Dewoitine D.510 / D.501 - from 53" John Blakely plan

[Mark Honman]

I've been looking around for a new "everyday" model, small enough to fit in the back of the car when assembled, light enough for 3S power, good for my continued quest to learn aerobatics, and if possible to scale. 

 

So this design took my fancy - with a 53" span and 1.3kg weight (not sure if this is going to be with or without batteries though) and full length ailerons it might just do the job. 

 

However as I started to study photos and a cutaway diagram of the prototype, I realized the shape of the fuselage is a bit off in various ways. Hence this build blog, to illustrate the ways that I'm trying to build a D.510 with more realistic proportions - and also notes on things I wish I'd known when I started.

 

Choice of prototype

 

It's actually simpler to build these plans as a D.501, which has a 2-blade propeller rather than the 3-blade prop of the D.510; also the undercarriage design in the Blakely plans is that used on the D.501 rather than the updated D.510 undercarriage.

 

Since it will help to know which way is up, the initial plan is to model one of the D.510s supplied to China in 1938. These have a green mottled upper surface, and "apparently natural aluminium" underside. Photos are very few and far between, and even the plastic scale modellers don't seem to agree whether the mottle pattern is mid green over dark green, or mid green mottle over natural aluminium.  However, combined with the big blue and white Chinese roundels the mottle-on-aluminum finish should be OK for visibility and still lightweight - if acrylics will take on covering film. 

 

However it's still tempting to built it as a D.501 with one of those multicoloured French not-very-camouflage schemes. 

 

[Mark Honman]

First thoughts

 

The first disappointment was the slab-sided lower fuselage (if Keil Kraft could do rubber models with realistic proportions in the 70s, I don't know why we can't do better in the 21st century).

 

Other things that made me go "hmm" were the lack of any provision for right-thrust, and the need for 56g of nose weight for balance. 

 

So I decided to shape the rear fuselage formers with some curvature as in the cutaway of the prototype, use 3/32 rather than 1/8" sheet. Also, rather than cutting the sheet with its top edge aligning with the main longeron, I used the full height of a 4" sheet. The aim is that the rear fuselage should be a curved shell which supplies most of the required strength (something I want to try anyway in the next project, where I'm thinking of a using full monocoque construction).  

 

However that means that the it would all become very hard to align... 

 

Fuselage structure 

 

Being one of those people for whom there is no design or procedure that can't be "improved" I decided to start by building an easy-to align box consisting of the doublers (1/32 ply rather than 1/64, I'm a cheapskate) and the formers immediately in front of & behind the wings. The longeron in the first photo is there for alignment purposes only. 

 

Separately, I pinned the longerons to the plan and attached the upper fuselage formers to them. F9 was split into upper and lower halves, and transverse braces added to maintain correct separation of the longerons.  

Edited December 23, 2021 by Mark Honman

[Mark Honman]

Picture time (didn't realize I couldn't come back to the posts and insert the picture links)

 

The picture from the plans. Now that I've studied so many photos of the prototypes, there are 3 things that stand out as being wrong:

1. the protoype has a higher "waistline"
2. The cylinder head fairings aren't on top of the fuselage, instead the fuselage sides should extend vertically to the beginning of the curve that defines the top of the fairings.
3. The top of the fairing should be straight and horizontal.   

Ribs, formers, and tail parts marked out for cutting. 

 

Fuselage parts after cutting (with former and longeron alignment marks drawn on the inside of the 3/32" fuselage sides).

The fuselage sides extend forward of the firewall F3. 

 

doubler box - this worked out well, I would definitely do it again. 

 

Longerons and upper formers on the plans. Plenty of water sprayed to help the longerons bend. The board is 20mm blue insulating foam. A crossmember is placed at the right angle for firewall F3 to be glued on. 

 

[Mark Honman]

The plans show a cockpit floor, and also a floor to the battery compartment behind the firewall. I ditched the idea of a battery compartment floor - to make space for bigger batteries, and improve accessibility to the motor etc. The cockpit floor is partial; that might result in a more realistic look to the cockpit when it's all done, but who knows at this stage? Though I do regret not finding a way to fit the cockpit floor beneath the longerons. 

 

Also in the name of "realism", F7 is moved forward of its marked position so that a separate instrument panel can be fitted. And it has a cutout for the imaginary legs of the pilot. 

