Pee Wee Horseshoe back plate

Shortened the tube in the back to allow more clearance for the needle valve from the firewall. Filed in some notches in the mounting legs to allow clearance for the needle and the fuel tube. I’m going to add that feature and beef up the other side of the notch to add strength.

This is a 7 minute run. The syringe sucks as a tank since the fuel height affects the draw. I’ll have to mount a decent tank so I don’t have to mess with the needle during testing. Nothing melted on this longer run. Too bad the TD .020 fine thread needle valve assemblies from Ace RC are hard to find. That would be an awesome addition to this setup. I have one I can try.

Cox 25% super power fuel and a black 4.5x2 flexible prop.  



A pic after the run.



Set off the smoke alarm this time.

3:23am? Man, that’s well after last call and you’re doing test runs of a 020 in the house! I mean, there’s nothing wrong with that, I am impressed with your work and willingness to work into the wee hours to accomplish amazing projects! Keep it up!

I wanted to see if it would run with an 80 TPI NVA from a sure start backplate. It runs but the needle setting is very touchy which is expected. Same prop and fuel as previous runs.

Video posted earlier today than well......... today.

Need to back off with the cell phone camera the static noise is messing up the audio app readings for rpm.
Back away and use the camera's software zoom feature so the mic is not so close need some clean audio
for future reference.

'RPM Gauge" for android phones.

Has presets for 2 stroke / 4 stroke and multi cylinders / low rpm and high rpm set to
what you expect to get out of the engine most times 30k is enough for TD engines.

Suggestion for next Pee Wee printer project:

Do a scaled down copy of a Kavan Baby Bee tank extender for the Pee Wee.

ticomareado wrote:Suggestion for next Pee Wee printer project:

Do a scaled down copy of a Kavan Baby Bee tank extender for the Pee Wee.

That would be like the SR-71......leaks fuel on the ground till takeoff.

Here is another run video. Zoomed in and the phone is further away from the prop.

Also the venturi mount failed in the 3rd run. I think vibration broke the mounting feet off.

18,087 rpm much better =)

I don't want to be the one to say it, but a machined mount from aluminum would be the way to go. Somewhat like the Surestart. Maybe some kind of air intake coming out the side for a throttle? Maybe not.  It would be more solid and stronger IMHO.  As a speed guy, an aluminum pan would always be good for a few revs too.  Having said that, maybe a fillet on the back tabs could be designed in on the next one.

Jason_WI wrote:Here is another run video. Zoomed in and the phone is further away from the prop. Also the venturi mount failed in the 3rd run. I think vibration broke the mounting feet off.

Nice going Jason, your experimental proof of concept indicates great potential with a few minor changes, IMO.

It shattered because of concentrated stress point being the near 90 degree angle of the mounting foot. This is why the British Comet jets were discovered to burst and lose cabin pressurization after so many hours of flight. The passenger compartment windows were rectangular. The window vertices at a sharp 90 degree angle in the aluminium frame and aircraft skin provided a high stress concentration point. The constant high frequency vibrations combined with expanding and contracting by cabin pressurization and depressurization coupled with fuselage bending by flight loads lead to material fatigue and weakness. After a couple blew apart in mid-air killing all, the aircraft were grounded.

The entire industry was benefited by the retrofits required to mitigate. This is why all subsequent airliners by all manufacturers now have oval shaped windows.

I remember the DC-9 cabin pressurization depressurization test when I worked at Douglas Aircraft in the 1980's. It had already passed one life time (20 years flying), when the after test bulkhead blew out. They repaired it, went another life time. Then the retired the test.

It's why you don't see DC-9's and derivatives (MD-80's, MD-90's) becoming convertibles in the air. Douglas prided themselves on their structural designs. (Not saying there were not other weaknesses; 1999, an Alaska Air MD-80 crashed when its elevator screw jack fatigued and failed.) It's why you still see an occasional DC-8 flying, now re-engined with turbo fans (basically ducted fan turbine engines). That aircraft was considered to have an unlimited lifetime. You don't see any old Boeing 707's flying.

As is, I gather that you had a very small cross sectional area to carry the weight of the engine (shear) and its bending moment (engine CG times distance to engine bulkhead, in in.-oz. or its metric equivalent). I'm thinking that if you were to move the mount legs inward slightly to a straighter position, provide a radius in lieu of the sharp inner corner near where engine bulkhead screw mounts, and may be even a triangular web outboard of the screw to strengthen the 90 degree at the foot, might be enough to mitigate point fatigue. This will increase the amount of material at the foot to carry the vibration load.
YMMV (Your Mileage May Vary).

Beefed up the venturi backplate. This should prevent the legs from breaking off. Will be a weight penalty by a few grams with the added material. Now that my 3D printer is back together I can resume testing.

Worth the weight penalty if the engine makes 800 rpm + more over a stock Pee Wizzer.

Here it is printed with a .25mm nozzle at .1mm layer height. This is after the upgrade to my printer using .9 degree/step vs 1.8 degree/step stepper motors on the x and y axis.




You can see the difference in print quality in the #3 text on the side of the part. Left is .9 degree and right is 1.8 degree.

The difference is indeed noticable. Either way, those that would like one, I am sure would be happy with either. Have you yet had a chance to test the new one to see if it is better in any aspects of running?

RE: Aspeed's aluminum suggestion. Also the reed valve "check action" would be much better on a smoothly machined aluminum surface

ticomareado wrote:RE: Aspeed's aluminum suggestion.  Also the reed valve "check action" would be much better on a smoothly machined aluminum surface

The reed seal surface is polished with 400 then 1000 then 2000 grit. This pic is hot off the printer without any prep work done to it. Aluminum probably is better but I don't have a 5 axis CNC mill.

A regular 3 axis mill would be fine. Even a standard lathe and mill. The time involved would have a high price. I have an old Hardinge turret lathe that could rough out one half and then the rest could be milled on a standard mill. I would guess a couple hours a piece with all the holes etc. Minimum wage would even be expensive for this method.

I have a lathe and mill also and I would not want to make these in any quantities manually.

I press print, come back 6 hours later and 4 backplates are done. I spend 30 minutes removing the support material and polish up the reed seal surface and they are ready to run. The only two critical smooth surfaces are the reed seal area and the gasket area which is already smooth. Yes the external surface is not as smooth as a injected molded or machined part but it works. I did have failure with the mounting legs. I believe I have that fixed now with the venturi backplate design.

Relative roughness of the 0.9 degree backs is no rougher than some of the bead blasted aluminium crankcases on some engines. I'd say you did well, Jason.

We now see the reasoning why Cox moulded backs initially out of Delrin. The machining was invested in making the mould. The injection process was quick, adequate, low cost, reasonably accurate, and sufficiently durable.