Wednesday, 20 March 2019

Checklist

I thought I would share my checklist for G-FUUN - which is type specific but generally applies to any fixed gear Twister with a UL engine.

Feel free to download and print out - it is set up to be A5, a good size for knee boards. PM me if you want the original PDF - I cannot upload it here as Blogger only allows images not files.

Strike out the cowl flap references if you do not have one.

I personally don't use this checklist anymore having committed all of this to memory but if you are starting out it would be wise to have a Checklist.





Sunday, 3 March 2019

Twister Pilot Notes

The following is meant to help anyone who has yet to fly their Twister and should not be taken as any kind of instruction - it is merely a guide.

As I've only ever flown G-FUUN this is type specific - your results may vary :)

(This is for a fixed gear, integrated tailwheel Twister with a UL260iSA engine.)

How I fly a Twister.

Preflight:

I start my pre-flight inspection at the prop, checking it for any damage and general secureness, I look underneath the cowl for any leaks and at all 3 tyres to see they are inflated correctly. 
Then the oil level is checked, then all cowling fasteners.
I then visually check the contents of each fuel tank. 
After that I walk around the wings running my hand along the leading edge to check for any damage before stopping at the pitot to check it is clear and the ailerons to check they are operating in the correct sense and are free. I also check the aileron linkage at this point. 
Proceeding back along the fuselage I check the static vents are clear and then that the fail safe screw for the tailplane attachments is in place. The elevators are checked next for integrity, full and free movement and no play.
Next the rudder is checked, that both pins are in and both cable attachment bolts and nylocks are secure, also checking for full and free movement.
The walkaround continues to the other side of the aircraft with the same checks as above where appropriate.  

Start: 

After getting strapped in and applying the parking brake the canopy is closed and locked (you should never even start a Twister, let alone run one up with the canopy not locked down). Throttle is set to just above idle and all switches are set to ON from left to right. A brief pause to check you have correct fuel pressure and then looking outside for any people who may have strayed close to the prop before shouting "clear prop". Starter is then engaged. Oil pressure is monitored closely and must come up in 5 seconds or less or shut down. Battery voltage should also begin to rise.

Checks:

Whilst warming the engine I perform all pre-take off checks (bar one). Controls, full free and in the correct sense (the most important check in any aircraft). Straps, secure and any loose ends tucked in. Instruments, set altimeters to local QNH, turn on radio and transponder. Flaps, set to first position, 10 degrees down. Trim, set to slightly forward of centre. Engine, check battery voltage is now 14.5, then check each ignition coil in turn by turning them off one at a time there should be a small rpm drop of only about 20-50 rpm each time. Oil temps coming up, I taxi when it is 35 degrees using only low revs, no more than 1,500rpm.

Taxi:

The view for me over the nose is possible but a shorter pilot will not be able to see over the nose so you should use the Pitts method of taxiing in a series of left and right curves. The turning circle on the integrated tailwheel Twister is quite poor so you need to take that into account before taxiing into a confined area, this takes up quite a bit of your attention when at an unknown airfield. Braking won't help tighten up a turn either as the standard brakes act on both main wheels simultaneously. If the surface is bumpy then it's best to hold the stick hard back to ensure the nose does not tip over.

Take off:

Final check before take off is a full power static rpm test after first checking oil temp is at minimum 50 degrees C. Depending on the outside air temp and wind direction I get between 2,500 and 2,600 rpm. 
Once the take off roll has begun do not be tempted to raise the tail too soon, wait until you have 30 knots or so and have good rudder authority, the integrated tailwheel will help you greatly to keep straight but not when it is in the air! Once the tail is raised your view ahead improves greatly and the usual footwork will be required to keep straight, at this point I glance down to check that my ASI is live, when you get to 50 knots things will become light and shortly after it will want to fly off. Let it accelerate in ground effect until you have 60 knots before easing it up into a climb attitude.

