3DlabPrint F-86 Sabre - a short lived Model

[Gil.E]

With the 7 days grieving period over, I was contemplating posting a retroactive build log for this model, however going through the photos and videos was too painful. Too soon. Too late. Queue in violin music.

So here is a single post summary about the build project and a bit about the model itself.

This was a collaborative project taken by Carlos (Co-printer/builder), Jeff (test Pilot), and myself.

*  Project Duration: 2 Winters (November 2018 to last Thursday)
*  Combined Printing time Estimate: 400+ hours - Can be accomplished in a month of overnight print runs. Printing was the easy part.
*  Actual Build time Estimate: 80+ work hours.


"Model Plan" or Printing files can be purchased here:

https://3dlabprint.com/shop/f86asabre/

More information can be found in builder/user guide document linked below:

https://3dlabprint.com/wp-content/uploads/2018/10/3DLabPrint-F-86A-Sabre-user-guide_.pdf

Key dimensions:

Length: 1427mm
Wingspan: 1427mm
Wing area: 45.76 dm2 / 4.92 sq feet
Wing loading: 104 g/dm2 / 28 oz/sq feet


Particulars specs for this Build:


Take off weight: 5.5Kg

Power System:

* 2500W peak power
* 12 blade 90mm Changesun EDF
* 200A 7S Opto-ESC

* Final System Battery: 3S, 850mAh with Hobbyking 8A BEC located at front

* Tried and failed System battery: 2S2P 2Ah custom Li-ion pack (4 x 18650 cells) located inside the fuselage tail section, behind airbrake bays. Remained in model to maintain optimal CG, acting as 160g tail weight.   

* Initial Battery: 6S, 5800mAh ( 2 Series 3S packs) - Tested and failed to take off.

* Flown with improvised 7S 3300mAh (3S, 5800mAh in series with 4S, 3300mAh)


Radio System:

* Full range FrSky X8R 2.4G 16CH SBUS Smart Port Full Duplex Telemetry Receiver Coupled via SBUS to  FrSky XM+ Mini Receiver for redundancy.

* Frsky FLVSS Battery sensor connected via SmartPort

* 8-channel SBUS-to-PWM Expander


Programming Setup - 14 Channels used as follows:

2xCH: Right Aileron & Left Aileron - Y Cable on single channel can't be used. Symmetrical servo bay printed in wings requires a reversed signal for each.

2xCH: Right Elevator  & Left Elevator  - Y Cable cable can't be used, servo arrangement in fuselage requires reversed signal for each.

2xCH: Right Flap & Left Flaps - Y Cable on single channel can't be used, symmetrical servo bay printed in wings requires reversed signal on one side.

2xCH: Right & Left Air Brakes - same problem: Servo arrangement in fuselage requires signal to be reversed on one servo.

3xCH: Nose and Wing retracts: Programmed  on 3 individual cloned channels to avoid the clutter of a 3-way splitter cable. Can be used in staggered delay operation to reduce inrush current 

1xCH Rudder 

1xCH Nose LG Steering Servo

1xCH Throttle

The use of an individual programmable channel, per each one of the 10 installed Servos, alleviates the need for fine mechanical trimming and centering.  Each control surface is  programmed (electronically trimmed) with its own center point and min/max endpoints right on the model by tweaking individual control curves in OpenTx (T16 Pro radio).   

Cost of Materials:
A crude Hardware Bill-of-materials for all electronics, construction materials,  plastic used (including bad prints waste material) sums up to about $650 CDN. Somewhat cheaper than an equivalent 90mm EDF Jet from MotionRC before Tax and Shipping.

Field Deployment
The model made 3 evening appearances at TEMAC with the intent of performing test flights prior to completion and painting. On the first visit we performed basic Taxi runs and 2 takeoff attempts, which failed with the model reaching the end of the runway before reaching its takeoff speed. Subsequently the power system was upgraded from 6S to 7S and the landing gear setup was reinforced.

On the second field visit the model was maidened and flown successfully, twice.

On the third visit the model fell victim to the 2020 Geotex runway:   On takeoff, the model hit a bump (or shall we say a 'pitfall'), skidded to the side of the runway, then picked up by a gust of wind which rolled it 360 degrees and crashed it breaking both wing tips and damaging the front fuselage section. Based on the OpenTX telemetry SD card log, the ailerons had nothing to do with bizarre fatal roll 1 meter above ground.

Cost benefit analysis:

The model flew two times for a total duration of 6 minutes.

Cost per flight: $300

Cost per flight minute: $100

Flight time to Build time ratio (Estimate):   6 flight minutes / 4800 construction minutes   =  1:800 (also known as Time kill ratio)

Entertainment value provided to field spectators: priceless.


Attached here are some photos taken along the way.


