Design Notes

[Andy Blackburn]

Thought it might be useful to give a bit of background to the Jet Provost design for those interested, and suggest a few minor modifications to the design after a few years of experience with it.

The Jet Provost was originally designed in about 2004 - when I had a lazy left thumb - for ailerons, spoilers, elevator although it could be persuaded to do fairly reasonable stall turns (into wind) on ailerons alone. However, I’m currently building another one (after unwisely selling the original JP to Scott Edwards at an ISA bring and buy) to a slightly revised plan, it has a rudder and so did Terry Lidstone’s example; I think it flies better with one.

The wing section is Selig 7055; a number of thinner and more efficient sections were tried but to be honest, the front view then just looked like a whale with a razor blade sticking out each side. Eppler 205 doesn’t really work that well on a tip chord that small (less than 4 ½ inches, Eppler 374 was rejected for the same reason. I nearly used Clark Y but chose S7055 on the basis that it was designed for thermal soaring and should therefore work well at low Reynolds numbers (this is a small model); happily, this has turned out to be the case. It has very good manners at low speed, this is partly down to the section and partly because it has a small - but very sensible - amount of washout; please don’t leave it out!

It goes better in light lift with the tanks off, and a bit further down the page I’ve added a plan fragment showing some properly-shaped plug-in wing-tips which will look much better and improve performance further by reducing tip drag and decreasing span loading.

There are a couple of deviations from scale, and some minor simplifications;

• The wings are enlarged very slightly in span and chord to give a wing area of 8% over-scale; not noticeable, but every little helps. The tail surfaces were enlarged by about 5% to keep everything aerodynamically and visually in balance.
• The wing root fairings have been left off; this makes the airframe a little bit draggier, but it’s simpler to build. Adding wing root fairings is straightforward but is a bit more work, although it does improve the appearance slightly.
• I can’t honestly remember whether the intakes have been thinned in plan view; on the basis that it looks fine as it is, I think that if they have been thinned, it’ll only be by a very small amount. I seem to remember that the tip tanks were also thinned by a very small amount.

The prototype was built as light as could reasonably be managed and weighs 24 1/4 oz ready to fly including 1/4 oz nose weight, giving a wing loading of 14.3 oz/sq ft. Terry Lidstone’s example had flaps, fully detailed cockpit, sliding cockpit canopy and about 3 layers of finish so it weighed an enormous amount (about 40 oz!) and felt as though it was doomed to failure when holding it at the edge of the slope, but it flies in a very similar manner to mine – maybe a little bit faster, but not a lot. However, if you want it to catch thermals, I’d suggest you keep it light.

 

Edited By Andy Blackburn on 01/01/2014 23:28:09

[Andy Blackburn]

Suggested Modifications

Having flown the original for quite a while and ruthlessly taken notes when other people have had prangs, I’d suggest the following minor modifications:

Its Better with a Rudder
If you want a simple model, it’ll be fine on two functions with one servo per aileron and will even – with the correct differential – do a passable(ish) stall turn. However, it’s more fun with a rudder. If you decide to fit one, I suggest an external horn (the rudder outline is marked on the plan) with the snake exit about 35 mm forward of F7, about 12 mm left of the fin. Mount the servo as far forwards as possible on the opposite side to the elevator servo, and support the snake every 2-3 inches.

Wing Bolt Fixing
Scott Edwards (Jet Provost) and Matt Brett (Macchi 202) had a mid-air a few years ago which caused the JP wing bolt to pull right through, rather than break as intended. The wing bolt head was quite small in relation to the thread, so I’d suggest the following minor modifications:

• Change the 0.4 mm ply bottom reinforcement to 0.8 mm;
• Use a steel washer under the bolt head.

Top Rear Decking Support
I’m just building a replacement JP and I think it really needs a scrap balsa support ( maybe 1/8” – 3/16" ) at the top-rear edge of F6 to support the 3/16” top rear decking, which is in two halves either side of the fin strake, otherwise there’s a tendency for it to dip in the middle.

Plug-in Wingtips
As discussed, the original model flies better in light lift without the tip tanks, and this can make the difference between it getting away and not. I’d therefore suggest that some plug-in wing-tips are used for those light-lift days when you’ve chucked it in the car on the off-chance. Here’s a (hopefully) full-size drawing:

Aileron Torque Rods
You might find it easier to shape the ailerons if the aileron torque rods that project into the (solid sheet) ailerons are cut-back to about 18-20 mm; I don’t think they really need to be 25 mm long (as drawn) and the holes for the shorter ones are less likely to break the surface of the aileron.

Wing Reinforcement
Terry Lidstone had to land his (heavy) JP out in a field at the bottom of Crooke Peak and - not surprisingly at that range - it cart wheeled and broke the wing at the dihedral joint. As insurance, I’d suggest that the wing root be given some light reinforcement, something like a ½” strip of gauze bandage fixed with PVA ought to do it.