Climb:

On the initial climb out I leave the flaps down at 10 degrees and climb until I have about 300 ft or more before raising them. Once raised the Twister will naturally assume a lower angle of attack and the speed will come up.
At this point I let it accelerate to 90 knots before turning onto the crosswind leg. I continue to climb at 90 knots, this gives 1,300 fpm in the Winter (and as little as 500 fpm in the Summer when hot, heavy and high). There are many good reasons to 'cruise climb' at 90 knots rather than try to climb out at a much slower speed. Firstly you have a much greater margin over the stall giving you much more time to react to any emergency, Secondly you have a better view over the nose for possible conflicting traffic, Thirdly the climb rate is no worse at 90 than it is at 70 because with a fixed pitch prop the engine is not able to rev out (and therefore produce more power) when you fly it slower, Fourthly when you are actually trying to go somewhere then a faster climb speed will make your journey time less, and finally, Fifthly, the cooling is better for the engine when at a higher airspeed and a lower angle of attack.
Caveat to this; my Twister is probably the heaviest one out there and I may also be the heaviest Twister pilot(!) so your combination may climb better at a slower speed, but I have found this is the best speed for me.

Cruise:

I climb above my target altitude by 100 feet or so then shallow dive down to the target altitude to accelerate quickly to cruise speed and get on the 'step'.
I typically cruise at 2,700 rpm giving a TAS of 125 knots with a fuel burn of just under 15 litres per hour. 
If I'm just flying around at low level for an evening 'bimble' then I'll reduce the revs to 2,600 or even less.
I change fuel tanks every half hour to keep the wings balanced.

Descent:

The Twister is a slippery beast so it is best to plan well ahead with your descents. I tend to reduce power a little and then head down at max rough air speed - which is 120 knots indicated.

Circuit:

The more familiar I am with an airfield the tighter I will fly the circuit. I tend to fly the downwind leg quite fast - at 120 knots indicated and then go below circuit height on the last bit of the downwind leg while setting the throttle to idle. Then as the speed comes slowly back I will raise the nose by quite a lot to slow it down further until the flap limiting speed. If you don't raise the nose it would take forever to slow down to approach speeds. Flaps come down in stages. Only use 20 degrees of flap if you have a strong crosswind and plenty of runway length available, otherwise I always use full, 30 degrees, flap.
If you're starting out then fly a big wide circuit and only go at 100 knots until you get used to everything.

Approach:

A steady approach at 60 knots is what you are after, nail that 60 knots and make sure you carry that speed all the way until the roundout. The Twister side slips quite well, so if you are too high then throw in a side slip to lose height.

Landing:

A minimum runway length of 500m is a good idea for a Twister - certainly in the early days of flying one. (In nil wind I personally would not try to land anywhere that was not at least 500m)
The Twister floats quite a bit on landing but don't try and slow down your approach below 60 knots to land shorter you will only end up running out of energy to round out and hold off if you do. Stick to 60 knots and you'll be fine, just accept the long hold off period until touchdown.
I virtually always 3 point for a landing. What you are trying to do is come down very slowly in the last foot or so above the runway so that you touchdown without much vertical speed at all. If you have a sudden drop to the runway it will bounce, the undercarriage does not have any kind of damping effect and is more like a spring so you end up going back up to where you came from. If this happens then hardly anything needs to be done, just be patient and let it come down again gently with a very small back movement on the stick at the last bit (do not pole forward!). Unlike most other aircraft you should not stall a Twister onto the ground as you are landing, the reason for this is that it has a sudden wing drop at the stall (for me it's the right wing) and if this were to happen with any height above the runway you would come down very quickly on one side. It has happened to me once and it was not nice.
Due to the Integrated tailwheel the Twister is a pussycat of a taildragger. As soon as that tailwheel is on the ground you have good directional control (unlike most other taildraggers). So that's why 3 pointing is a good idea, even in a crosswind. I have done some wheeler landings (where you touch down on both mains in a level attitude and then lower the tailwheel later on) but to be honest they are no better than a 3 pointer if you have the integrated tailwheel.


Aerobatics:

So you're ready to have some fun then? Good stuff. Here are a few things I've learned while aerobating the Twister.

Safety first: Do it high so you don't die. If you're trying anything for the first time then do it high - minimum 3,000 ft or preferably 5,000 ft.

The main thing to watch out for in a Twister when doing aerobatics is speed control when you are heading ‘downhill’. The Twister accelerates much faster than more draggy airframes so if you are not careful it can get away from you a bit. If in doubt always roll out of any manoeuvre rather than trying to pull out (if you are upside down) if this is happening.

