A return to model building and flying. Part 38
Note: if you follow any external links from this page, use the browser back button to return to this page.
It Works!..... I think?
Regular readers must be heartily sick of me rabbiting on about carbs, but at present that is the major cause of problems with all the carrier models I have built. As I would like to eventually compete in a carrier competition, even if it's one I have to organise myself, it's something I have to sort out. I did say at the beginning of this epic that I would write about all the problems I encounter; and this has been one big consistent headache that I have to sort out, or at least understand.
After all the problems with throttles last month I finally seem to have made some progress. The extra movement to the throttle arm appears to have doing it's thing, to the extent of giving me control whilst flying on a wind blustering between 24kph/15mph and 40kph/25mph; not a particularly pleasant experience, but it proved a point.
The Enya 30 BB carb that gave me so much grief last time, still played up at first. The idea of using this engine again was the thought that, if I can get this to work anything will. Unlike the other carbs I know and love, the Enya adjustments are very critical, alter the main needle and it affects the behaviour of the midrange jet and vice versa. Just when you think you have it all sorted and all it needs is a sight tweak in the main needle, everything goes to pot, and the some part of the middle range goes very lean. I got it together in the end, but this is not a forgiving carb to adjust, it makes me want to laugh out loud when I hear people moan about MDS carbs. It still feels very non linear in it's response, but it does work now.
The other important thing was the fact that removing the throttle barrel bias spring completely, does not seem to have affected anything unduly: at least that I can notice. It also has the advantage that now the throttle tends to close if unattended instead of opening, this in turn means I may be able to do solo launches without my bit of wire and string to restrain the model.
Next thing was to try it on a calm day. Everything worked, even the engine defaulted to a tickover, not flat out, which enabled me to leave it unnatended and not have it hurtling across the field as soon as my back was turned.
The only fly in the ointment is the strange lag when the throttle is closed, It takes half a lap to settle down, then, although it responds well enough to being opened, it has a peculiar non linear accelleration, which makes it very difficult to to control the transitions from slow to fast. This is not helped by the engine developing quite a lot of power at low revs. I know the carb is working now, better than it has ever done, but the effect still feels like an on/off switch. There is obviously nothing wrong with the engine itself, but I am not happy with the carb at all; so I am reluctantly going to try another engine in the model.
Playing with a Wildcat
To restore my faith in engines, I had a play with the Wildcat to remind myself how a carb should work. For those that have not read the previous page, the engine is an MDS.18. and you can see some of my antics in this short video. Apologies in advance for the out of focus model but I can't film and fly at the same time.
Back to the Crusader
Next up for the Crusader, the MDS.28 as a replacement for the Enya. Again it worked better than the Enya but still it was a pain to get the full travel on the throttle lever, this being aggravated by something binding in the bellcrank area at one extreme end of the movement, does not not make for a sensitive throttle response.
I am now certain of what is fouling up the whole system but cutting out the whole bellcrank assembly, redesigning it and making it all from scratch is a task I don't want to undertake if I can get away with it.
As the Enya and the MDS had quite adequate power, I decided that a smaller engine would do just as well and maybe not quite so inclined to start heading for the stratosphere as soon as the the throttle starts to open.
Enter stage left, an O.S.25FX I have had hanging round for some time waiting for something suitable to put it in. This has been, on the whole, a slight improvement, in that I can now fly the model with a reasonable amount of control; that's after a considerable amount of piddling about with the carb settings to get it to throttle evenly. Yes folks, even O.S. carbs need a lot of tweeking to get them to run right. and there is a certain lack of forsight in the design, when the the actuating lever is so close to the engine body that it's difficult to attach the pushrod clevis and not have it fouling something. The engine also has a rear mounted needle assembly which I much prefer, even if it is fixed in a somewhat vulnerable position; unlike it's Chinese counterparts. I get the distinct impression O.S. took their eye off the ball during this period.
Now, hopefuly? I can finally get on with trimming the model.
Although everything is now nearly working as it should, there is still an unpredictable delay in the throttle responce at the low end of the rev range. Whilst I can just about get away with it on a calm day, it could very rapidly turn into a nerve wracking if not suicidal event in a wind. I am now out of options, I have to rip the wing apart to try and find out what is causing the problem.
The carnage above was unavoidable in gaining access to the control system. The maddening thing is, I still cannot find out what is causing the bellcranks to bind at one end of their travel. I am so fed up with problem now, that I have scrapped the original system. and will now take the radical, if not very elegant, step of mounting the bellcranks externally; there is just not enough room in the wing to fit larger cranks that it obviously needs to surmount the gearing problem with the throttle arm.
