GWS F-15


Until recently electric ducted fan (EDF) propulsion has been a rather pricey and complex way of getting a scale jet into the air often resulting in less than jet-like performance. Frantic four-minute flights with wing loadings that would make the space shuttle seem lightweight have been the order of the day.

Things are moving on though – Lithium batteries have had a huge impact as have brushless motors. Indeed you’d be hard pressed to find an EDF pilot who still uses NiCads.

Another thing that has changed model flying over the past 5 years (for the better in my opinion) is the gradual proliferation of polystyrene based models or ‘foamies’. They’re light, cheap and allow small models to be flown in spaces that would have been unthinkable just a few years ago. Which brings me neatly onto the GWS15 (for all intents and purposes an F-15) a scale-ish, foamie EDF jet.


Enjoy more RCM&E Magazine reading every month.
Click here to subscribe & save.

It’s a great choice of aircraft for a model that needs to look good and be fun to fly. All the F-15 models I’ve seen have been easy to fly as the stability and generous wing area of the design compared to other jets have made it a firm favourite.

All present and correct, the plain box seems to be a backward step compared to previous GWS offerings. Can you imagine anyone wanting to fit those props?



The plain white box is a rather underwhelming sight, a small sticker showing a plain picture of the model. While this did not deter me at all, it’s not on par with other model manufacturers or indeed previous GWS packaging. All of the parts are nicely wrapped in their own plastic bags to stop any damage during transport and all of the heavy objects are separated from the foam airframe components.

With most models there is something that isn’t perfect and in this case it’s the outer edge of the air intakes where the two fuselage halves join. They do not line up properly – not difficult to put right during the build, but annoying still.

Other than this it all looked pretty good. Scattered over the entire airframe are small hexagonal groups of moulding dots, they’re about 4mm across and easily removed with some very light sanding. The GWS15 features moulded panel lines that do seem overdone, they’re a good 2mm wide, quite deep and did make me wonder how much drag they would create.


One thing that becomes clear as soon as you set your eyes on the kit is that it really isn’t all that scale, sure it has a similar outline to the real F-15 but the engines are way oversized, as are the wings and tailplanes, which to be honest, had plenty of area in the first place! This makes the vertical fins and nose look undersized, giving the whole aircraft an almost cartoon aspect. One wonders why GWS didn’t just build a bigger, near-scale F-15?

This is quite a large model for a foamy so the removable wings make good sense. This is achieved by 4 plastic ‘sockets’ for want of a better word, that are installed in the fuselage and 4 carbon rods that protrude from the wings. On one of these sockets, and its opposing rod there is a barb/catch system that locks the wings onto the fuselage whilst allowing for a nice quick release.


Included in the kit are two fan-units of the 64mm size. They are GWS items and have a slight funnel affect on the intake side. The shroud still feels very flexible and could use a spine running around in the same fore/aft position as the fan, just to ensure the shape is retained. GWS also include servo and motor/ESC extension leads would you believe! Two GWS brushless 2028 Green 3900Kv brushless inrunners are included and just require the additional purchase of two 25amp ESCs, I opted for two GWS units.

There is an option to build the model with a fixed tail and conventional elevators or with full moving tailerons. Fixed undercarriage components are in the kit but wouldn’t be something I would be using. Finally, some large three blade props (would you believe) are also included although it’s difficult to imagine anyone seriously considering using these in a pusher prop-jet configuration.

Just to touch on batteries for a second. This F-15 is designed for the popular 2100mAh 3 or 4-cell Li-Po packs that now fly a multitude of models. The motors and ESC supplied are rated up to 4-cell, so obviously that is the way to go if you want all-out performance. However I decided to conduct the initial flights on the more commonly found 3-cell packs.

The instructions, which are of entirely pictorial form, remind me of something from an Airfix kit, there are no real written instructions and where there are, they’re badly translated. Thankfully it’s quite easy to glean all the information you need from the images.


Whilst not going into too much detail, I will just highlight some construction steps.

