Another Radial - this time 18 Cylinders

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Wow Terry. it is looking really good. I am getting antsy to hear it run.

I know if it were me I would start it , and worry about boots later :) The patience you exhibit is as impressive as the build.
Not to mention the time you have taken to document it for all of us. It is truly appreciated !

Many thanks
Scott
 
I drove around to the local auto parts stores and purchased all the Dorman #47408 1/8" x 7/32" right angle vacuum fittings I could find because, to me, they looked like T-18 spark plug boots. Then, I experimented with various modifications to adapt them to both the CM-6 and the 1/8" diameter 20kV plug wire that I'm using. This wire is distributed by S/S Machine & Engineering.
Straight out of its package, the small diameter end of this fitting is a perfect size for gripping and strain relieving the 1/8" diameter plug wire. However, a simple shop-made insertion tool is needed to actually get the wire into the boot. The i.d. of the body, though, is too small to fit over the CM-6, and so it must be opened up some. This is much easier said than done in rubber, and I was only able to come up with a passable solution using tools I had on hand.
There are several ways, involving various degrees of complexity, to make an electrical connection to the top of the spark plug. The simplest way is to pull about four inches of wire through the boot, strip off 2-1/2 inches of insulation, and then coil the bare wire up into a small pillow. The wire is then pulled back through the boot until the pillow is at the bottom of the boot where it can contact the electrode when the plug is pushed into the boot. The boot will grip both the wire and the plug tight enough to maintain a reliable connection on practically any model engine. I torture tested a few boots made up this way because I was really interested in using this simple connection for my application. What I found, though, was that after twenty or so installs the ultra-fine strands making up the 1/8" diameter plug wire began to break away. I was more worried about this debris finding its way into a combustion chamber than I was about the integrity of the electrical connection. For most users where the plugs won't be frequently removed from the engine, this simple contact using this particular wire is probably very adequate. In my case, though, several plugs will frequently come in and out of the engine; and so I wasn't entirely comfortable with it.
After trying unsuccessfully to make compression springs work, I finally arrived at a tubular contact design similar to that found in plug wire kits for early era custom cars. The sequence of photos shows the steps I followed to make up four plug wires using the last four fittings left over from my experiments. Additional fittings for the rest of the wires are on order from an online supplier.
I started by sliding the boot onto a 3/16" diameter mandrel in my lathe and cutting off a portion of the body to give the boot a final height of 3/4". As mentioned earlier, the i.d. of the body must be opened up to fit over the plug body. The next photo shows the three tools I used to do this: a 1/4" rougher, a 5/16" rougher, and a 9/32" annular cutter. I found that 3-flute cutters produced a nicer hole than the 2 and 4-flute cutters that tended to create polygons. I manually held them in end-mill holders to reduce hand fatigue and marked the depth of cut directly on the cutter to avoid going too deep. A bi-directional twisting motion was used to grind away the rubber and open up the hole. I found it useful to first chill the fittings in a freezer for an hour or so. I started with the annular cutter using the rear of a drill bit as a pilot to get things started straight. I then went deeper using the 1/4" rougher, and followed that with the 5/16" rougher. I tried to grind a special purpose cutter, but I wasn't able to improve upon the grinding action of these cutters. When the boot slid onto the plug with the 'right' snug feel, I stopped and went onto the next part.
The contact I designed starts as a length of 9/32" o.d. brass tubing purchased from a local hobby shop. After cutting a .360" long piece, both ends were slightly beveled by manually rotating a 45 degree countersink in them. The bevel on the top end will later help guide the contact onto the plug. On the mill, a 1/16" wide slot was cut along the entire length of the tube. As shown in the photos the corners were chamfered using a pair of sharp side cutters. The bottom-end chamfers are wide enough to provide a clearance slot for the plug wire. An .040" diameter hole was drilled for the plug wire by passing the drill through the slot and drilling close to the bottom end of the contact. A half-round file then cleaned up all the sharp edges. Finally, the contact was slid onto a #3 drill and pliers were used to carefully reduce the diameter and close up the 1/16" slot while maintaining the circularity of the contact. The chamfered wire clearance slot was re-checked.
To prepare for soldering, the wire insertion tool was pushed into the boot. This tool is just a short length of .160" o.d. x .130" (reamed) i.d. brass tubing pressed into a convenient holder. This tool is a very snug fit going into the boot, but it will allow the plug wire to easily slide into the boot where it can be grabbed with pliers, or preferably, a hemostat and pulled through the boot. The insulation was was then stripped back about 1/8" and the wire was soldered into the hole that was drilled earlier. Care has to be taken to not allow excess solder to flow down inside the contact where it can block the insertion of the spark plug. Any wire left protruding outside the contact after soldering was clipped off and the sharp stub was rounded with a file.
The plug wire passing through the wire insertion tool was then rotated to orient the attached contact so the slot runs parallel with the insertion tool. The contact insertion tool was then used to push the contact down into the boot. This critical dimensions of this tool include a .218" dia. x .250" long nose followed by a .250" dia. x .250 long shoulder. The tool was inserted into the contact and then, with a coordinated effort between both hands, the contact was pushed into the boot while the excess plug wire was continually pulled out through the wire insertion tool. When the contact was fully inserted, the wire insertion tool was removed. The boot was then slid over a test plug several times to check the mechanical fit, and an ohmmeter was used to verify continuous electrical continuity while the boot was twisted on the plug. If the hole in the boot was enlarged correctly there will be enough spring force created by the surrounding rubber to hold the contact snugly closed around the plug electrode.
I found it interesting that all my NGK plugs showed zero resistance as expected while the Rcxel plugs I have typically showed tens to hundreds of ohms of resistance. I got the exact same non-zero results measuring directly across them. I have no idea where the resistance is coming from. The CM-6 is not a resistor plug and the RcXels I have are not labeled as being iridium plugs.
If there is any problem with the contact, it can be removed without damaging the boot by snipping the plug wire outside the boot and grabbing the contact with a pair of long nose pliers. Using a twisting motion the contact can be collapsed and pulled out of the boot so a new one to be installed. - Terry

