From the Editor of | ||
|
|
|
|
The Maggie MuggsTM low-speed ramjet project - |
August 2003 - TODAY!
(last content update: 24 Jun 2005) |
|
WORK IN PROGRESS LOG | |
Preliminary Design | |
Concept submitted: | 2003-07-10 |
Design complete: | 2003-08-07 |
Working Prototype | |
Construction started: | 2003-08-28 |
Construction completed: | 2003-10-03 |
Testing started: | - Not started - |
Testing completed: | - N/A - |
E X P E R I M E N T A L
--- R E A D T H I S N O T I C E --- |
In viewing this site, YOU ARE AGREEING TO THE FOLLOWING TERMS AND CONDITIONS: |
|
______________________________________________________________ I. O R I G I N A L S C A L E D R A W I N G ______________________________________________________________
Disclosed on Kenneth Moller's Ramjet Forum - 07 August 2003: |
______________________________________________________________ II. W O R K I N P R O G R E S S P H O T O D I A R Y ______________________________________________________________
Photo Diary: How I Built the "Maggie Muggs" Low-Speed Ramjet Prototype at home in my basement, using ordinary hand tools and a handful of discount-store materials, with NO WELDING, for under $60 US - PART I by Larry Cottrill, Editor, jetZILLA Online Magazine - All photos this page Copyright 2003 Larry Cottrill - Table of Contents [Sections I-IV below]: I. The parts lineup [28 August 2003] II. The tubular truss engine mounts [28 August 2003] III. Building the diffuser section [04 September 2003] IV. Bonding the diffuser/flameholder subassembly to the combustion chamber/nozzle shell [12 September 2003] Table of Contents [Sections V-VIII go on to Page 2] V. Final major assembly steps [22 September 2003] VI. Details, details [22 September 2003] VII. Fuel and ignition system work [29 September 2003] VIII. Finishing up [03 October 2003]: Table of Contents [Maggie Muggs Test Firing Page go on to Page 3]: I. Ben Koopman's Maggie Muggs Clone [April 2005] II. Maggie Muggs Testing Plan, Hints and Tips I. The parts lineup [28 August 2003]: Next section Back to Contents Top of page Page 2 Page 3 Subscribe to jetZILLA Online Magazine (it's FREE!) II. The tubular truss engine mounts [28 August 2003]: Previous section Next section Back to Contents Top of page Page 2 Page 3 Subscribe to jetZILLA Online Magazine (it's FREE!) Gluing up an N-strut in my first gluing jig - the opposite strut, semi-finished, and the fully cut rails are to the left. The rails are .5 x .75 x 1/16 inch aluminum angle; the N-struts are cut from the .25 OD titanium tubing contributed by my friend Mark 'Thixis': The N-struts resting against a cardboard jig to align them properly, with just enough J-B Weld to tack them to the rails: The finished engine mounts, fully reinforced with J-B Weld - left mount in foreground, right mount behind: III. Building the diffuser section [04 September 2003]: Previous section Next section Back to Contents Top of page Page 2 Page 3 Subscribe to jetZILLA Online Magazine (it's FREE!) Here's a shot of cutting the groove around the mug shell. The mug is rotated counterclockwise as seen from the top, with the saw oriented as shown [for a left-handed blacksmith, set it all up just the opposite way round]. I oiled the hacksaw blade, but couldn't tell whether this really helped any - it was slow going: I tried starting the cut with the jeweler's saw, but found it impossible to get a good starting cut, so I just ended up working my way around it with the hacksaw until I knew I'd gotten fully through it all around. The piece had to be moved in the vise jaws four or five times: Oddly, the lower part of the shell was not that easy to remove, because of the friction of the insulating foam tightly formed inside: Finally, though, it does break free: The edge requires a few minutes of work with a flat file and a small half-round file to make it safe to handle: When I scraped off the non-skid ring attached to the end, I was pleasantly surprised to find two little holes, so it was easy to thread the jeweler's saw through to get a good start on the cut. I just worked my way around, a bit inside the edge of the flat end. Again, the piece had to be re-chucked in the vise a few times; and, as predicted, I went through 6 or 7 saw blades to get the whole end cut out. Unfortunately, neither the blade nor the progressing saw