in Germany and it did not posed any problem but this trip is different. I rented the car for the whole duration of the trip and planned on driving a lot, through hills, mountains and coast. And not primarily on auto-bahn.

After collecting the car, I got lost. The first was because I wasn't aware of the right turn out of the carpark and across the yellow box. Didn't think one would entered a junction by performing a right turn, so I missed it the first time and had to circle the airport compound. I have the google map turned on, but that did not seem to help at all. The second was when I was out on the road and somehow I missed a fork, resulting in driving across what seems to be underpopulated sub-urban districts.

Google map took a while to get used to. The instructions were corrected very quickly, so quickly that I didn't noticed I was already on an alternate route.

Driving on the main road was fine. Go with the flow, and there isn't much problem. The problem I had were because I was used to driving the other direction. Many times, my car would be positioned too far right in the lane, sometimes, I could hear the right tyres running on the lane markings. There are little humps at the beginning of each dotted lane marking to provide audio and ride feedback.

Driving through single lane winding roads was challenging but not an issue. The speed limits imposed are safe and there are taking over bays for the slower vehicles to drive on and allow the faster vehicles to overtake them.

Temporary post on Hyundai i-30

11 July 2018

Registered on 26 June 2018.
Car received, after a long delay sorting out the Qi charger, on 3 July 2018.
It is no Mercedes but it is fine.
From last thoughts: Where's the manual to the GPS/DVD and the blind spot detection on the wing mirror?
No emergency triangle or First Aid box.

Just WhatsApp Terrence.
He said manual given liao. With the 3M's warranty.
Asked pdf, he said he will get.

Office has used up allocated number of car lots. Additional lots, such as my case, is subject to the office building's management office decision.
While it is convenient and conventional wisdom to park at office, due to the lots constraint and perhaps also the cost per month (220+), many people who work in this building parks elsewhere. And they are usually cheaper, maybe 180 per month.
There are coupon parking near office, 60cents per half hour starting 8.30am.
So to convince myself, I did this calculation.

Find number of days that I would be parking at the office.
365 calendar days in a year less 52 Sundays, 52 Saturdays, 11 public holidays, 21 days leave, 5 days MC/outstation/jobsites.
365-52-52-11-21-5=224

Find the coupon cost per day.
8.30am to 6.00pm is 9.5hours.
9.5x2x0.60=$11.40
Since I use parking.sg, I get refund if I don't park that long.

Find coupon cost per year.
224x11.40=$2,553.60

Yes, it will cost me about $2,500, it is that expensive, but compare to buying season parking at the office, 12x 220 = $2,640.00, it is lesser. Coupon parking is at roadside, office parking is at basement, when it rains, obviously office parking is preferable. On dry days though, if I were to park at office, the additional distance and time to go home are: to drive down to its basement to park, drive out from its basement to leave, making a U-turn, against the walking distance and time from coupon parking. It is close, but coupon parking wins.

Now, comparing coupon parking against season parking at another further location (URA season parking not available):
12x180=$2,160.00
I will therefore save 2,553.60-2,160.00=$393.60.
I have to work a distance of about 224x2x0.6 km=268.80km in around 224x2x8min=59h44m
$/hr is $393.60/60h=$6.56/hr
However, this saving potential but may be offset by the $0.50~$2.00 ERP I might incur when going back.
If ERP is $0.50 (i.e. 6.00~6.05pm), that's 224x$0.50=$112.00
If ERP is at $2.00 (i.e. 6.06pm onwards), that's 224x$2.00=$448.00
This going home ERP cost is currently avoided because I leave the office earlier than 6pm (I compensate for this by arriving earlier in the morning), but there has to be some limit as to how much earlier I can swap without incurring remarks and comments.

30 April 2018

This is to record my thoughts and the promises conveyed by the salesman of Komoco, culminating in the 15k deposit placed for one "i-30 Wagon" yesterday, 29 April 2018. It will probably be more than a month before they will call me up to collect the car. Even though traditional wisdom would say I should have everything he promised written down and signed, we did not go for this route because we do not wish to be stressed by all these and have an enjoyable experience turned to one that reeks of distrust governed by fear.

In anycase, what comes with the 'standard' I-30? The features and accessories are not listed, so what are they selling and what am I buying? Listed below the accessories/features that salesman, Terrence Soh, said will be incorporated into my I-30. He promised it's the best he can offered, in terms of price and features/accessories.

• Steering column and be adjusted longitudinally. This came when I was test driving the Accent. which lacks this feature.
• Leather wrapped seats but not electric/memory. Although the seat is not electrical/electronic, the general ergonomics cannot be that bad to drive long distance. I have in mind the Ford Focus I drove in Taiwan.
• Auto opening/closing wing mirrors (indicator and blind spot detection).
• The screen will be changed to 9".
• Comes with GPS, dvd, rear parking camera, front/back camera, sensors (forgotten to ask about front), stability, abs, front ventilated disc and rear disc brakes, 7 airbags, pull out cover to the wagon boot, cargo net. Wife likes the two roof rails, calling them angel wings. Terrence assured everything is complete, solarfilm, diamond coat, sound clad.

Wife chose this car, after going through Citroen C3, Mitsubishi Attrage (zero interest), Kia K3 (lack of interest), Hyundai Elantras (a midgen bit of interest), 1l Kona manual (too small),  Accent (zero again), Santa Fe (like but too big), Tucson (big but why?).

Terrence said the chosen 'moon-rock' is not in the immediate shipment. Choice of colour, externally and internally is important for her. Since they can't deliver 'moon-rock' (2nd shipment coming in July/Aug), wouldn't sell the 'moon-rock' car at the same price when it becomes available (post July has 10,000 VES scrapped), wouldn't sell the exhibit 'moon-rock' car (suggested by wife and he declined because his senior said so), an option should be the refund of deposit, not just choose another colour which Terrence suggested. He suggested white, which was clearly rejected by wife just yesterday, black and silver. He forgot that my wife does not like the two tone internal. Out of the available colours, micro grey and silver have 2 tone interior.

Wife asked, White or Black? "Black, same as yours." I replied, and some time later she said we go to the showroom tomorrow to choose.