741.jpg


742.jpg


743.jpg


744.jpg


745.jpg


746.jpg


747.jpg


748.jpg


749.jpg


750.jpg


751.jpg


752.jpg


753.jpg
 
Heloo Terry!
Congratulations for you building, its is a stat of the art craftsman, amasing what you have done. Really impressing. I wish to have that skills.

Please could you explain how you have cut the gaskets? I`m guessing it was cut on CNC mill. Right?
How you did that and what kind of tool/cutter have you used?

Thank you.


Edi
 
Real nice Terry. Looks like you have the boots all done up & provisions for reproducible spares based on accessible commercial parts. Thought Id throw out some links just in case.

Paz shows his trials & tribulations with silicon. I'm not exactly sure why the issues but Ive heard can be fussy with any combination of incorrect release agent, mix ratios, curing conditions, shelf life...
http://homepage2.nifty.com/modelicengine/k090102.htm

Ive not messed with silicones as much as equivalent durometer urethanes for composites work & they seem quite easy & forgiving for custom parts like this. But maybe there are some electric/insulation or gasoline/breakdown reasons why they wouldn't be suitable as rubber. I use this supplier (Cdn) but all the goodies come from USA. There is a whack of instructional stuff on youtube particularly SmoothOn & Aluminite. Maybe not for this particular application but something to keep in mind for similar custom items down the road.
http://www.sculpturesupply.com/index.php
 
Heloo Terry!
Congratulations for you building, its is a stat of the art craftsman, amasing what you have done. Really impressing. I wish to have that skills.

Please could you explain how you have cut the gaskets? I`m guessing it was cut on CNC mill. Right?
How you did that and what kind of tool/cutter have you used?

Thank you.


Edi

Edi,
I cut the gaskets on my Tormach CNC mill. I used a 60degree carbide vinyl knife that Tormach sells as an accessory. - Terry
 
Real nice Terry. Looks like you have the boots all done up & provisions for reproducible spares based on accessible commercial parts. Thought Id throw out some links just in case.