kerf shows in the picture, since they are about the width of a horse hair. It took about 45 minutes of sawing and changing blades to complete the cut: Even though the jeweler's saw is capable of accurate work, I didn't try for real smoothness or precision. A fairly coarse half-round file was perfect for evening up the hole - this only took ten minutes or so: Then, a much smaller and finer half-round was used to smooth up the opening for safe handling and inlet streamlining: The finished diffuser behind the 'draggy flameholder'. All the small parts of the sink basket strainer are removed and discarded. Note the wonderful array of apertures in this high-quality stainless strainer - a near-perfect 'burner grill': Both the inside edge of the diffuser [large end] and the entire rolled rim of the flameholder have been roughened with sandpaper. A small bead of J-B Weld has been applied to the INSIDE of the diffuser rim. Here, I apply a HEAVY bead of J-B Weld over the whole surface of the flameholder rim: When the diffuser is inverted and lowered into place on this rim, the two beads of cement will blend together to form a smooth, strong 'glue fillet' that will lock the pieces together without the chance of even the slightest gap, with just the slightest bead of J-B Weld visible all around the outside. IV. Bonding the diffuser/flameholder subassembly to the combustion chamber/nozzle shell [12 September 2003]: Previous section Next section Back to Contents Top of page Page 2 Page 3 Subscribe to jetZILLA Online Magazine (it's FREE!) There is a significant gap [a couple of millimeters] between the OD of the rear of the diffuser assembly and the ID of the front edge of the combustor shell. A ring of solid J-B Weld is formed around the diffuser rear edge to a depth of about 3/8 inch [about 8mm] to take up this difference. Believe it or not, the TOP inside plastic rim of the Wal-Mart mug looked like just the right dimensions to work as a mold for this epoxy ring. The trick is to make sure the J-B Weld sticks to the stainless but not to the plastic; so I smeared a thin coating of vegetable oil all over the plastic rim internal surfaces. Then, a heavy coating of J-B Weld was run around the rear diffuser where the flameholder is bonded on, and this is then lowered, flameholder down, into the top of the mug. J-B Weld is added by toothpick to finish filling the gap, right up past the top of the plastic rim: The fingertip is used to smooth out the top edge and remove the slight excess. The thin film of J-B Weld left on the outside of the diffuser shell is of no consequence: After allowing full curing time, the finished molded ring is popped out of the mold by wiggling the diffuser [it comes out easily] - note the inevitable small air voids; these will be individually filled in with J-B Weld during the operation of bonding to the combustor shell: Now the Wal-Mart mug is disassembled, and the top rim of the stainless shell is smoothed with a small file to remove slight burrs. This is a trial fit of the epoxy ring into the edge of the shell. Note the formed ridge of epoxy; the shell slides easily over the main body of the ring, but balks at this ridge, which will assure that the shell is perfectly aligned with the diffuser assembly: The epoxy ring must be cleaned thoroughly with strong alcohol to make sure there is no residue of the cooking oil. Using medium sandpaper to roughen the inside edge of the shell to ensure solid bonding [this area is also de-greased with strong alcohol]: Finally, the entire outer surface of the epoxy ring is coated with J-B Weld and the little voids are all filled in. Then, a HEAVY bead of J-B Weld is applied all around the INSIDE of the front edge of the shell. The diffuser assembly is inverted and the shell is lowered into place around it, until it comes to rest on the molded ridge. This is the orientation it is left in until the bond is fully cured. Here, I wipe excess epoxy off the outside with a dry paper towel; the residual film will be removed by a little fine sanding: Previous section Next section Back to Contents Top of page Page 2 Page 3 Subscribe to jetZILLA Online Magazine (it's FREE!) |
|
"To God Alone be the Glory"
- Johann Sebastian Bach 1685 - 1750 |
The Maggie MuggsTM miniature aviation powerplant
- now under development by