B J Craft Anthem build - 2 m FAI F3A class aircraft and Debowski TMCR contra

[Philip Lewis 3]

Just wondering if the servo gears had simply worn out due to their age Peter, or do you think there might be another reason?

 

I would definitely recommend you check for equal throws on the new elevator servo's if you haven't already done that, I also use them for the elevator servo's and in both of my planes it required a curve to be appied to one side to match the other throughout the travel.

 

Edited May 24, 2023 by Philip Lewis 3
Spelling

[Peter Jenkins]

Could have been a contributory factor Phil but BJ was quite insistent that while the 3301 was a great servo the metal gear set should be used.  However, as the 3301 is a bit of an old design I couldn't find anyone selling nylon or metal gears.  Sadly, they will have to go in the WEEE bin.

 

I did check for equal movement over the full range of the servo.  After tweaking the clevis to get as close as possible to both servos matching position at neutral, I had to use the X-Bus facility to tweak one servo.  There was an absolute match on up elevator at full throw but there was a slight difference on full down elevator.  Unfortunately, having gone the X-Bus route for the elevators I could not find a way of putting a curve on the throw to get alignment at the extremes.  Apart from the slight misalignment at full down (for spinning purposes) the servos match each other extremely well throughout their range of movement.  Had I used the non-X-Bus solution then there is a servo match function on the JR XG11 that allows you to slave one elevator servo to the other and get them to match movement throughout the range.  I'm sure with an X-Bus servo that would be possible but a non-X-Bus servo being fed by an X-Bus converter doesn't seem to offer that option.

[Peter Jenkins]

Just to finish off this story, I flew the Anthem in competition for the first time last Saturday.  The aircraft performed very well and it was just a shame the pilot wasn't up to doing it credit.  The other Anthem pilot did much better coming 3rd in only his first season in FAI(P) while I brought up the rear!

 

There are a couple of tweaks I might carry out, one being to have a windy day setup with a faster datum speed to try to reduce the difference in ground speed going into wind and down wind.  This will mean re-binding the Rx for the Windy or Calm day as the case may be.  

 

The combination of the Anthem and the TMCR has definitely provided a slightly better performance than the Element with the CRS contra and it's now up to me to up my performance to do it justice.

[GrumpyGnome]

Whilst it's not my cup of tea, I like the build threads, and am also gaining an insight into the complexities of this aspect of the hobby.  Nice one!

[Peter Jenkins]

Thanks to a PM on another forum, I have found that the XG11 does indeed allow you to set up 2 or more memories to be treated as the same under the same bind.  I can now switch by just changing the model memory from windy to calm with no need to bind.  I might even decide to have 3 settings if I find that there is too much of a difference in flight speeds between the two current settings.  I'll also have slightly different elevator trims for each of those memories and, possibly, slightly different snap settings to get the same snap speed for different flight speeds.

[Peter Jenkins]

I wasn’t going to post again on this thread but something came up!  A couple of weeks ago, I was at my 2^nd flying club site to get in some practice of my aerobatic schedule with the Anthem.  This site has a row of 20 ft trees at the eastern boundary so for today’s wind conditions, I would be landing over them.  I had reached the manoeuvre called the reverse Immelmann – basically an Immelmann carried out from inverted and at the bottom of the aerobatic box and at the downwind end.  There was quite a brisk wind blowing as well.

As I started the half outside loop and started to ease up the power, I heard a loud crack and the power seemed to go.  I immediately pushed hard to get the aircraft heading back to the patch and my immediate worry was would I get the Anthem back to the patch.  As soon as the aircraft was upright and heading back towards the patch, it became apparent that it would not be able to maintain height and that I would lose sight of it once the 20 ft treeline blocked my sight line.  At that point, I had not touched the throttle as I felt the problem was a power pack issue.  Given the potential total loss of the Anthem, I closed the throttle completely and then slowly advance it until there appeared to be some power being delivered.  The amount of power seemed sufficient to get the aircraft back to the patch but before I could add any more power a large brown puff of smoke exited the fuselage.  At this point I wondered if the flight pack was on fire!  However, no other smoke exited of the fuselage – thankfully!