Paz shows his trials & tribulations with silicon. I'm not exactly sure why the issues but Ive heard can be fussy with any combination of incorrect release agent, mix ratios, curing conditions, shelf life...
http://homepage2.nifty.com/modelicengine/k090102.htm

Ive not messed with silicones as much as equivalent durometer urethanes for composites work & they seem quite easy & forgiving for custom parts like this. But maybe there are some electric/insulation or gasoline/breakdown reasons why they wouldn't be suitable as rubber. I use this supplier (Cdn) but all the goodies come from USA. There is a whack of instructional stuff on youtube particularly SmoothOn & Aluminite. Maybe not for this particular application but something to keep in mind for similar custom items down the road.
http://www.sculpturesupply.com/index.php

Peter,
Here is another link that I studied before deciding to modify the vacuum fittings:
http://www.homemodelenginemachinist.com/showthread.php?t=14539&highlight=Ignition+boot+molding
I was considering moulding my own, but I wasn't sure anyone had come up with a solution for getting good black dielectric parts. - Terry
 
Peter,
Here is another link.... considering moulding my own, but I wasn't sure anyone had come up with a solution for getting good black dielectric parts.

Ah, that was the other link I was looking for, sorry. I notice in the SmoothOn line they make a color tint specifically for urethanes, but specify not for use for silicone
http://www.smooth-on.com/Urethane-Plastic-a/c5_1119_1213/index.html?catdepth=1
but looks like they do offer an equivalent for silicone
http://www.smooth-on.com/Silicone-Rubber-an/c2_1128_1190/index.html
Presuming urethanes could meet the electric insulation + gasoline/oil criteria, they offer a wide range of durometers, cure times, kit sizes, established releasing properties & bit less expensive. I'm not sure if these referenced builders chose silicone from the outset, completed the parts, mission accomplished & didn't look back. Or maybe they had some negative results with urethanes?
Notwithstanding selection, I'd be inclined to try lower viscosity (pourable) varieties, although the putty demonstrates more than one way to skin the cat. I base this on the detail level the sculpture crowd is molding with reverse curves, sometimes thin delicate features etc.
If you go explore down this path, I'd highly recommend a vac pot. There are inexpensive turn-key solutions. Made a huge difference to my results. Explanation + vendor example below. Sorry for the 'squishy-stuff' side tour, back to metal talk!
http://www.bestvaluevacs.com/index.html
http://youtu.be/GKddrZI4qAo
 
Last edited:
Ah, that was the other link I was looking for, sorry. I notice in the SmoothOn line they make a color tint specifically for urethanes, but specify not for use for silicone
http://www.smooth-on.com/Urethane-Plastic-a/c5_1119_1213/index.html?catdepth=1
but looks like they do offer an equivalent for silicone
http://www.smooth-on.com/Silicone-Rubber-an/c2_1128_1190/index.html
Presuming urethanes could meet the electric insulation + gasoline/oil criteria, they offer a wide range of durometers, cure times, kit sizes, established releasing properties & bit less expensive. I'm not sure if these referenced builders chose silicone from the outset, completed the parts, mission accomplished & didn't look back. Or maybe they had some negative results with urethanes?
Notwithstanding selection, I'd be inclined to try lower viscosity (pourable) varieties, although the putty demonstrates more than one way to skin the cat. I base this on the detail level the sculpture crowd is molding with reverse curves, sometimes thin delicate features etc.
If you go explore down this path, I'd highly recommend a vac pot. There are inexpensive turn-key solutions. Made a huge difference to my results. Explanation + vendor example below. Sorry for the 'squishy-stuff' side tour, back to metal talk!
http://www.bestvaluevacs.com/index.html
http://youtu.be/GKddrZI4qAo

Peter,
The urethanes might be an even better choice. I don't have any experience with them, but I have run tests on silicone rubber and it is not compatible with gasoline. Some silicone fuel line samples I tested dissolved in gasoline within a day. It seems to be OK with alcohol and oil though. Fortunately plug wires don't come into extended direct contact with fuel. - Terry
 