The Elevator has a nice feel to it, the Ailerons are heavy (spades are needed really) and the Rudder is much too light and so quite sensitive.

The roll rate is pedestrian - 90 degrees per second - which is very slow for an aerobatic aircraft. Also it has a low power to weight ratio which makes it lose energy going 'uphill' quite quickly. Conversely as it is so slippery it accelerates and gains energy going 'downhill' very quickly (as cautioned about above). 


120 Knots is a good all round entry speed for most manoeuvres.

Loop, Stall turn, Half Cubans etc.

A big input of right rudder over the top of any Looping manoeuvre is needed to stay on heading.


The Immelmann requires a higher entry speed, 140 or even 150 if you can manage it as the Twister loses so much energy and has a relatively slow roll rate at the top.

Any other figure which requires a long up line - first half of a Goldfish or reverse half Cuban for instance - also require a higher entry speed - 140 is good.

Snap (Flick) rolls should be entered at a much lower speed, well under Va. 80kts is fine for those.

Normal aileron rolls and hesitation rolls can also be done at a lower speed. Around 100 to 110kts or so gives a clean look to those.

Stall turns should be to the Left. Once the nose has passed the horizon on the first half of the turn then I go to idle with the engine. This helps in two ways. It lets you make a long down line without gaining too much speed, it also gives you much more right rudder authority to 'stop' the turn on a vertical downline with the wing leading edges parallel to the horizon, it also reduces the torque effect of the engine.

With spinning I prefer to spin to the left as it gives more control and enables you to stop it on a heading better (more right rudder authority at idle). Spinning to the right it happens quicker so is harder to stop on an exact heading, normally you should start recovery from a spin 1/2 of a revolution before you want to stop to get it to come out on the right heading.

Enjoy your Twister and be progressive in exploring it's limits (and yours).

Eye Candy Part 2

Some more pics that others have taken.

At the Sywell Rally in 2018

At Breighton for the Club class aeros comp.

Parked up in Reykjavik Airport, Iceland on my way across the Atlantic.


Parked up next to Johan's Twister at Sanicole for the EFLEVA weekend.

I made FlightRadar's Year in Review with the Spitfire outline drawn over Southern England as my tribute to the RAF100 anniversary.

A more detailed view of the path I flew. The coverage is not perfect hence the jagged lines here and there - the actual SkyDemon path flown was much smoother than this.

Landing at Lee-On-Solent in October 2018.

At Oshkosh in the flypast.

At Oshkosh, and featured on Kitplanes 'Guess what' section of their website.

Over Lake Tahoe as part of the Kitplanes early morning photoshoot.

A 'break' from the photo ship.
The best shot from the Kitplanes photoshoot. I like the reflections off the prop disc.

Close up from the Kitplanes photoshoot.
Landing at Conington during the Aeros comp.

Thursday, 28 February 2019

Fuel overflow

When upside down the fuel overflow outlets are in an area of low pressure and so vent out a lot of fuel.

This has been a problem since day one and I've only just got around to a long term workable solution.

My idea was to restrict the size of the outlet hole which as standard is quite excessively large at about 6mm.

For a while I was just using a piece of duct tape over the outlet with a small hole pierced in it but that did not last very long as the fuel soon got to the sticky part of the tape and made it fall off. I needed a more permanent solution to the problem.

So I made up two small discs out of thin ali sheet and drilled them in the middle with the smallest drill bit I had - 1mm.

These were then glued to the outlets using 5 min epoxy.

A recent flight test proved that these to work well - there is still a tiny bit of fuel leaking out during and after inverted maneuvers but it is hardly anything compared to before.

I never fill my tanks till they are overflowing so there is plenty of space in the top for expansion should it sit in the sun - the fuel can still flow out of the outlets at a reasonable rate anyway with the 1mm hole so I don't think expansion is going to be an issue.

Fuel venting out whilst upside down. This is before the fix.
Ali plate restrictor with 1mm hole drilled in it.


Thursday, 21 February 2019

Wheel pant mod

With the fixed gear Twister it is impossible to pump up the main gear tyres while the wheel spats are in place. You even have to remove the wheel pants just to check the tyre pressures, which is quite impractical.