It’s best to assemble and test the power set-up before starting on the airframe. Initially everything went together smoothly until I tested the fans. The fan blade clearance was fine when stationary but as soon as I started the motors the blades rubbed against the shroud. To fix this I spent a fair while re-centering the motors in relation to the fan shrouds. Once I was satisfied I’d got the motors as central as possible, I then tested them again. While they were both slightly better they were still binding above half throttle. As far as I could see this was due to the blades stretching slightly under centrifugal force or the shroud changing shape due to the airflow, either way it was annoying. I gradually took more and more off each blade-tip using some emery paper until I got to the point where there was no longer any binding at full throttle. I then set about balancing the blades using my Dubro prop’ balancer. Amazingly even after my wrangling, both of them seemed to be pretty well balanced, a small dab of paint on the light blades is all it took.

This F-15 has what’s known as ‘cheater holes’. These are large openings in the topside of the fuselage which provide more airflow to the fans and therefore more power at low speed. They are only really needed if running a low power set-up as they will create a lot of drag once up to speed. I blocked these off with the clear plastic covers provided for the exterior, these fit exceedingly well but they leave a large gaping hole on the inside top surface of the intake. I surmised that this may not be good for the all important intake airflow, so decided to cover the holes using some very thin plastic card.

Talking of glue, GWS provide their own glue, albeit with no instructions as to use. I tried using it in several different ways and none of them really worked. I was left with a soft, apparently non-setting, chewing gum-like glue that really didn’t want to stick anything. I tried leaving it to go tacky and then letting the two surfaces contact each other, but this made no difference, I could easily peel them apart about an hour and a half later. After all these experiments I decided that I would use foam friendly cyano and possibly 5/30min epoxy for any load/stress sections.

I needed to glue the wing mounting sockets into the top half of the fuselage before I could install the servos and the extension leads. Thirty-minute epoxy was used for this and provided a supremely strong bond for what is an important structural component. I finishing the wings by gluing the wing spars and the wing mounting tubes into place, again I used 30-minute epoxy.

The task of gluing the fan units into the top half of the fuselage was achieved with foam friendly cyano and I joined the fuselage halves with this also. It is certainly one of those moments where you need to be decisive, if you make a poor attempt at it you will most likely end up with the glue going off before everything is in the correct position. Luckily everything went together surprisingly well.

Once the glue had set, it was time to test the fan units and servos to make sure that everything had gone according to plan. The fan units spooled up and down very nicely, there were some slightly odd harmonics going on between them but they weren’t binding which was the main thing! As soon as I started checking the servos it became clear that only one of the servos was responding. I double-checked everything I could think of, including changing the receiver for another I had lying around, but with no effect. There was no other choice but to take a scalpel to the hindquarters of the F-15 and try to extricate the servo. This was accomplished relatively quickly and the problem identified as an intermittent loose plug connection in one of the GWS extension leads supplied with the kit. I used the old extension lead to pull a new Futaba one through the fuselage and for piece of mind repeated the process on the other side. While the operation left a few scars, I felt a lot happier.

The fins came next. Some large black plastic fittings are provided that glue to the fuselage and allow the fins to be bolted on. I found these to be unsightly but storage and transport wasn’t a problem for me so I decided to glue the fins directly onto the surface using 5-min epoxy, this worked well and looked a lot better.

The all-moving tailplanes screw directly onto the servos using a special servo head attachment. I must admit that I didn’t particularly like this way of doing it. It must be putting a lot of stress on the head of the servo and seems a poor piece of design. I would have preferred some kind of simple axle to take the flying load and then have the servo moving an arm attached to that. I also noted that at full elevator deflection, the leading edges of the tailplanes are below the underside of the aircraft – something that would lead to carnage if they caught the ground when landing!

Moving towards the painting stage, the final part of the main assembly involved cutting out the canopy and then gluing it to the foam cockpit hatch-cover. This went pretty smoothly and I could have had it done very quickly but I decided to spend some time on the cockpit just to add a little detail. A small HUD was made from some clear plastic, and plastic card was used to create a dashboard shade like a real F-15. This is when I realised that I had no pilot! I really wanted to get this part of the build finished as soon as I could, so decided to carve one out of foam. I didn’t do a particularly good job, but it’s better than not having a pilot at all. I glued the canopy on having first applied the waterslide decal for the dashboard. I used the GWS glue for this and for once it worked!