I took a couple weeks off from my build in order to drive up and visit my newborn grandson some 1200 miles away in Ohio. Our visit happened to overlap the Zanesville show where I had the honor of meeting and chatting with several builders and seeing a lot of really nice models. The models of Steve H. and George B. look even better close up and in person. A third highlight of our trip was the Wright Patterson AFB Museum in Dayton which I highly recommend to anyone who finds themselves in the area. There were actual full-size versions, some of them cross-sectioned, of nearly all the popular aero engine models that I've come across.
Getting back to this build, though...
While I was searching the auto parts stores before our trip, I also came across a Dorman 47403 straight 1/8" x 7/32" vacuum fitting that fits nicely on the 1/4" diameter high voltage towers on my distributors. For the HV connections to the towers, the previously pressed-in bronze electrodes in the cap towers had been center drilled with a 1/4" deep .078" diameter hole. These holes provide a snug fit to some gold plated beryllium copper connector contacts that I found in one of my junk boxes. I don't know the manufacturer or part number, but similar contacts are commonly available from electronic parts distributors.
These contacts are soldered on the distributor ends of the plug wires after trimming and forming their strain relief crimp barbs to into a 'tunnel' for the tinned wire. The soldered joint was then covered with two short pieces of shrink tubing to strengthen this weak area of the contact and to build up its diameter for a tight fit in the 1/8" end of the boot. The boot was shortened as shown in one of the photos before pushing it over the contact. The contact is left protruding about 1/32" beyond the 7/32" end of the boot before a piece of shrink tubing is shrunk down over the rear of the boot to seal and secure it to the plug wire. I included heat shrunk numbered labels on both ends of each wire because of the large number of wires in the completed harness.
I made up a set of wire looms to help organize the wires and to keep them out of the paths of the hot exhaust gasses which is where they tended to settle on their own. The looms also keep the wires away from the engine and effectively raise, even higher, the 20 kV breakdown voltage of the plug wire with respect to ground. Old school wax'd lacing cord was used to tidy up and complete the plug wiring harness.
It turned out to be very difficult to grip the plug boots and pull them off the plugs while in their recessed cavities in the heads, and so I made a tool to pry them off the plugs. While I was at it, I also made a puller to help remove the boots from the distributor cap.
The spark plug harness concludes the assembly of the engine. I would have added the carburetor, but even though I thought I had designed the rotisserie to clear the whole carb assembly, I ended up with an interference between it and the fuel bowl. So, the carb assembly can't be installed until the engine is moved to the display/running stand. I added a temporary cooling club to help when positioning the crankshaft, but it will be replaced later for running by a three blade prop. With the compression that showed up when the spark plugs were installed, the engine now requires a minimum of 150 inch-lbs to be 'slowly' turned over by hand. My 1/2" diameter crankshaft is starting to feel small. The keyless chucks on both of my 18V battery drills can't grip my starter adapter tightly enough to spin the engine now. I Loctited a sleeve with milled flats on my starter adapter to help the drill chucks get a grip, but I haven't yet tried it out. I can feel thumb suction on the 1/2" carb adapter port when the prop is manually rotated, and that seems like a good sign since the carb I'm planning to use will have only a 3/8" throat.
To reduce boredom during my wife's driving portion of our trip, I tallied up a total of 1057 shop-made parts and 1069 commercial fastener components that make up the engine as it currently sits. The T-18 engine file folder on my computer currently contains just over 900 CAD/CAM files created during the past fifteen months. The 'scale' of what I've done, though, didn't really hit home until I weighed the engine while on the rotisserie and found out it weighs 42 pounds.
Before attempting to start it, the next several weeks will be used to design and build the various support components including a firewall, throttle/advance linkages, fuel /oil tanks, fuel pump, tach, and electrical control panel. - Terry