So I decided to mod the pants by making a small panel which is easily removable.

This panel allows enough access room to both check pressures and pump up the tyre if necessary.

I used 3 layers of carbon for the flange with small patches to make 4 layers where the riv-nuts were to be located.

The panel is flush and the two small M4 dome head Allen screws don't make much drag.

Size and location of the panel. Tape was left on till I painted the flock (which was used to fill the cut gap).
View of the flange and riv-nuts.

Once finished it is hardly noticeable and adds very little drag.


Monday, 18 February 2019

Cowling outlet experiments

I long had the suspicion that my outlets on the cowling were too big. 

I erred on the side of caution when I was making my cowling and made both the inlets and outlet quite a bit bigger than the calculations in case things did not work as I first thought, it being easier to make them smaller than make them bigger afterwards being my thinking.

Until recently I had not bothered to restrict the outlet to see if it made much difference to temps but finally got around to doing that this Winter.

For the first trial I blocked off approx 50% of the outlet (the lower half) using a piece of foam pipe insulation. This didn't seem to make any difference to the temps at all.

I then tried blocking off approx 80% (some of the side outlet being blocked off this time). This did make a difference - I saw a 10 degree increase in the CHT's with a slight rise in oil temps (as the cylinder heads are oil cooled - when they run hotter the oil does too).

On the second test the hottest CHT in the cruise was 99 degrees or some 60 degrees below Max continuous. It was 115 degrees in the climb. All CHT temps stay within 3 degrees of one another (thanks to my 7 odd iterations of baffles on the front cylinders).

These tests were conducted at 5,000 feet with an OAT of approx 2 degrees.

It goes to show I went a bit overkill on the inlet and outlet sizes when building the cowling and it also shows that the sealed plenums work very well.

It also shows that the lower outlet area doesn't do much - it being in a high pressure region does not encourage flow out of the cowl. However the side outlet area works very well as they are in a low pressure region which means the low pressure air coming out of the cowl is more likely to flow to this region.

The beauty of the foam pipe insulation is that it is very easy to remove when the weather gets warmer.

I've also conducted some True Airspeed Tests to see what the exact cruise is at certain RPM's. I used to guesstimate at 125 knots True when light and 120 knots True when full of fuel and baggage (when touring).

I flew 4 way compass headings then added up these speeds (from the GPS groundspeed) and then divided by 4 to get the average. At 2,800 rpm (max continuous) at 5,500ft this resulted in a speed of 129 knots True.

I then flew the same 4 way test at 2,700 rpm (my usual cruise setting) at 5,500ft and this resulted in an average of 125 knots True.

Flying at 2,800 burns around 17 litres per hour but at 2,700 I burn just 15 litres per hour so that is a better rpm to cruise at for economy (and the engine seems to like it better too).

In hotter weather it will not cruise quite as fast due to reduced power from the higher inlet air temp for the engine and the extra drag from the open cowl flap - which I have found adds about 2 knots to the cruise from fully open to fully closed.

Approx 80% of the outlet blocked off with foam pipe insulation. The first test was without the small additions of foam at the top on the sides.

View of the left hand side. The approx 20% remaining outlets are more than enough in Winter to adequately cool the engine.




Monday, 4 February 2019

Thursday, 31 January 2019

Canopy repair - again

In December the Twister had it's annual inspection and after sending off the paperwork I did not get anything back through the post - the LAA normally turnaround a Permit the same day if all is in order.

So after a week I phoned the office and was told - "Oh yeah that's on hold as there is a Repair that has not been approved yet".

This Repair was the one done to the canopy frame and hinges after I got back from the Oshkosh epic in 2017. The Repair paperwork was done to satisfaction (or so I thought) in February of 2018 - so what was the hold up?

Their excuse was that they never received the email I sent which contained all the photos and answers to their questions. Quite frankly I do not believe this story as if the email did not arrive then it would have bounced back to me - which it didn't. It is clear that the LAA don't have any kind of system in place to ensure something is not processed internally. I had assumed all was well as I had not heard back about the Repair.

Anyway on sending the email again it was swiftly approved - except now I had to get an inspector to look at the repair and sign off what I had done - more paperwork!

