By Jupiter

[Vixen]

During the last few weeks, I have been making good progress with the two dummy magnetos for the Jupiter engine.

The exterior of the two magnetos is almost complete as you can see in the next photo. The tiny 8BA wingnuts which secure the ignition points cover, were interesting to make due to their small size and also because any means of holding them quickly disappeared. The final cuts were made with a single 8BA (2.2mm) clamping screw.




The input shaft and coupling were next items to be made, together with the 28 tooth gear blank. These items are all keyed to the central shaft




Most of this work was done using my 4th axis unit. As you can see, I have built the 4th axis unit from a redundant EMCO Compact 5 lathe. The spindle motor was replaced with a powerful stepper motor with a 3.6:1 toothed belt drive. I use the 4th axis unit mostly as a horizontal dividing head for gear cutting etc. The spindle can be positioned precisely at any angular division using the stepper motor, no need for a dividing plate. This will be very useful when I cut the 63 tooth gears in the distributor drive as 63 hole dividing plates are not exactly common.




Using an redundant lathe as the basis for the 4th axis unit has the advantage of being able to accept my existing three jaw, four jaw or collet chucks, as well as the ability to hold long shafts between centers. I also have a fixed steady available to provide additional support for very long whippy shafts






For vertical workholding, I like to hold small round items like this distributer spindle in an ER25 collet. The ER25 collet is in a Stevenson block held in a small machine vice.




The four mounting holes in each distributor spindle are coordinate drilled and tapped 8BA. The 63 tooth gear blank is then coordinate drilled in the same setting to ensure everything is concentric. The gear blank is then bolted in place so that the three kidney shaped holes can be machined, all at the same setting




Here is the distributor 63 tooth gear blank, sitting in place in the heart of the Magneto. I ordered some 0.6 MOD involute cutter from Aliexpress in China but they failed to arrive, I eventually got a full refund. I propose to make some single point cutters from hardened gauge plate following the advise from other forum members. A single point cutter should be adequate as the 63 tooth gear is made from delrin and the 28 tooth gear is made from some soft, free cutting mild steel (EN 1). Neither material should make huge demands on my home made involute cutters.




The final photo show some of the collection of parts made for the two magnetos. I often make a few extra pieces so I can select the better ones for the engine and to allow for unexpected disasters, which may be lurking just around the next corner.



Stay tuned

Mike

[Ye-Ole Steam Dude]

Hello Mike,

Those mags are too beautiful to be called dummies, how about decorations or accessories ?

Have a great day,
Thomas

[fumopuc]

Hi Mike, brilliant as ever.

[Vixen]

Hello Mike,

Those mags are too beautiful to be called dummies, how about decorations or accessories ?

Have a great day,
Thomas

Thomas, they are only dummies in the sense that the magneto coil part does not generate the sparks. The other parts of the  magnetos will still be earning their keep, acting as the working 9 cylinder distributor. I am also using a magnetic Hall sensor inside the magneto, in place of the contact breaker points, to trigger an external 12 volt spark coil. I have actually made provision in the design for a working magneto coil to be added at a later date.

Mike

[Art K]

Mike,
This all looks great. They are only dummies in regard to magnetos, I love the idea of hiding a modern CDI ignition with a hall sensor inside. It only makes it better if you can put the coil inside as well.
Art

[michelko]

What a great looking work 
Will follow this thread for sure

Michael

[steamer]

Noce work Mike!.   Gage plate cutters work fi e, but with steel, consider gashing the spaces first as it really lightens the load on the gear cutter.    That worked well for me.

Dave

[Vixen]

It's time to do some gear cutting for the Jupiter's two magnetos.

Some of you will recall from my topic "0.6 Module Involute Cutters" www.modelenginemaker.com/index.php/topic,8489.msg184641.html#msg184641  the Jupiter distributer rotor arm is driven via a pair of tiny 63T and 28T gears giving an overall reduction of 2.25:1. The only affordable source of the 0.6 Module involute cutters I could find, was Aliexpress in China. Unfortunately the cutters became lost in transit. I did eventually get a refund but that is no substitute for actual hardware.

Jason gave me good advise and the inspiration to consider making a pair of single point cutters, adequate for the small number of gears required. Zephyrin provided the necessary cutter theory and geometry. Ivan Law's excellent book "Gears and Gear Cutting" provided very practical advice and the "how to"

Following Ivan Law's two buttons method, The 63t gear cutter can be formed by two circular cuts, each of 12.93mm diameter and the 28t pinion cutter by two circular cuts of 5.74mm. It's not quite watchmaking but smaller than clockmaking.

Silver steel 'drill rod' is normally round, so I chose to make the single point cutters from flat gauge plate. Make sure you specify the gauge plate is delivered in the soft 'annealed' condition, otherwise it will be too hard to work, except with a grinder. The circular cuts were made with a 5.0mm end mill with the annealed gauge plate blank held in the machine vice at a slight angle to provide some cutter relief. I machined a single point cutter on each end of the gauge plate blank. The cutters were hardened by quenching in water from cherry red, followed by tempering, to reduce brittleness.