754.jpg


755.jpg


756.jpg


757.jpg


758.jpg


759.jpg


760.jpg


761.jpg


762.jpg


763.jpg


764.jpg


765.jpg


766.jpg


767.jpg


768.jpg
 
My battery-powered drills now seem to be able to grip the flatted sleeve I made for my starter adapter; but neither of them, not even the 18V Ryobi, has enough torque to spin the engine over. My big Bosch electric drill with its key chuck has no problem with it, but it requires both hands to hold it.
Hopefully, the engine will start easily by rotating the prop 45 degrees or so by hand because there won't be any manual prop 'slapping' on this engine. But if it's a hard starter, I'm afraid I'll be tempted to pull out the big Bosch drill and that now raises a big concern about my starter adapter design.
My crankshaft is 1/2" in diameter including the front portion that's keyed to the prop sleeve. However, the end of the crankshaft that's LH threaded for the spinner is only 3/8" diameter. According to the handbooks the maximum recommended bolt torque for this thread is on the order of 17 ft-lbs. I measured a thread yield of 30 ft-lbs on my own test rod which did not include the crankshaft's 1/8" diameter through-hole for crankcase ventilation. Approximately 14 ft-lbs torque is currently required to slowly rotate the crankshaft, but at cranking speeds the torque is likely higher because the compression pressure has less time to leak away. Therefore, my starter adapter which grips the hex on the rear of the spinner is dangerously close to twisting off the threaded portion of the crankshaft. The adapter needs to be re-designed to act upon the prop and not the spinner.
I should have anticipated this when I decided to add the additional nine cylinders. A single cylinder four stroke engine spends roughly 180 degrees of its 720 degree cycle in compression. With 18 cylinders, I have a cylinder firing every 40 degrees which means that at any given time I have 4-5 cylinders in some state of compression. All those individual cylinder compression pressures instantaneously add up to produce the resistance felt when rotating the prop. As the engine passes through its firing order this continual summation produces a seemingly continuous resistance instead of the familiar and distinct compression "bumps" felt on engines with fewer cylinders. So, it looks like a new starter adapter has to be added to the list of things left to do. The really frustrating part of all this is that there was absolutely no reason to have reduced the diameter of the threaded portion of the crankshaft. Fortunately though, the problem was caught before some really serious damage occurred. - Terry

769.jpg
 
Hope your drill solution works so it doesn't require 'the next $tep'. Maybe there's a surplus type 12v high ratio GD type solution? I'm not familiar with torque delivery vs rpm curve on typical drills, but suspect the max rating is at full rpm & it cant get there trying to overcome initial resistance.

Only speaking from maybe non-comparable RC experience, but I see these pulley adaptions in the field. The (+/- same wattage) direct drive motors just stall & that's with the dangerous pre-windup-and-spinner-bump Keystone Cops routine. But the reducers turn them over. Seems the crank doesn't have rotate particularly fast before ignition kicks in, getting 'over the hump' is the challenge.

2014-10-29_21-01-57.jpg
 
Hi Terry,
Looking at the end of your crankshaft I see the .500 dia. area with a key slot cut into it. The front portion as you stated is reduced to .375 dia. (with threads).
When the prop is mounted to the shaft is there a shoulder in the prop hub which sits against the step on the crankshaft so that the hub can be tightened to the shaft without putting load against the front engine bearing? If this is the case here's and idea for you. On my little radial I made a hardened threaded sleeve which has a through hole (radial) for tightening. The O.D. of the sleeve is made so that my starter with a one-way bearing rides on the sleeve. In your case the sleeve could have a flange that would press against the prop hub thereby imparting some of the torque against the hub and not all taken up by the threaded area. I hope this makes sense. Here is a video I made of my engine which shows what I'm talking about better than explaining it.
gbritnell
http://youtu.be/V8SDqPE4kZA?list=UUPvNzXJm9KOlaQwjAmYW9Xw
 
George,
Thanks for your reply. You're right about your understanding of how my prop hub mounts, but I'm not 100% sure I understand your suggestion of how to take the torque starting load off the small diameter threaded portion of the crankshaft. But let me define a few parts first to make sure we're both talking about the same things. I have three parts that go on to the end of the crankshaft. They are an inner prop hub that is keyed to the crank and an outer prop hub that bears against the offset on the crankshaft. The prop is sandwiched between these hubs and secured with six 10-32 bolts. The shoulder on the crank against which the outer hub bears insures there is no actual load on the front bearing. The third part is a spinner that threads onto the threaded portion of the crankshaft and bears against the outer prop hub. There are some flats on the rear of this spinner that my starter engages when it applies the starting torque to the weakest part of the crank.
So to take the load off the threaded section of the crank during starting I would need to make a new outer hub that included its own larger diameter threaded section onto which the spinner (re-tapped for the new larger diameter thread) would thread. Is this your suggestion or am I misunderstanding? - Terry
 