The first test flight was with the MDS.28, as it's the one I have the least problems with regarding throttling. Everything worked apart from the arrestor hook doing it's usual impersonation of a plough and making take-offs unpredictable and long winded. Removing the throttle bias spring had also meant not having to use a release device. I'm still not sure if this was affecting the throttling in the air or it was something else, but the full throttle responce was not what I expected and seemed wrong for some reason I can't explain. But bearing in mind this was the engine that originally had it's liner installed 90° out of true, I'm not expecting too much from it.
The thing that caught me completely off guard and nearly wrote off the model was the elevator throttle bellcrank interaction being 180° out of phase. All the previous models have given an up elevator kick to a greater or lesser degree as the throttle is shut. This is really only noticeable if the the throttle is shut suddenly from flat out. Whilst it can come as a bit of a surprise if you are not expecting it, it is at least safe in that the model is moving away from the ground. The modified crusader however, went in exactly the opposite direction. One moment I was was flying fast and low, the next the model had buried it nose into the ground and bits of prop where flying into the air. For once I was glad that the model is overly strong. The only damage was a broken prop and a bent nose leg. I have now put the elevator horn on the underside of the tailplane to correct this behaviour and done a bit of pushrod bending to stop it fouling the leading edge of the all moving tailplane. At the same time I have redesigned the arrestor hook to sit behind the fuselage end instead of underneath it in an attempt to stop it's ploughing antics.
I noticed during the modifications to the Crusader, that the tank feed and pressure pipes had been crossed over. This may have been the cause of the strange throttle setting and response on the MDS. This Proved to me once again, that MDS engines can take a lot of abuse and still work.
Harking back to a previous point, I also removed the throttle bias spring from the SC.40 carb on the Hellcat. this worked perfectly, with no problems, the engine settling into a nice tickover if left unattended. This engine has always been well behaved anyway. The only problem I can see that might arise is that, if line tension is lost, say up-wind flying slowly, the engine will have tendency to shut down instead of open up which could aggravate the loss of control. Only time and experience will tell me if this is a good or bad trade off. For now the Hellcat is finally beginning to fly as I want it to; or maybe I am just getting more used to it? The one thing I have noticed is that on a 10x6 prop the SC.40 goes like stink, but has a tendency to roll the model round the engine if the thottle is blipped quickly, causing a lot of unstable roll and pitch activity. An 11x6 Taipan prop (I have no idea how it came into my possession, and I have not seen any on sale) tames this activity a bit, probably because of it's large blade area absorbing some of the power. Didn't you just guess this was going to be the best prop, just because I only have one.
Brodak - Buster
Regular readers will know that this is a kit I won on eBay some time ago. I am in desperate need of a semi decent stunt model, so that I can at least keep my hand in at flying the schedule. I am by necessity restricted in the size of model I can build, but this one should fit the bill nicely even if it is, marketed as racing model, come combat trainer, come stunt trainer. Not quite inclusive of the kitchen sink; but close. When I first opened the box I was impressed; well into building the wing, my enthusiasm has cooled a little. The wood selection is a trifle erratic; the railing edge sheet comprised of one piece rock hard and one piece very soft, and centresection sheeting is much too hard and heavy for a stunt model. However if it was meant to take some heavy punishment as a combat model, then maybe it isn't so far off. If this is the advertised contest quality balsa, I would like to know what sort of competitions it's used in, the mind boggles!. I am still enjoying building it though, it's a fairly strait forward simple model, and I am certainly not knocking the only major producer of CL kit's in the world. You can see the full range of Brodak CL kits here, it's impressive.
During the course of building the wing I ended up with two ribs missing. Whether this was a manufacturing oversite, or the previous owner of the kit had removed them for some reason I will never know, but it was easy enough to replicate them using one of the originals as a template. Of course Sods Law comes into play at this point and I find the missing ribs two days later, the die cutting was so feint I thought it was a piece of sheet wood.
Putting the cap strips on the wing was becoming a bit of a chore, as half the wood supplied was completely the wrong type and density, hard quarter grain is not the easiest thing to bend or pin.
This was the time to try a technique we had recently talked about at the club field, namely using heat to bend the wood. I had read a piece in an Aeromodeller about using a tin can and a candle. This turned out to be a bit of a mistake as the tin got knocked onto the floor and the candle splashed hot wax over a new pair of shoes I was wearing. As I have awkward shaped feet, finding a pair of shoes that fit is a rare event. We were not amused.
A somewhat safer option was to use my plastic film heat sealing iron clamped in a vice. This is very effective as you can see from the pictures using some of the quarter grain strip. It's even better with some soft wood.
I am putting off some hard decisions on how to modify it. The model, will, be nose heavy, and the fuselage moment arm is very short, it looks much longer than it actually is in the box illustration. This makes me wonder if the prototype did actually have a longer moment than the kit? The tailplane and elevator are also very small by modern standards. So I can extend the rear of the fuselage and/or enlarge the size of the tailplane.