Finally, here is a youtube link to a short video Jeff put together using WhatsApp media gathered from Temac members:


https://youtu.be/v2mc_qkWj7I

[Andy Hoffer]

Hi Gil, Carlos and Jeff,

I am truly sorry to see this casualty of COVID-19 which precluded our installation of a new runway this season which might have prevented this tragic loss.  It is a beautiful plane in every respect, and you are to be applauded for your fine team work in bringing this superb plane to fruition.  The ghost white is eerily prophetic.  My heart was in my mouth watching the twitchy maiden flight.  Bravo Jeff!!

It would be interesting to hear your analysis of the crash survivability of this construction technique and design, given the apparent high take-off (and landing?) speeds required by the high wing loading.

I look forward to your posts on the repair process and to seeing your prized creation fly again.  I hope you draw inspiration from the two successful test flights.  The second flight was magnificent.  So glad you had these videos and still photos, and thank you for sharing this with us.

Andy

[GuyOReilly]

Good Morning
I have a project in the works about a Cessna 150 to be 3D printed.
So sad about your loss..  perhaps parts could be re-printed and thus be repairable?
Guy

[davidk]

Admirable effort and success, guys.  You are part of the pioneering for this type of construction and fabrication, so, you are to be applauded.

I'll be printing a plane one of these days... and so appreciate the effort and documentation you've provided.

[bfeist]

Nasty snap. Maybe some flaps on takeoff?

Put it aside and I'm sure once you're feeling industrious again you can take a blade-tipped soldering iron to it and print the replacement sections you need. The advantages of 3d printing!

[Gil.E]

   Thank you all for the kind words and encouragement. I am sure this model will eventually come back to life - especially if enticed by a new flat ironed runway in the future

Yes, the next step is to wait for the first snow fall and experiment with repair methods. Re-printing the entire model is doable but not an attractive proposition.

Removing sections of such models is tricky to say the least. All fuselage and wing segments are glued together with CA glue. Here is a starting point for some reading on the subject:

https://3dprinting.stackexchange.com/questions/2772/how-to-debond-cyanoacrylate-glue-from-pla

While Acetone dissolves ABS plastic,  it's supposed to be harmless on PLA (the model is printed with eSUN PLA+ filament).
I am not fully convinced using a relatively non-toxic material like acetone won't weaken the plastic parts as well.

[ one hour later....]

I just run a quick experiment on two bonded wing sections from an old crashed trainer. See attached photos.

Some key observations:

1.   The parts are glued on both internal and external seam lines. Acetone has to be soaked into the joints over multiple applications using a brush or a syringe. It took 10 minutes of repeated wetting and slowly forcing the two parts apart.

2.   White residue forms on the plastic surface after acetone is evaporated off it. I assume it suggests  there is some interaction between PLA+ and the acetone, although it seems minimal and cosmetic.

3.   Separated part edges did not turn brittle or dissolved. So far this looks promising and smell nice too...

Next step:  wash the parts with water, dry them and attempt to re-glue them with CA  to see if we get back the original bond strength

[sihinch]

Gil and team, you are truly pioneers. I admire your vision, patience and resilience.

I wish you very very good luck. And say "keep going! Bravo!"

[Frank v B]

Gill and Team,

Bravo.  Full marks for trying.  I love the analytical stats on the original post, especially the Time:Kill ratio.

Your quote comes to mind that you mentioned during your presentation at the Pilots' Meeting a few years ago.... "If you want to get into 3D printing model airplanes, check the status of your marriage first"

Put all crashed airplanes into a corner and they will start to repair themselves.

Get back on the horse.

Frank

[Andy Hoffer]

Gil (@Gil.E), Jeff (@flying saucer), and Carlos (@Palkina),

OK guys.  You've proven the design.  You've had a couple of great flights.  Now it's time to really scale up your design.  Here are some good EDF power plant deals for you.  Pick one, scale up your design to fit, order 100 lbs of filament, and let her rip!  (Put the printers on auto, take a 2-week vacation, and when you come back all the parts will be waiting for CA.  Call me to do photos BEFORE you fly. 

https://www.rccanada.ca/rccforum/showthread.php?t=596779

Andy

[Gil.E]

Thanks for the vote of confidence Andy, but you should be aware of our team's budgetary constraints.

While we try to strive and maintain a "skunkworks like" modus operandi, we are not supported by any defense department contracts. The beautiful 128mm EDF fans you suggested are above and beyond our meager means.

We are more likely to hoard stockpiles of broken vacuum cleaners and leaf blowers to supply spare parts for our next EDF project 

[Gil.E]

Get back on the horse.

Thanks Frank!
We already fully recovered and moved on to plan the next wreck.

You had me confused for a moment with the horse comment. Almost thought I posted in the wrong Club type.


PS: Horse Riding Club rates are also out of our budget range...