 

The upper frame fits to the ply box with a twisting action in which the longerons engage in the cutouts in the U-shaped ply former F4. (the plans have this as a full-fuselage former, but as it's going to be cut to create the battery compartment hatch there's no point to making it go all the way around. 

 

However... if I was doing this again, I'd extend the sides of F4 vertically above the longeron to support the fuselage sides. Also, the profile of F7 should have about 1/2" of nearly straight sides above the longeron, before curving inwards to its top - and F5 and F6 should be totally straight-sided up to the point where the cannon cover meets the cylinder head fairing.  

 

One the assembly of upper frame and ply box had dried, I added F11 and the bottom half of F10 to the frame and pre-bent the lower front corners of the doublers inwards (copious water as usual). While this was drying, I lad out the lower stringers on the plans, slighltly overlength and with a sacrificial cross-member to help get the right shape. That's not entirely to plan, because of the inward curve at the bottom of F10 and F11. 

 

Note: F8, F10, and F11 are all cut with a curve at the bottom, so that the under-side of the fuselage will be curved when sheeted.

[Mark Honman]

With that cut to length and installed, the fuselage frame is really taking shape. The only part still to come is the spine from the cockpit to tail. And - it's slightly curved.

 

Rather than trying to duplicate the shape from the plans, I found it easier to use an over-width strip of 1/8" balsa and then sand it to shape. What this photo doesn't show is that I later sanded away the top 1/4" or so of F9 (the former at the back of the cockpit) and the "spine" to create a flat base on which to mount the cockpit headrest/fairing. 

 

[that's another fix to the shape of the formers that I would want to make...]

 

Now when it comes to that fairing I'd say: ignore the plans, it looks nothing like the real thing - which is a simple, small, tapered affair which joins the fuselage top in what is pretty close to a straight line. Just work from photos. Although I've spent hours sanding a fairing that looks like the one on the plans, it would have been simpler to fabricate one using 1/2" or 3/8" sheet glued either side of a piece of TE strip. I might yet do that... 

[Mark Honman]

Now I don't know whether the pandemic has caused the local posties to discard rubber bands in the street, or whether it's caused us to take more walks around the village - but I've built up a stash of 8" rubber bands over the last year or so. 

 

These have come in handy for the next step, which is to attach the 3/32" side sheets. I used wood glue on the doublers, and Gorilla Glue elsewhere - as anyway it was necessary to wet the fuselage sides inside and out so that they are more flexible. It turned out one sheet was softer than the other, but it was the softer sheet that gave the most problems, because it was more bent inwards at the top edge. In the end I clamped it the top edge against a piece of hard 1/4" balsa, and that kept it from getting bowed. When the wood had dried I cyano'ed strips of 1/4" square inside the sheet on that side of the fuselage (which caused its own problems, as I hadn't thought to profile them to get the right shape. I had prepared the harder side by wetting it and then strapping it to a glue bottle of more or less the right diameter.

 

Note to self for future projects - would have helped to have an intermediate former to take the load. 

 

It was quite hard to get the curvature at the firewall right; and the curvature of the rear of the engine compartment didn't work out as expected - F5 and F6 were too small - so that needed re-doing later. It would have helped if I'd realised the top half of the fuselage didn't need to curve inwards here. 

 

In the background of that photo you can see the front upper fuselage being pre-shaped by the "strap it to a glue bottle" method. 

 

The next step was to complete the fuselage akin aft of the cockpit, using offcuts from the bottoms of the fuselage sides. This helpfully left an opening a bit smaller than the footprint of the headrest/cockpit fairing. 

 

Since I was using Gorilla Glue here, I wrapped the fuselage with cling film to prevent it expanding everywhere. 

 

[Mark Honman]

The calm before the storm... after a lot of sanding, it's looking good. The wheeze of building the rear fuselage sides with a curve has paid off, they look good and the fuselage is very straight and rigid. 

 

Realization then dawned that it would be VERY hard to make the cylinder head fairings in the right shape, with a front-to-rear taper, and blending seamlessly into the fuselage. 

 

Work stalled for a week while I thought about the problem (yes, this bothers my employers too!). 

