Gerrit's Eureka Relieving device build

[gerritv]

It is fortuitous that Bjorn posted his build thread recently. I am in process of building a Jan Ridders Glass Cylinder engine and it requires T2.5 timing pulleys. I am not fond of spending money (it is a hereditary thing I think) so I intend to make my own pulleys. Originally I was going to use the fly cutter technique but with the tiny tip size I am concerned that the tool will not last cutting 3 pulleys.

So, I need to make proper cutters and then relieve them. As Duplex states in his original article series in ME 1949, more teeth is easier on the machine. Not wanting to cave in and use the alternatives, I need a relieving device. I researched a lot of patents, including the Hendley cross slide version but it turns out the Eureka is more feasible with the tools I have. I found Balzer's original 1895 patent for a previous version of the Eureka but the article I am basing this off is from the Centennial ME archive.

Off to the metal pile to find the chunks of steel from which to carve this thing. I opted to make the anchor and ratchet plates first so that the mandrels can be turned for a good fit. Not having a large boring head I bolted the 2 plates together after laying out and drilling a .125 guide pin hole in the centre. This was mounted in the 4 jaw and centred on the hole. A boring boring operation later and I reached the 1.500 dimension.

Off to the bandsaw for some rough trimming and then to the horizontal milling machine for cutting the  slots and cleaning up the straight bits. The round portions were trimmed on a belt/disc grinder and filed to the lines.

[steamer]

Cool!    It's on the list.....

watching along!

[Don1966]

Great start Gerrit.....


Don

[gerritv]

Feeling confident, I started on the main mandrel. This lump has 2 eccentrics and 34 diameters, of which 2 need some accuracy. The eccentric offsets also need to be accurate otherwise the relief won't be enough (or too little) and the ratchet mechanism won't index properly.

Having only metal that is too big by a margin, it was first a tedious turning exercise. I started using a carbide insert tool so that I could take .010 cuts to get the majority of the metal off. I then offset the lump 0.168, checked that several times and finished the 1.75 and 1.500 main diameters. The 1.500 was finished to a what turned out to be .004" loose fit :-( Whoopsie 1 of more to come.
The mandrel was re-centred and the 1.000, .620 and .375 sections were turned. All fine so far except the slightly loose fit above.
Then came the time turn things around and to offset the other eccentric, which determines the amount of relief. This was turned to 1.12 to be held eventually in the 3 jaw when using the tool. I also drilled the centre hole for the small end, around which the tool will oscillate. I checked this offset 5 times, perhaps in hindsight I should have only checked 3 times! After finishing this section I did some checking and found I had offset in the wrong direction! Good thing I had made that 1.12 as I had enough to turn it correctly and down to 1". Also off course I ahd to remove the incorrect centre hole from the small end, also feasible since I had left this longer than needed.
And then it happened, the cutting tool dug in and up popped the work piece. I though all was fine but realized after some other checking that the .375 portion was bent. It now sits on my desk as a paper weight!

[gerritv]

Having learned in life that learning is a good thing, I did a deep think about how to make this mandrel manufacturable. A fascination article from WWII on making the Bofors 40m anti aircraft gun ready for mass production came to mind. I also wanted to avoid producing another 5 lbs of swarf.

The result is to make the mandrel in 3 parts: the main eccentric with 0.168 offset, the main shaft with 0.030 offset and the small 0.375 shaft. this would assist in making adjustments later as well as reducing the risk of mistakes. And even better, a lot less cutting.

This resulted in proper parts and dimensions. The 0.375 hole was reamed with a .374 machine reamer, the resulting fit on my drill rod is great. I marked the 4 quadrants on the back of the main eccentric to assist in lining up with the main shaft later on. I still need to drill and tap a few set screw holes for the eccentric and the small shaft. The eccentric will also be glued using Weicon green retaining liquid (this is an affordable alternatie to Loctite, less than half price for same specs)

I also single pointed the 1/2-40 tpi thread although later realized that I should have ground the cutter at 55° instead of 60°. I turned the spindle by hand, not used to running the spindle in reverse with tool upside down yet ( a method that avoids the threading to a shoulder risks.)