 

I had to stretch the “glide” as much as I could to maintain a visual line of sight to the Anthem while it slowly made progress against the wind.  As the Sun was shining, I saw the shadow of the Anthem on the 20 ft trees as the aircraft passed over them – there wasn’t much of a gap between the aircraft and its shadow!  The aircraft’s flight path back to the patch was at about 20 deg to the runway direction.  However, as the speed was low and the aircraft quite low by now, I was loathe to turn the aircraft onto the runway heading.  I felt there was just enough space to land and then use the rudder to steer away from the long grass (3 ft high) along the edge of the runway.  I was almost right!  The aircraft was almost straight when the port wing tip caught in the grass and spun the aircraft into the tall grass.  Phew, all in one piece when I had been looking at a total loss had the aircraft hit the trees or worse still not made it over them at all flown into some unseen hazard in the field below it.

When I reached the aircraft, I lifted the canopy expecting to see a possible fire but, thankfully, there was no sign of fire although the ESC on the LHS of the aircraft had disintegrated and there were some scorch marks on the fuselage structure and a very strong electrical burning smell!  The great advantage of having a mobile phone with you, set to Flight Mode of course, is that you have a camera with you at all times so took a couple of photos of the situation I found.

In the photo, you can see the rear motor is covered in carbon dust and the ESC reduced to its charred component parts!

 

At that point, I also noticed that the port u/c leg had been bent backwards.  No other damage was evident.  Interestingly, the rear prop was completely loose which led me to wondering if there had been an ESC fault that stopped and then restarted the motor causing the prop to loosen its retaining nut.  The D3 ESC is one of the fastest reacting ESCs there is and when you close the throttle the props abruptly stop.  Indeed they abruptly start when you advance the throttle and that speed of reaction is there throughout their operating range.

 

The 2^nd ESC appeared to be fine.  Remember, this is a twin outrunner motor contra rotating setup with 2 ESCs one driving each motor.

Thank goodness for having this redundancy in the power set up!  If I had only had the one ESC the outcome would have been an irretrievably damaged model. 

Having emailed Adam Debowski, the supplier of the motors and ESCs, he suggested visually inspecting the motors for any discolouration of the copper windings and then if no sign of that to remove the props and try each motor using the good ESC – obviously I swapped over the 3 wire connections from the good ESC to the rear prop motor to check it out.  Both motors were undamaged following a visual inspection and both worked as expected with the good ESC powering them.

I relayed this back to Adam.  Apparently, this is the first D3 ESC failure that he has had reported back to him.  Why did I have to find it?  Anyway, he’s going to ship me a new D3 but that was going to take around 7-10 days.

Neither prop was damaged as a result of the ESC malfunction or being driven into the grass!  Another “Phew!”

As you will see from the photos, the failed ESC came apart completely and the heat generated in its demise melted the solder holding in the 2 power leads and the 3 motor leads.  That probably prevented any further electrical damage.  What remained of the 2 large capacitors are at the bottom left of the photo and the shiny bit is the heat sink.

 

Having got the aircraft home, it was apparent that the u/c plate had fractured but only on the port half of the plate.  Twisting the leg to put the plate back in it’s correct position allowed me to unbolt the leg and inspect for damage.  Thankfully, the leg was undamaged.  So, I just needed to get the u/c plate repaired and a new ESC and the Anthem would be back in business.

The big question was how best to repair the u/c plate that is impossible to access without major surgery.  I came up with one repair scheme but chatting to Keith Jackson (he writes the Aerobatics column in RCM&E) at the next competition I was schedule to fly, he came up with an alternative solution.  I’ll describe that as I show how I did the repair.

[Colin Carpenter]

Excellent piloting skills in a difficult situation, Peter. I have had ESC failure in a HK SBach where the only option was to land ahead over a hedge into grass. Got away with that with similar uc damage. Well done Colin 

[Philip Lewis 3]

I've had to do that excact same repair, my landing plate one side was toast with it's cheap rubbish ply. Luckily the formers either side were perfectly OK so it just needed a new plate. Taking a reciprocal saw to an expensive airframe is not for the fainthearted but there was really no other way to cut it out cleanly.

 

The new ply plate went into the existing holes in the formers with 1 mm carbon plated either side and the joints where the formers was reinforced with carbon ribon tape and laminating epoxy which makes the joint very strong.

 

To my mind the secret is to make the undercariage leg the weak point rather that the attachement point so whilst the landing gear legs on these planes are pretty fragile not to mention expensive I'd rather they break before something inside the suelage does.

 

I 'm interested to see how you repaired yours using Keith's method.