I’m rather pleased with my paint job!


With the main construction finished it was time to get on with the part I really enjoy – the painting. I would be using my airbrush for this and spraying Tamiya acrylics. Tamiya paints aren’t the hardest wearing of all acrylics but they are easy to find and come in a very wide variety of colours.

I have no real love for the standard grey-on-grey U.S. military scheme, so decided to go with sand and brown ‘Aggressor’ colours from an aircraft based at Nellis AFB near Las Vegas.

The sandy colour went on first as a base coat using a wide-area covering nozzle then I swapped to a fine-line nozzle and painted the outside edges of the brown camo’ finally swaping to a medium area coverage nozzle to fill in the brown camo areas outlined. The whole process took about 4 hours including drying time.

One thing that I don’t like about painting foam models is the inability to mask anything off – masking tape invariably pulls the paint off when removed. Even de-tacking the tape beforehand doesn’t help much. Because of this, I brush painted the engines and the large ‘metal’ panels on the underside. These were done in aluminium, the engines with gunmetal and titanium.

The final painting stage is the one I enjoy the most and that’s weathering the aircraft. I went over the model in black with a low flow nozzle whilst using real F-15 reference photos as a guide. The engines and the underside panels received a lot of attention at this point and I’m really pleased with the result. The finishing touch employed a metallic blue to give the underside panels a heated metal look. I painted the nose cone separately and stuck it on at this point. The waterslide decals supplied with the kit seemed to be of reasonable quality and went on well. I particularly liked the band of yellow and black that wraps around the top of the fin.

The only thing left to do was to install a Futaba 146 limited range PCM receiver and the flight pack to see where things balanced. I used self adhesive Velcro to attach the receiver to the side of the cockpit in front of the divider. I could not put it where suggested in the instructions as the two ESC’s take up all the available space. With a 3-cell pack it balanced about 1cm back from the advised 160mm point, but with a 4-cell pack it was near spot on. I attached some lead to the front of my 3-cell pack and this seemed to bring the C of G into the correct position. With that done it was time to set up the controls. I was quite shocked with the 45 degree pitch control and 30 degree roll control suggested. This seemed a little excessive, but I followed the instructions anyway. I did however set my usual 60% rates on both as a back-up in case the model was too sensitive.


A lovely calm evening came along soon after I’d finished the model – perfect for the first flight. With the help of my father I completed a range check with and without the motors running, checked the control movements and direction, checked that both motors were running and that both would start when asked to.

Here I noticed something slightly odd. Every 4-5 times I opened the throttle, at least once a motor would not start at all. A different motor was affected each time. This concerned me a little but with the first flight looming I pushed it to the back of my mind (a mistake) and got on with the pre-flight checks. After changing the pack for a fresh one it was time to see if this ‘15’ would fly. With my father holding it over his head pointing into wind, I selected full throttle and gave him the nod, a second later it was away… and descending, with not a great deal of thrust. Luckily I was flying from home and there is a good 30-foot drop between the hill we were launching off and the lower field where the model was now heading. With the throttles still fire-walled, I gently turned it away from the trees at the edge of the field and tried to coax a climb. Over the next minute or so I managed to climb it up to altitude so I could at least relax and plan a landing approach.

While at height I tried to point it directly towards and away from me a few times and sure enough I could see that one of the fans was stationary. The fact that it was maintaining altitude on one engine was amazing. With that big wing I knew that speed would not be a problem when landing so I throttled back and started a descending downwind leg. Turning finals I blipped the power again just to stretch the approach and then throttled back and gently flared. The F-15 floated on for quite a while before gently settling into the long grass.

Phew….. It was getting dark so I decided to quit while it was still in one piece. Back in the workshop I grabbed the speed controller instructions and proceeded to re-calibrate both of them for the throttle-stick position, just in case that was the cause of the problem – it did seem to fix it.