[zeeprogrammer]

I'd never seen such a thing before and had no idea what it did.
A little googling got me a video of how this tool works.

Very interesting! 

[gerritv]

Yes, a fascinating device that is difficult to figure out from the drawings, it pretty much needs to be seen in a video to comprehend the motions involved. The original patent from 1895 is even more perplexing. We have so many tools such as 3D Cad to help us visualize and design, in those days it was quiet reflection and drafting skills.

[gerritv]

A bit behind in posting!
With the indexing sleeve cut off, I can now use that to machine the nut for it. I ground a 60° internal threading bit for a small boring bar and incrementally cut the thread until the sleeve just fit. The nut blank was cut off and off to the mill to machine the flats. I prefer the look of 2 flats over 6.
Then performed the stepped internal boring per drawing. It is useful to figure out which internal face is the locating one as the others then don't need as much attention for depth. In this case the locating face is at the end of the .620 bore. I thru drilled 5/16 knowing the in the end a part of it will be 0.375. Then brought the 1.500 bore to a very close fit on the mandrel. Finally drilled out the 5/16 to just under .375 and reamed with a .374 machine reamer. This was later finished reamed using an expandable reamer for a close fit on the drill rod used for the mandrel. A popping noise rewards me when putting the sleeve onto the mandrel so very pleased with the fit.

[gerritv]

Next up is the retaining ring for the anchor plate and some studs for pawls.
Nothing special here, I used 1/4-28 and 10-32 threads instead of the BA ones called out since I have those taps and dies on hand. Unlike the 1/2-40 and 5/16-40 threads, these are just generic.

It is starting to look like the desired end result. A few more parts (pawls an.ds springs) and the 12 ratchet detents before a test. I am using Weicon 306-38 retaining compound for assembly, it is an equivalent to Loctite 638 at less than half the price so the bank account is pleased.

There will be a short intermission while I install DRO's on my milling machine, I don't trust my skills in centering the sleeve correctly for cutting the detents. Too much time in that piece already to blow it on miss-location. One paper weight per project is my goal and I already reached that early on :-)

Gerrit

[gerritv]

That was a short intermission, not even time for popcorn!
The mill now has 2 of 3 DRO's mounted. Not totally sure how I want to do the last axis and don't need it for this project.

Wow, what a convenience! Trivial to centre the cutter over the indexing sleeve. I cut the 'teeth' in one depth, changing the spindle to slower speed to take it easy on the dovetail cutter. 3/32 depth of cut, a bit of oil and the teeth are done. And yes, they are pointing the right direction, not paper weight here. While it was set up I also did the wrench flats.
A bit of file work to break the sharp edges and this part is done.
That leaves the 2 pawls and springs to make and then some fitting to get things to click.

The Anchor Plate pawl was up first, a bit of work in the mill, again the DRO makes it so easy to line up holes with edges. Some band saw work and then filing resulted in an almost finished pawl.
I also cut the screw driver slot in the Anchor Plate pawl mounting screw.

[gerritv]

I used some old school lay out methods for the arcs. Its great to have lots of different sized stock on hand for marking out.

[zeeprogrammer]

One paper weight per project is my goal and I already reached that early on :-)

I'm still targeting 3. I've got a ways to go to match your goal. 

Following along. It's an interesting tool. 

[Don1966]

Looking good Gerrit and pretty close to finishing.....

Don

[gerritv]

It's alive. Still have to add a spring for the lower pawl, finish cleaning up surfaces and securing various parts (the mandrel is 3 pieces instead of original 1) but the ratchet works.

<a href="https://www.youtube.com/watch?v=Eou5lnRIGSE" target="_blank">http://www.youtube.com/watch?v=Eou5lnRIGSE</a>

Gerrit

[Steamer5]

Hi Gerrit,
That's cool to watch! I hope once you get it up &  running that you will show us it doing it thing!

Cheers Kerrin