[Graham Bowers]

Another good save! Probably a good time to put the lottery on ;-)

[Peter Jenkins]

What I have done before, on another 2 m F3A airframe, is to cut out the entire u/c box assembly and make up a new one using proper ply and allowing the u/c attachment plate to have a longer tongue to fit into the vertical ply formers.  On my Element, which I bought from Keith, he had cut out the entire u/c box assembly from the beginning and used a carbon composite sandwich glued across the fuselage with a much longer length to spread the torsion load imposed by the long landing gear.  As it happened, I was using the Element to fly in a competition where the turbulence on approach.  As I had caned the battery to overcome the strong wind that day, my drive pack low capacity alarm had been going before I completed the last manoeuvre!  The approach coincided with a very gusty period and the aircraft was being thrown all over the sky!  My caller said "Overshoot" but I was concerned I wouldn't have enough battery capacity to do this safely.  So, I continued with the approach and had a hard landing that broke the starboard carbon fibre u/c leg.  When I looked inside the fuselage, Keith's carbon sandwich composite plate was completely undamaged - Phil's point exactly.  That was what triggered my conversation with Keith.  When I sent him the photos of the Anthem's u/c box damage, his suggestion was to make use of the good part of the u/c box structure but to cut away the top part and glue the carbon plate to the now flat top and the fuselage sides as per the Element design. 

 

This struck me as a great idea as it both saves cutting out the u/c box frame and makes use of it for added strength.  I'll do a drawing of the proposed repair scheme and post that together with photos of the Element u/c mounting and a series of photos showing what I have done to effect the repair.

Edited July 9, 2023 by Peter Jenkins

[Peter Jenkins]

Well, I didn't manage to do a drawing of the repair scheme or take a photo of the u/c plate in the Element but I did finish all the wood work on the Anthem using my Dremel very carefully!  The net result is shown in the photo.  I also used a piece of depron card to make a template for the carbon plate that I will cut out using the template.  The great advantage of the template is that you can make mistakes and it doesn't break the bank!  I also took the opportunity to locate the u/c bolt holes.  I will check these again tomorrow by bolting the u/c in place through the template to check that the holes are correctly placed.  So the photos show these two activities. 

 

The plate lies flat against the top of the cut down u/c box.  I will glue the plate to the box and to the fuselage sides.  The key thing about the plate is that it will spread the torsion load placed on the u/c box over a much longer moment arm.  I will be reinforcing the plate to the fuselage with some carbon cloth to further strengthen the fuselage/plate join.

 

  

 

As you will see from the photo on the right, the template extends along the length of the carbon reinforcement that is already in place.  I have left the front vertical brace in place but cut away the rear one so that I do not have to cut into the carbon plate template to accommodate it.  I have rebated the template for the front vertical member.  

 

The next job will be to transfer the template to the carbon composite sandwich and cut that out.  I will check that the u/c attachment bolt holes are correctly positioned before then gluing the plate in place.

 

My new D3 ESC arrived in the post today so that all the bits I need to get the Anthem back into the air are now to hand.

[Peter Jenkins]

Discussing my proposed repair with Keith, he asked how long the carbon plate was in the Element.  That was 100 mm, so I looked to reduce the size of my template to 100 mm spaced equally either side of the bolt holes.  I then used that template to mark out the plate on my piece of carbon composite.

  

The reason for the larger clearance slot on the left hand side of the photo is related to the large piece of support structure that you will have seen in an earlier photo above.  That broke off on the port side of the u/c plate (poor gluing) whereas the one on the starboard side remained firmly glued in place.  I did think of leaving the plywood construction in place but that would have left the bolt on the extreme starboard side without the carbon support.  So, I resorted to my Dremel to grind out this piece of laminated ply but trying to get all of it removed without damaging the vertical support or the fuselage side meant that a small piece remained so cutting away the carbon plate was the only way of getting a flat gluing surface.

After cutting out the carbon plate, it needed further fettling to get it to fit.  After that I did a dry fit bolting the u/c legs in place through the carbon plate.  That indicated that a couple of bolt holes had not quite aligned with the existing bolt holes in the old ply plate so a gentle tweak with a needle file had everything fitting.

 

Time for the epoxy.  I was intending to use 30 min epoxy but when I came to pour it out I found that the adhesive part was like jelly!  As it was now too late to rush into town to buy some more 30 min epoxy, I decided to use my 12 min epoxy!  I reckoned that it should be possible to get the whole job done in no more than 5-6 mins so decided to go for the fast solution!  Keith had recommended mixing in balloons in with the epoxy and as I had some (quite a lot really!) left over from the last time I used them all was fine. 