So I arranged an inspection and he discovered that the reinforcing carbon I had used to thicken up the hinge brackets was cracked at bottom end - not surprising since I had done this repair 'in the field' and so the reinforcing had a weak point which was at the end where it was not really bonded to the existing hinge as it was just butted up to it.

Actually I was happy in a way to have another opportunity to repair the hinges properly this time, so I decided to take the canopy off and repair it at home.

I started off by taking away some of the reinforcing carbon, totally so in the region where it had cracked.

I then made up two small blocks of wood that would sit inside the hinges and allow a clamp to force the carbon up into the corner where it had cracked before - thereby ensuring a complete bond this time round.

The back hinge was not cracked but I decided to do this repair to both hinges anyway to make them right.

The front hinge got a further layup of 3 layers of carbon in a strip which was laid over the entire surface - see pics - to closeout the layup and add even more strength to the hinge. The front hinge is subject to much more flex than the rear one due to it's distance from the retaining wire which supports the rear hinge much better as it is closer to it.

After the carbon work was done I primed and painted the whole area, making a much better job than I did before.

The canopy is now back on and signed off by my inspector - I am now waiting for the Permit to arrive but this has taken more than a week already. In the meantime I have used the time to touch up a few things here and there but hope to get the Permit soon.

If I could start from scratch and build again I would consider reinforcing the inside of the hollow canopy frame with runny flock in the area around the hinges. I would also take away some of the micro in this area and replace it with carbon and also reinforce the hinges as I've done here too.

Canopy off and at home to do the repair - a much better idea than trying to do it 'in the field'.

This shows the crack that developed in the end of the carbon reinforcement - it was just butted up to the end and therefore not really bonded to the round portion of the hinge.

Two blocks of wood, shaped to fit and wrapped in Duct tape to release, enable clamps to force the new carbon reinforcement into the corners of the hinge.

Many layers of carbon were used here - it is easy enough to take away any excess afterwards.

One further step with the front hinge with a layup of more carbon over the entire hinge to close out the reinforcement.

Front hinge with it's new reinforcement. Much stronger now.

Rear hinge. You can see that the carbon has added about double the original thickness of the hinge so now it is 3 times as thick overall.

Time to prime.

Front hinge after paint and back in place on the aircraft.

Outside looking better now - I have since feathered off the edge of the paint too.

I am a donkey

I tried moving around an oil hose recently and that required me to first drain the oil. 

I ended up having to do that twice and each time I did not bother to use a new copper crush washer on the sump plug, also I was not using a torque wrench on the sump plug and have been over tightening it as a result - made much worse by the use of an old washer which provides no 'feel' as to the torque as it does not have any crushing properties once it has been used.

So I ended up stripping out the thread in the sump!

What a donkey...

I discussed the possible fixes with Jonathan the UL Power agent here in the UK and decided that a new sump was the best fix - particularly as it ended up being cheaper than a Helicoil type kit to fix the thread (the best versions of this coil - TimeSert - are very expensive).

So I've had to put a new sump on. In doing so it gave me a chance to see inside the engine. 

I noticed the cam lobes had a bit of wear on them which did not look right to me - the engine has just over 300 hours on it.

I consulted Jonathan about this and he shared it with Patrick in the factory too. This was his response (I also asked about the exhaust valve pic below too).

"Morning Andy

Feedback... 

1. Wear on cams nothing to worry about.  A bit more than normal.. but OK.  (Always check your oil filter mat and magnetic plug for material tho!)

2. Your valve picture... apart from evidence of a tiny leak due to uneven bolting of the exhaust manifold the rule of thumb is 'uniform burnt pizza is good'.

Here is an unofficial  guide to valves..

White deposits from lean  / hot
Black deposits from rich / cool
Brown/reddish burnt pizza is Goldilocks  zone 

With the caveat of 'uniform distribution '

Green is bad. Holes are bad. Uneven is bad.

If your compression/leakdown us good and CHT / EGT good you are a winner! 

Normally all 4 exhaust valves should be similar in appearance.


I wonder if this wear was caused by all the Swarf that was present in the engine, when it was new, after machining? - as evidenced in the first few oil filter changes when checking the filter element (if you remember).