 

It was clear I would have been better off, and with less work, if I hadn't wrapped the fuselage sides over the formers. The only solution (vs bodge) was to cut into those carefully shaped fuselage sides and let in vertical sheet extensions.  The first cut is the hardest...

 

The vertical extension had 45deg angle strip pre-glued to the top edge - this keeps it straight (because there is going to be a twist in the sheet at the firewall) and makes a way to attach the inner face of the cylinder head fairings. The 45deg angle turned out to be about right, and 3/8" sheet worked out well for the inner faces of the fairings. 

 

The 3rd photo shows a temporary spacer fitted between the front of the vertical sheets needed to maintain the correct separation (the clamp at the rear end is out of shot). Wedges of 45deg triangle were glued in to hold the outside front at the correct angle.  Photo 4 shows the "end" result (it's from a bit later in the build). 

Edited December 23, 2021 by Mark Honman

[Mark Honman]

Success at the front end revealed a new shortcoming - the low waistline of the design. This was making it impossible to blend the cylinder head fairings into the fuselage. 

The route I took was to pre-shape some really soft 1/8" balsa into approximately the right curve (good thing I have glue bottles of all sizes), stick it on, and sand, sand, sand. 

And fill... and sand... etc. 

 

The happy/easy part of the progress shown in these pictures is the battery hatch, which is cut along the inside edge of the cylinder head fairings, and perfectly aligns with the shut lines cannon access cover on the prototype (I hope...). Of course it came with its own complication, that formers F5/F6 didn't align with the end of the hatch cut - but who cares, the skin is holding its new shape and the formers aren't strictly required (they will likely return "in some form" when time comes to figure out how to secure the hatch). 

[Mark Honman]

As an aside, you might have noticed that I use easy to find adhesives - Gorilla glue, PVA, and Toolstation's high-viscosity cyano. I haven't found a hardware-shop source for thin cyano, but I'm learning to live without it - and much happier I think, because the viscous cyano gives one time to get the parts in just the right place. 

 

For filler, I'm using Wickes' lightweight filler (surely not as lightweight as hobby filler for balsa?), taking care to minimise its use aft of the C.G. 

 

And you might have noticed this is all happening on the dining-room table... 

 

Just to encourage anyone that might be considering getting into the construction side of the hobby - this is only the 2nd R.C. model that I'm building from plans; other than that I've built a Wots Wot kit and done some balsa ARTF repairs. The great thing about working with balsa is that there's almost no problem that cannot be rectified by either sanding, filling, or insertion of a suitably sized bit of wood (such as the small block you can see in the bottom centre of the last photo above). But I would not recommend the D.510 as a first build, due to the inaccuracies and sketchy instructions. I learned so much from building the Tim Hooper Courier, with his excellent blow-by-blow instructions and education on important but ultimately invisible principles like washout. [No, I haven't finished it - got stuck trying to dream up a scheme for a sufficiently realistic, lightweight, and strong dummy 7-cylinder radial]. When it comes to building the D.510 wings, I'll be borrowing some ideas from Tim's design. 

 

 

[kc]

A really interesting build!

Mark, please don't think that the lack of comments so far means nobody is following this - I reckon many are interested.

 

The plan seems to be the MW3507 from RCModel World April 2011.   Plan is now available from Sarik Hobbies but I see there is also a copy of the original magazine with plan for sale on Magazine Exchange.

[DaveyP]

Indeed I think this is a very interesting build, watching with great interest ?

 

[Jonathan M]

Very nice build Mark - look forward to seeing more!

 

Jon

 

PS - it's always easier to follow builds if there are line spaces between the photos (just a little visual pedantry I admit!) ?

[EarlyBird]

9 hours ago, Jonathan M said:

PS - it's always easier to follow builds if there are line spaces between the photos (just a little visual pedantry I admit!) ?

Agree as then I focus on the individual photo content rather than working out where each photo starts and ends. Just an old brain thing I guess ?

[Mark Honman]

Thanks for the encouragement, everyone. Yes it's MW3507 and the plans are from Sarik. I'm also open to comments along the lines of "next time, do it like this" as I do have a tendency to do things the hard way. 

 

As to putting the pictures in-line, I'll experiment using this post. I can't see any way to insert images in-line, though. What I've had to do is save the post with the picture as attachment, then edit the post rearrange the text...