Tail Surfaces
I had to dump the prototype JP on a path on the north face of the beacon one day when some walkers unexpectedly appeared during a landing approach (we have to be careful at Ivinghoe because it's a public site), which cracked a tailplane half, so I now routinely reinforce sheet tail surfaces with full-depth strips of 1/16” thick spruce – in this case, 3/16” x 1/16”. Terry Lidstone followed his usual practice of replacing the flat-sheet tail surfaces with built-up sheeted fully symmetrical surfaces (SD8020, I think) and it certainly had a nice control response, but it’s rather more work than I was prepared to put in. Something to think about, though…

Tip Tanks
I haven’t seen the laser-cut parts but I understand that Traplet have decided to “thicken” the cut tip-tank parts slightly; all the original models had tanks that were OK and fitted properly – they were a tight push-fit onto the tips. However, if (for any reason) it turns out that the tanks aren’t enough of a push-fit onto the wing-tips, a good fix is to epoxy some 6mm x 2mm magnets (from Maplins) into the wing tip and wing tank; if the wing-tip magnet is flush, you’ll still be able to swap the tanks for streamlined tips as discussed above. It’ll require some care to make everything stick properly, I suggest the following process:

1. Roughen-up the gluing face with a file or emery paper and clean with alcohol or acetone, and
2. Use a small amount of epoxy first to get everything to stick in position, and then fill in the gaps with more epoxy when the first lot has set.

Edited By Andy Blackburn on 01/01/2014 23:27:42

[Monz]

Hi Andy, thanks for the details on your JP. Makes for interesting reading!

[RedBaron]

Woops should have read this thread before posting on Phil's intro thread.

Rgds

RedBaron

[Andy Blackburn]

<cross-posting from phil's intro thread>

Redbaron wrote:

That said it just has to have a rudder for me. Maybe glue hinged, but haven't quite figured out how to keep rudder actuation hidden yet. Maybe a removable section in the rear bumper area to get at control stuff easier?

It's possible to have a hidden rudder horn, but you haven't got a lot of space to work in (about 15mm between the underside of the tailplane and the elevator snake). If my rudder horn was going to be internal, the approach I'd consider is:

1. Permanently mount the rudder torque rod and clevis + snake below the tailplane; it'll need a slot in F7 to clear the actuation horn.
2. Make a long thin slot in the tailplane to let the torque rod through so that it can be glued on after covering
3. Make the fin built-up and a bit thicker (maybe 1/8" x 1/16" structure with 1/16" skins) and a slot in the bottom to let the torque rod through; the fin strake slot in the top rear decking will need to be a bit wider as well, of course.
4. Make the rudder like the ailerons.

[RedBaron]

Thanks Andy, maybe keep it simple as you say, will save hidden control for a twice size one?

Rgds

RedB

[John A H]

On the subject of weight......... I've just put my version (which we flew at the Great Orme and Lleyn PSSA meetings this summer) on the scales and it's a bit of a bloater at 37oz . I think this is due to two things; not been very careful about wood selection (the quadrant I used was quite heavy) and the amount of tissue, dope, talcum powder and paint I put on the fuselage. In good lift it helps it penetrate but it's no floater. The video below was taken at the Lleyn PSSA meeting in August (filmed by Rez) and gives an idea of it's flying characteristics.

My version does not have a rudder but you might be able to offset the elevator snake/torque rod slightly to one side to avoid it clashing with a torque rod from the rudder.

[Andy Blackburn]

Nice model, that, John.

The prototype was covered in Solarfilm,Solarspan & Profilm, and I seem to remember that I went to the local model shop with some digital kitchen scales and selected the lightest wood they had (6 to 6.5 lb/cu ft) for the fuselage sheet, triangle section and tail surfaces. I also think there was about 1/4 oz of nose-weight, although that was with a 35 MHz Rx and a trailing antenna, so I'd expect to get away without any for a 2.4 GHz set-up without a rudder.

At just over 24 oz it would thermal quite happily - but it's tiny, so it's easy to lose sight of it.

Edited By Andy Blackburn on 08/01/2014 13:25:34

[Andy Blackburn]

Sorry, one other thing - aileron torque rods.

The plan has 3mm carbon tube epoxied inside 1/8" aluminium tube, which works really, really well - the carbon tube is useless in torsion but really good in bending, and the aluminium tube is the opposite. If you can't get 3mm carbon tube, 1/8" dowel sanded down to about 3mm will work nearly as well, the only issue is that you have to be a bit careful when drilling the holes at each end for the wire attachments; I'd suggest a pilot hole with a scriber and drill by hand with a pin-vice.