Anyway some lessons learnt. Particularly about always using a new copper crush washer and using the correct torque values - 25nm for the sump plug. All the torque values can be found in the Illustrated parts manual from the UL Power website.

Thread stripped out by Mr Donkey - me.

Old sump coming off - took a few 'love taps' with the rubber mallet to come away.

Oil pickup in the old style iSA engine. The long threads limit movement of the pickup.

A chance to look inside the engine.

This cam lobe wear did not look right to me so I asked.

Another lobe with wear. Strangely the front No.1 Cyl lobe did not appear to have any wear.

Overview of the cam.

Wider view of the above.

No.1 Cyl exhaust valve - even 'burnt pizza' is good. The other 3 looked just like this one.

New sump in place.




Wednesday, 5 December 2018

Maintenance matters, fuel leaks and TPS (again)

I'm pleased to say that yesterday G-FUUN passed it's annual inspection again, so it's good for another year.

Unfortunately I've also had another series of fuel leaks.

I had originally thought they were only coming from the Ali end plates - specifically the central bolt through the fuel sender that tightens the plate (via an ali tube with a thread inside the tank) but as it turns out that was not the only cause.

I've since replaced the ali end plates with new ones and ditched the fuel senders altogether as I now have solid state fuel sender probes. The rubber seal on the Falcon gauges was changing over time and caused some leaks.

The other thing I've done to improve the end plate sealing is making up some thick rubber washers which have a slightly undersized hole drilled in them. Screwing that onto the thread and then putting a large metal washer in front on the outside helps to seal things better than just relying on Proseal. I also made up a smaller washer for the inside which is squeezed between the plate and ali securing tube.

No leaks from the plates anymore!

However recently I got another fuel leak, very small, just weeping out, but very annoying as it took me some time to find out where it was coming from.

To start with I thought it was further up the tank and fixed that supposed hole from the inside with flock - see pic below.

But that was not the cause.

I then bought a UV dye kit which consists of a UV dye and a UV torch. You mix the dye in with about 10 litres of fuel then pour that into the tank. Wait a while then in the dark shine the UV torch and you are supposed to see only the fuel and find out where the leak is from.

Well this did not really help and I ended up leaving it feeling quite frustrated as it still was not clear where the leak was coming from.

When I came back to the job the dye had actually helped as it is orange and visible in the daylight too so does point to where the fuel has been.

A tell tale orange mark from the underside of the wing near the drain finally pointed me in the right direction. It was the ali plate that is bonded into the tank and lower wing skin that was the problem. So I used runny flock to fill in the cavity around the drain and overflow/breather fitting on the inside. Being runny this found it's way into every nook and cranny of this area and sealed up the leak good.

No more leaks again (for a while anyway - fingers crossed).

I will do another blog post about the fuel senders when that is finally sorted as it too has been a bit of a saga and is hopefully soon to be resolved. The manufacturer has admitted to a faulty set of modules for these and the replacements are finally here but stuck in Customs at the moment. 

Finally the Throttle Position Sensor - it has started to fail on me again. Happening when I took of from a bumpy hard runway the other week at Great Massingham.

I checked when I got home and found that it has failed at almost exactly the same number of hours as the previous one - 123 hours approx.

Ironically the next week UL Power released a Service Bulletin which said they are no longer fitting this type of sensor to any new engines in the future. Which to me says they admit there is a problem! 

They are converting to a Hall Type sensor - which does not have the sensing parts touching so therefore is much more robust to the vibrations of an aircraft engine installation.

They are offering the Hall Type sensor as a mod to us older engine users but at a price! Around £600 which is a bit cheeky considering the Hall Sensor is a £25 part. True a new inlet block is needed and an ECU remap and additional wire but £600 is a bit rich.

G-FUUN with it's clothes off for it's annual inspection. It passed again 'without comment'.
 
New fuel sensor probes - more about those in another blog post.


Fuel tank end plates - now without the Falcon senders and with rubber sealing washers.
Rubber washers I made up to help with the sealing of the fuel tank end plates. These are working well.

Port wing off to fix this fuel leak.

I thought it was coming from this 'hole' part the way up the tank.

Which I fixed with flock but it turned out was not the cause of the leak.
This area around the drain and overflow/breather connections was the cause of the leak and has been fixed with runny flock allowed to run into the area.