 

 

I wanted to return to this because I forgot to mention a trick that I used to get the shape of the front of the hatch correct at the firewall. This was to wrap a bit of cling film over the top of the firewall before the upper front fuselage sheet was fitted. Then I glued a balsa offcut between the side sheets, forward of the firewall, sanded it to match the shape of the firewall, and glued the top sheet onto it. Thus when the hatch was cut free, I also cut through the beam, releasing the hatch together with the middle section of the beam, which continues to hold it to shape. 

Edited December 26, 2021 by Mark Honman
trying to insert picture inline

[EarlyBird]

Hope this helps.

Post in the above topic if you need more help.

[Mark Honman]

Thanks EarlyBird, just what I needed!

 

Today's topic is "making up the nose", and it involves rather a lot of "making up". I wasn't confident in getting the right combination of strength and space for the motor with the suggested scheme of building it up from block glued onto the front face of the firewall. I didn't know quite how far to extend the nose, and how to create and blend the various curves in the underside of the nose, its sides, and where the cylinder head fairings meet the nose.

 

I found it really useful to study the 3D modes in the game "War Thunder" where one can move the camera around the plane, and also close-up photos of a 1/32 kit (where one hopes that the rivet-counting customer base will ensure that the manufacturers do a good job). 

 

Fortunately the scheme of extending the sheet sides forward of the firewall paid off. I was able to shape pieces of 3/8" sheet to mate with the inside curve of the sides, and align with the outer face of the cylinder head fairings. 

 

 

 

Although not a perfect fit, with the foaming glue it was good enough. 

 

As you can see I had already started to build up the underside of the nose. I've used strips of 3/8" sheet because I had this plan to leave a hot air outlet hole on the underside of the nose (exiting underwing via the radiator housing). Of course that's pointless because there's an open cockpit further along the fuselage. 

However the use of strips rather than solid sheet means there will be a nice slot for the motor wires, maybe even allowing the ESC to be shuffled well forwards.

 

The next step forward (sad pun intended) was to extend the inner flanks of the cylinder head fairings, and shape another piece of 3/8" sheet to fit inside the "cheek fillers", establishing the right distance to the front of the model, and also filling the "V" at the front of the cylinder head fairings. These pieces have vertically oriented grain. 

 

If I had known at the start, I could have just cut 220mm long inner fairing flanks and saved myself a good bit of time. 

 

Here we are, with one of the cheek-fillers clamped. 

 

Even though the model will be powered by a titchy motor, I've been concerned about getting it into this tight space, so I decided to fit a single 3/8" front "former" rather than the two 1/4" formers F1 and F2 of the plans (so the cheek fillers project to take up the missing 1/8"). I've marked out both their outlines on it to assist with sanding.

 

On the bottom of the model there is a 3/8" cross-beam behind the modified F1,  with some dabs of foaming glue in the corners to fill the missing 1/8". 

 

 

 

 

 

[Jonathan M]

Proper job!

[Mark Honman]

After fitting the bottom of the nose (more 3/8" sheet), the next challenge is to sand the nose to capture the complex curves of the prototype. 

 

There's good news though, which is that the curves are not as complex as I first imagined. To a first approximation, on the lower half of the fuselage they are circular arcs, meeting the front face at around 45% at the sides and maybe 70% below the spinner. The radius of the surface is almost constant. I found it useful to draw a vertical line down each side corresponding to the prototype's join line in the engine cowling - the arcs end at this line. 

 

 

[I've found it very hard to capture the 3D shape in these photos, I'm afraid - there is no substitute for turning the model in one's hands]

 

The shape of the upper half of the nose is dominated by the cylinder head fairings.

 

 

 

This is actually not hard to sand, as their front faces appear to be at the same 45 deg angle that the curved sides meet the front face of the fuselage. The front faces of the cylinder head fairings are straight, with just a small curve where they blend into the main part of the fairing. 

 

But you may have noticed they are not too prominent here, nor in the photo above... 

[Mark Honman]

All that cross-checking against photos and 3-views revealed an important visual point - the cannon cover slopes downward to the nose in such a way that the spinner is an extension of that visual line.  

 

Here's a snip of a historic photo to illustrate the point:

 

That's caused a re-evaluation of the motor centreline and spinner size. I think the motor centreline should be about 4-5mm higher than in the plans, and after comparing 38mm and 45mm spinners, 45mm seems to be about the right size. 