 

My plan was to bolt the u/c legs in position to clamp the carbon plate in place.  So, to avoid gluing everything together, I put Vaseline on the top surface of the legs, the bolts and the underneath of the large repair washers I intended to use on top of the carbon plate.  So, after a second Vaseline fuelled dry run, I mixed up the epoxy and balloons and set about applying the glue to the existing u/c structure and the edges of the carbon plate.  Having put the plate in place, I bolted the u/c legs in place and left the thing to cure.

 

  

 

 

 

Once the glue has cured, I’ll put some carbon cloth along the join between the carbon plate and the fuselage side.

[Ron Gray]

Nice work, Peter. 

[Peter Jenkins]

With the carbon plate glued in place, Keith had recommended a layer of glass cloth along the plate/fuselage join.  I was about to do that when I spotted an offcut from the carbon sandwich and thought gluing a short length of that to the rear of the u/c plate would both spread the load into the already carbon reinforced side fuselage and act to prevent the u/c plate being rotated by the u/c forces.

 

As my Hysol epoxy had nor turned up, I used that to glue these side pieces in place as its thixotropic form meant it wouldn’t dribble away.  I also use the opportunity to fill in the gaps caused by the normal epoxy running down inside the fuselage and leaving at glue gap at the top of the plate/fuselage junction.

 

       

 

As the Hysol takes 12 hours to set, I left it for 24 hrs and then refitted the motor, ESCs and battery tray and reconnected the ESCs and the telemetry leads to the Rx.  I gave the Anthem a quick ground run to check the new ESC’s functioning and all was well.  I ran the motors up to an indicated 2,600 watts to check that the ESCs were working.  At that power level, I was having difficulty in holding onto the fuselage by the u/c legs – it was restrained but I didn’t want the tail planes holding the aircraft with that amount of power being used!

 

The subsequent air test showed that the only change in trim caused by slightly less downthrust in the motor setting requiring a bit less up elevator trim.  Since I had not changed any settings for the wings, the zero aileron trim was spot on and the vertical up and down lines were also spot on.

 

So, I can now re-start getting re-acquainted with my 2023 aerobatic mount.  The slight increase in all up weight is a disappointment (I forgot to do a before and after repair weight check on the fuselage but I would guess the weight increase is of the order of 50-80 g. 

 

It is surprising how many differences there are between the Element which uses a different motor and ESC and the Anthem. 

 

 

 

 

 

[Adrian Smith 1]

Crikey, Peter I don't know how I missed this thread. Very interesting from start to finish 👍

[Peter Jenkins]

10 minutes ago, Adrian Smith 1 said:

Crikey, Peter I don't know how I missed this thread. Very interesting from start to finish 👍

As you can see Adrian, I have a good deal more space I can use than you have in the Capiche for moving battery packs around!  That was why I decided not to electrify my Capiche 140 especially as it was a bit porky to begin with and tipped the scales at almost 6 Kg!  When I put the DLE35RA in it, the weight climbed to 6.25 Kg and it just wasn't able to fly at a constant speed.  Full throttle verticals always ran out of puff far too early!

[Adrian Smith 1]

I see. Just purely an observation the Capiche 50cc doesn't seem to be over-engineered compared to some I have owned (Pilot RC for instance), although I am trying to keep the weight down as much as possible. I accept that compared to the very modern 3Ders like Extreme Flight and SkyWing, the Weston Capiche 50cc comes out heavier despite having what I consider to be a narrow fuselage compared to some on the market today. 

[Peter Jenkins]

I thought I'd update this thread to say that I have now finally trimmed out the pitch to canopy that was present when I first set the aircraft up with 0.7 deg main wing incidence.  The TP was at 0 deg.

 

I tried a mix coming in above the datum throttle setting to introduce a smidge of down elevator but that wasn't entirely satisfactory.  So, I took the other route of increasing the wing incidence which requires an increased down force from the tail plane to maintain S&L flight.  In the vertical upline, with no wing lift force, the up force from the tail plane is still there and should act against the pitch to the canopy.  It has taken several trimming sessions, with the final one being today, to arrive at a wing incidence of 1.25 deg at which the aircraft will fly vertically upwards hands off.  I now have the thrust line sorted and just 2 clicks of left rudder that I will dial out with a turnbuckle adjustment in due course.  I zeroed the elevator trim by adjusting the elevator control rod length but after this morning's test I now have 1 click of down elevator.  The ailerons are level with the wing and the aircraft flies dead level.  

 

I shall now just focus on learning how to fly the new schedule with the Anthem in an excellent state of trim.  

 

I will keep doing trimming checks from time to time as things tend to change a little over time and these aircraft immediately react to the tiniest bit of out of trim setting.