Here is one of the single point cutters mounted in a suitable radial tool holder.




The single point cutter was mounted on the vertical spindle of my Emco F1 mill. The gear blank was mounted in the horizontal axis of my 4th axis indexing head. Each tooth was cut in a single pass with the indexing head advanceing by one tooth after each pass. The tooth depth for 0.6 module is only 0.053" (53 thou), it was hardly necessary to gash the spaces beforehand to lighten the load on the single gear cutter. Besides I do not have a suitable small, thin slitting saw available.

I have created a generic G-code listing (LinuxCNC), which can be used for any tooth count or cutter size. Here is my code for 63T 0.6 mod. You simply enter the gear tooth count and tooth depth and other parameters in the appropriate # pidgeon holes . The program does the angular division and tool paths, by looping around until the specified number of teeth have been cut. You are all welcome to use this code if you wish.

#1=63(number of teeth)
 #2=[360/#1](tooth increment angle)
 #3=-.1(y clearance)
 #4=0.2(X start of tooth)
 #5=-.01(X end of tooth)
 #6=0(starting A axis position)
 #7=2(x and y feed rate)
 #8=0.053(depth of cut 2.25 x 0.6 mod)
 G0 X#4 Y#3
 G0 Z0
 G0 A#6
 M3 S200 M8
 o200 do
   G1 Y#8 F#7
   G1 X#5
   G0 Y#3
   G0 X#4
   #6=[#6+#2]
   G0 A#6
 o200 while[#6 lt 359.9]
 M30

Here is the setup on my Emco F1 Mill. The tapered ER25 closure nuts, made by Emco, are a distinct advantage when gear cutting these small gears. They allow you get get right in close and reduce overhang etc.










Here you see the results of about three hours of gear cutting. Four nice, deeply dished, delrin gears for the magneto, 63 teeth, 0.6 module, about 1.5 inch (40mm) diameter, waiting for the steel 28T pinions to be cut.



Stay tuned,

Mike

[Ye-Ole Steam Dude]

Hello Mike,

That is some neat stuff. The gears look great.

Have a great day,
Thomas

[kvom]

What feedrate did you use, and how did you calculate it?

[Jasonb]

Mike, when you tilted the tool blank at an angle did you compensate by cutting an eliptical shape as cutting a circle only produces the correct profile when the bit is horizontal otherwise you get too long a cutter profile.

[Vixen]

Kvom,

The feed rates are programmed in parameter #7 of my G-code program. In the example shown, I set the feed rate arbitrarily at 2.0 inches per minute. In practice, for the first gear, I throttled the feed rate back to a cautious 1 inch per minute, using the feed rate slider. I wanted the single tooth cutter to survive all four gear wheels, some 250 teeth, so decided to keep the cutter loading extremely low. The tool cutting diameter was a little over 1 inch and the spindle speed 2000 RPM, feed rate 1 inch per minute. I know I could have increased the feed rate considerable, but 15 seconds per tooth did not seem unreasonable. Time is no longer money, when you have retired.

Jason,

Quite correct, a circle becomes slightly elliptical on an inclined plane. But remember the perfect gear flank shape is actually an involute curve, so a circle is only an approximation. An ellipse is actually a slightly better approximation. However, as we discussed in the other topic, the tooth face is only a very small part of the cutting (or button) circle and you have quite a bit out of latitude with the circle diameter before it makes any measurable difference to the tooth profile, especially with a gear as small as 0.6 module. To illustrate the point, the #2 cutter from a standard set of 8 is capable of producing perfectly adequate gears over the range 55T to 134T. My single point cutter was designed and machined specifically for 63T geometry. I'm happy.

Mike

[steamer]

Looks good Mike!   I did something quite similar, using the same references, only the button worked out to a 5/32 end mill diameter, so I made the cutter using the end mill instead of making the buttons.

It worked quite well.   It's a little tougher going in steel....gashing them first with a saw worked very well in steel

Dave

[steamer]

A couple of shots of what I meant

[Vixen]

Hello Dave,

Yes, we all walk the same paths. Good looking gears.

Actually the profile of the cutter is probably a second order issue. You can get far bigger and measurable errors if you are a thou or two out when setting the cutter centre height or the outside diameter or PCD. I could probably be in trouble with those fractional inch sizes.

One of the biggest sources of error can be getting lost with the vernier wheel on a dividing head. Then you get to ask yourself "How do you want that last tooth to look? One extra wide tooth or two very thin teeth?"

I'm now looking for a small slitting saw to gash the steel wheels before the involute cutter. Or I could make and use an undersize cutter, that has the advantage of maintaining centre height with the same tool holder.

Mike