While thinking more about the best way to solve my starting problem, I decided to continue forward progress by working on the fuel pump since it's relatively straight forward and pretty much identical to the one I made for my H-9. I plan to use a recirculating fuel system that pumps fuel into the carburetor bowl from the fuel tank. A constant fuel level is maintained with a standpipe in the bowl by continually returning excess fuel back to the fuel tank through a return line.
Using this system has three important advantages on a large size model engine. First, it allows complete freedom in the placement of a large fuel tank. Second, it presents a constant fuel level to the carburetor for a consistent fuel draw. And third, it provides a safe and convenient way for removing left-over fuel from the tank when preparing the engine for storage. The disadvantage for a small model, though, is that it is a relatively large component that many will probably want to hide or disguise.
The key components come from a thirteen dollar RC fuel tank filler commonly available in hobby stores. I chose this particular unit after examining several others because it will run on less than 6Vdc, and my own personal experience shows the pump, itself, is compatible with gasoline. I removed the key internals, consisting of a dc motor and mechanically linked pump, in order to repackage them. I machined an aluminum housing with divider walls between the pump and motor while maintaining their mechanical alignment. The housing is well vented and contains a drain, but one should be aware there is still some hazard due to internal sparking at the motor's brushes. Even when powering the unit from a 6V battery, it's necessary to add a 50 ohm series rheostat (to be physically located elsewhere) and a .022" fuel line restrictor to maintain fine control over the flow to the carb bowl. - Terry

770.jpg


771.jpg


772.jpg


773.jpg


774.jpg


775.jpg


776.jpg
 
I thought they made a distinction between gasoline & methanol versions, but you mentioned it worked for you. Just mentioning FWIW in case it was packaging fine print.

http://www3.towerhobbies.com/cgi-bin/WTI0001P?I=LXVZ40&P=8
Current draw: .8A loaded, .2A with no fuel passing through Pump can move 1oz of fuel every 4 seconds on 12V supply Pump is for GLOW FUEL ONLY (NO gasoline, diesel or smoke oil)
 
Peter,
I'm familiar with their spec, but I suspect they don't want their pump used with gasoline with its high volatility near that dc motor for liability reasons. I tested the materials compatibility of their pump with gasoline pretty thoroughly and saw no issue. All the filler pumps I came across that were spec'd for gasoline were manual crank varieties. I'm definitely not recommending their product for use with gasoline to others as the manufacturer understands their own product much better than I. - Terry
 
The fuel tank was the next item on my list. For my H-9, I used a 10 oz polyethylene tank purchased from a local hobby shop. I appreciated the ability to be able to easily monitor the fuel level through the translucent tank, but I was never entirely happy with its appearance. I purchased an identical tank for this engine several months ago when I happened to be in the same shop. A full tank of fuel lasts only 4 minutes or so in my nine cylinder engine, and with this engine's 2X displacement the same amount fuel will last only about half as long. So, I don't want a smaller tank.
I machined a pair of mounting brackets to grip and secure the tank to the back of the engine's firewall similar to what I did for my H-9. After staring at the completed assembly for awhile, I decided to add a pair of metal end caps to try to improve the tank's appearance. I turned the end caps in aluminum and then modified the mounting brackets with a shallow recess around the tank openings so that when the end caps are inserted they are held in place with the bracket pinch bolts. This mounting arrangement has the additional advantage of removing the bracket clamping forces from the fragile plastic tank which now floats safely within the end caps. The appearance was a bit improved, and I still get to have my view of the fuel level in the translucent portion of the tank between the brackets. - Terry

779.jpg


780.jpg


781.jpg


782.jpg


783.jpg


784.jpg


785.jpg


786.jpg
 
Hi Terry
The Fuel tank looks just as classy as the rest ! Nice job.

I saw in your band saw photo that you used a filler at the back of the vise jaw for the short piece you were cutting. Here is a quick and easy mod that will save you trying to find the same thickness filler. Just drill and tap a 3/8 hole in the back of the vise jaw and put a piece of all thread in it. It works great and saves a bunch of time.

Scott

P1030140.JPG
 
Back
Top