Later that week I flew the F-15 again on 3-cells with two motors running this time. The performance was okay but not ballistic by any means and the climb quite gentle. Fun could be had once the model was wound up. That huge wing meant that speed bled off rapidly when too much elevator was pulled in. Loops required a dive beforehand to make it over the top. I checked the stall – trying to hold the nose of the model on the horizon at the stall just caused it to nod forwards and begin flying again. If I gradually fed in more and more back-stick as the speed decayed it started mushing downwards with a slight tendency to drop a wing, if I released the elevator the model would unstall itself and begin flying again. All-in-all, very benign.

I was eager to try it on 4-cells. Extended climbs of about 30 degrees were possible now and I could really throw it around but when I pulled to the vertical it ground to a halt in an alarmingly short space of time. Any speed lost in the turns was quickly regained in level flight and loops could just about be flown from level flight too. A fast low pass followed by a climbing Derry turn brought a smile and even though it’s not perfectly scale, the model was quite pleasing to look at.

After beating up the circuit for about 8 minutes I noticed a slight drop in power and decided to land. As before, landing was a non-event, that large wing slowing it down nicely and creating plenty of lift at the flare.


A week or so later it was time to fly again. Launching the model is quite easy, a half decent level throw is sufficient, although the lack of hand-holds wont help if you’re holding a transmitter as well. This time I could see that one of the motors had not started up just as soon as the model had left my hand.

Unfortunately it was quite breezy, nothing that would have stopped the F-15 from flying on two motors. But one motor wasn’t going to win in a fight with the wind. The aircraft quickly got on the wrong side of the drag curve during a turn. As she turned downwind (towards the landing area) the rate of decent increased substantially and the model mushed down in a deep stall and landed rather hard into a thorn bush. The damage was repairable, and it will fly again but the paint scheme was in tatters!


I’ve come up with various different solutions for this single motor start problem. If I was building it again I could replace the two speed controllers with a single controller of the 60A range whilst accepting that not all ESC’s will operate two motors. The other solution I have been advised to try is to use the throttle trim to start both motors at idle before launch, confirm that they are indeed both running and then leave the trim there and fly as normal thus not allowing the motors to stop and be re-started. Once both motors are successfully started they should stay running so the theory goes.

As you can see I’m still experimenting with the GWS15 but where do we stand so far? That the model isn’t suitable for novice pilots or as an introduction to EDF flight should be obvious – there are better performing alternatives available. Although I did enjoy flying the ’15’, the problems with the speed controllers and fan-units have distracted from the fun. It is easy to fly, the advised control deflections being just about perfect but the model does not feel as precise as I’d have expected which I think that may be due to the fully flying tail. The conventional tailplane set-up may be better in this respect. The 4-cell set-up is the way to go I feel, a 3-cell pack will fly it, but not with any real authority and you don’t have any reserve power. Saying that, even on 4 cells, the performance isn’t really jet-like.


  • Name – GWS15
  • Type – Twin ducted fan foam jet
  • Made by – GWS China
  • UK distributor – J.Perkins Distribution Ltd (01622 854300)
  • RRP – £69.99 (includes motors)
  • Wingspan: 900mm (35.4in)
  • Wing area: 23.9dm2 (370.45sq in)
  • Length: 1050mm (41.3in)
  • Flying weight: 850-1000g (30-35.3oz)
  • Wing loading: 35.5-41.8g/ dm2(10.7-12.6oz/sq ft)
  • Servos used: GWS PARK x 2
  • Motors used: 2 x GWS 2028 3900Kv brushless inrunners
  • Flight packs used: FlightPower 2170 3S 11.1v 25-50C and FlightPower 2170 4S 14.8v 25-50C

    Editors note: This model was supplied direct by GWS themselves and not by the UK distributor. The kit received did not contain motors but the model now available in the UK and listed in the datafile does contain two brushless motors. ESC’s must be purchased separately.

  • Article Tags:

    About the Author