 

At this point I'd like to invoke the wisdom of crowds - what do you interested parties think?

 

 

This particular spinner is a bit more "pointy" than in the photo, but I think it will work out okay. That's because it is secured by an allen bolt in a recessed hole. It should be possible to file the front of the spinner back to the bottom of that recess, round off the corners, and viola the bolt becomes the cannon muzzle. But I expect the proportions will still be a bit wrong. 

 

As a bonus a metal spinner should conduct heat away from the motor.  

 

The final bit of the spinner-choice puzzle is that despite my best efforts even the little Sunnysky X2220 motor is too long to slip between the nose-piece and firewall. The 45mm spinner makes it possible to hide a big enough hole in the front of the fuselage that the motor can pass through when fitted to its X-mount.   

 

Edited December 26, 2021 by Mark Honman

[Mark Honman]

I've subsequently found this page which is the single best set of D.5xx nose pics that I've come across. 

 

http://www.letletlet-warplanes.com/2016/07/08/dewoitine-d-500-510-camouflage-paint-schemes

 

Several interesting aspects can be seen:

• The nose of the D.500 is altogether more pointy than the D.501 and D.510. The cannon-armed aircraft have a deeper chin and shorter spinner. The nose of the D.500 might be slightly extended vs. D.501 onwards. 
• It could be that the prop centreline moved upwards from D.500 to D.501, to accommodate the cannon.
• My 45mm spinner is probably is bit too large, 40-42mm is probably the Goldilocks diameter. 
• It looks like I've given my model's cylinder hear fairings too rounded a top (don't know if I'm going to let that bother me...). The engine cowlings of these aircraft have a distinctly slab-sided look and they seem to have quite a small radius at the outer top of the cylinder head fairings.
• D.510s have two types of landing gear strut, the original D.500 straight strut that joins the fuselage struts at the inner wheel hub (as on the Blakely plan), and a strut with a "bicycle fork" (this is apparently what was fitted to the aircraft sold to China). 
• It would be legitimate to model a plane without spats although the later type of undercarriage is really ugly without them. 
• The D.500 has a slightly different shape to the read of its spats. 
• The pictures all give a good indication of the rear fuselage shape, especially the second photo. 
• The final photo shows an unusual exhaust arrangement with a sharp downward turn in the pipes. This photo might be of the experimental D.510 night-fighter. 

Big thanks to the Armee l'air for painting the aircraft type on the tail...

 

I'm starting to think that the D.500 would have made for a prettier subject to model. 

 

 

Edited December 27, 2021 by Mark Honman

[Mark Honman]

Wow, a month has gone by and not much seems to have changed in the build. I'll blame the resumption of "real work" but every weekend I'm making some progress on the model. So there will be more updates when I get the photos organized. 

 

It seems there has been an incredible amount of sanding and comparisons with historical photos, but the closer I get to completion the more nuances I discover (maybe I just don't know how to draw the "good enough" line). 

 

So this rather out-of-sequence update is triggered by today's effort to find out the real shape of the radiator under the nose.  That led to this rather interesting photo (okay, it's a D.500 rather than D.510) where you can see that aft of the last exhaust port, the sides of the engine compartment curve increasingly inwards at the top. A big and somewhat belated a-ha moment which came just after I had drawn the engine cover panel lines on the model to visualize the relative position of the radiator. 

 

 

It's a very subtle blend...

 

Also one can see that the radiator sides are inset very slightly relative to the cowling.  

 

Here's the model - looking very flat, but as usual sand-able!

 

 

The aft end of the side cover aligns with the position of the wing spar. 

 

I've used finishing resin to get a smooth finish on the nose forward of the side cover - takes lots of little sanding sessions but after a bit of sanding the shiny surface of the resin clearly shows up any depressions.  But it's still hard to see whether the shape is right... probably only when it's time to paint.

 

Edited January 30, 2022 by Mark Honman

[Mark Honman]

Further comment on that photo:

 

The photo also reveals that the fuselage widens towards the cockpit (which chimes with the description in the NACA circular No 180 which states that the fuselage is widest at the cockpit). 

 

Also one can see that the fuselage curves inwards below the bottom of the side engine cover. It's probably too late to capture that on the model!