Crossley Otto Langen

[Jasonb]

The latest version adds shadow and highlights to the parts, not to the same extent that Fusion does though definately not a full render but more than it did upto a couple of months ago. Line weights are also different as well as several other tweaks and fixes.

You can switch these off which is like running the older version in the same way the latest update to Fusion 360 can be reverted to the old look.

I design in much the same way as you doing groups of parts into sub assemblies and then the main assembly going back and forth tweaking fits and sizes, this is how things look now.

[Brian Rupnow]

Nice 3D work guys.I worked the first 30 years of my career on a drafting table. I thought all this computer stuff was just science fiction. Now after 25 years on CAD I wonder how i could have been so naive'.---Brian

[Craig DeShong]

I have heard that the folks at the Smithsonian Institution have an Otto Langen engine that was destroyed through ignorance when they incorrectly tried to run it years ago.

I was at a show this weekend and though I don’t recall the series of events that lead up to it; I managed to overcharge my model and though it didn’t blow the piston out through the top of the engine; when the piston hit the flat table above the cylinder so violently the force sheared off a clevis pin holder that attaches the rack to the piston and I was out of business for the remainder of the day.

Today I’ve been making repairs and the model is running again.

While I had it apart I used the opportunity to do a little “what – if” in answering some questions I need answered to move ahead with the design of this Crossley engine.   The bore of my Otto Langen is a tad over 2 ¼ inches and the scale bore of the Crossley would be right at 2 inches so I figured my current Otto Langen ought to fire with the same characteristics as this Crossley model I plan to build.

My existing model lifts the piston around 7/8th inch and the resulting charge provides a stroke of nearly eight inches in height.  This length stroke is way to long for the Crossley if I want to maintain scale so I fiddled around with adjusting the lift height of the piston and measuring the resulting stroke.  When I had reduced the piston lift height to around one half inch the resulting stroke had been reduced to around four inches- much more in line with a stroke length for this Crossley model.

So I’ll be setting the piston lift to around ½ inch on the Crossley and this “should” allow the scale cylinder height to accommodate the resulting movement of the piston in the bore.  One more issue resolved in the design of this model.

[Craig DeShong]

The design of this model has been moving along.  Most of what is left to do is cosmetic in nature, so that the model, though made from bar stock, preserves the “flavor” of the original.

Before I get too involved in the time consuming nature of this later design phase I thought I would concentrate on some of the parts of this model which might need design refinement.  Since this design refinement might result in an alter of frame measurements I’m thinking it prudent to address these issues before I get into the final frame design and construction.

One of these issues I discussed earlier up thread was the design and construction of the bevel gears that will drive the fly-ball governor. 

In this photo I’m forming the gear blank, taking a facing cut to reduce the piece to the correct thickness.


The next task was to cut the 45 degree shoulder for the face of the gear teeth.  I’m doing this with the compound slide, having set its angle to the crosshead guide at 45 degrees.  I’m also running the lathe “backwards”.


Moving on to the mill.  Here I have my indexing head mounted to the table at 45 degrees in order to cut across the face of the gear blank and form the individual gear teeth.


I failed to provide a photo of the gears I produced so I give it below.  The larger (drive) gear is the final gear I’ll use on the model and it has been cut from steel.  The smaller gear is still aluminum because I need to fabricate it on the end of the governor itself.  I will make it again, but in steel, when I build the governor.



Having worked out potential problems with the bevel gears, I thought I’d look into fabricating the sprag clutch.  I guess I could purchase one of these but I made the one on my first Otto Langen and it works well so I’m going to attempt to make this one also.

Here  is an image of the clutch internals.  You can see that it is comprised of an outer shell with a series of ramps cut into the inner surface.  Inside this shell is a round rotor, attached to the drive shaft.  There are cylindrical rollers placed on the ramps between the outer shell and the internal rotor.

When the central shaft is rotated clockwise the rollers are rotated to the thicker back of the ramps and the central shaft can rotate freely.  When the central shaft is rotated counter clockwise the rollers are rotated to the thinner front of the ramps where they jamb in the ramp.  This causes the outer shell to be carried around with the central shaft.


This is my drawing of the outer shell with the ramps.  This is one of many drawings I’ve made to assist me in the construction of the ramps.  With this drawing I can locate and drill the four outer mount holes as well as some “starter” holes for the individual ramps.



Finally, here is the setup on the mill.  I’ve started work on the outer ring blank and have drilled all the holes indicated in the above drawing.  Next I’ll start fabricating the ramps.

[b.lindsey]

Making good headway on the design and fabrication both. Still following as closely as I can ...good stuff!!

Bill

[crueby]

Following along here too!  I love the diagram of the clutch, very interesting. Are the rollers balls or cylinders?

[Jasonb]

Craig, did you cut both those gears with the 45degree setting? I would have thought you need something more like 20/70 as anything other than 1:1 are cut at different angles for pinion and wheel.

Have a look at Simon's lane and Bodley  thread

http://www.modelenginemaker.com/index.php/topic,7093.msg205179.html#msg205179

[Craig DeShong]

Bill and Chris; thanks for your comments. 

Jason, thanks for the input; getting helpful input is one of the reasons I post here.  According to a calculator I found, the angles on the gear cones should be 71.5 degrees and 18.5 degrees so you were pretty much spot on.  Though the gears I made seem to run well, I'm going to look into this.  Thanks.

The rollers are cylinders.  I'll make the clutch itself out of steel and the rollers out of brass as I did with my first Otto Langen.  This allows the rollers, which can be easily replaced, to take the wear as opposed to the machined surfaces.  Hardening the clutch would probably be a better solution but I haven't seen the necessity.

I'm appending a PDF of the clutch that you should be able manipulate.  I don't know if the board will allow this type of PDF for fear of transmitting a virus and if the board doesn't I can understand why.

[Craig DeShong]

Moving forward with the design of this model…

As I stated up thread, I’m keen to hash out some of the potential problems with this build; problems that might end up changing the design.  I wouldn’t want to put a lot of effort into the actual build, only to find out that a re-make of a lot of what I’d completed was required because  of a design change.  I’m going to go ahead and make the clutch and .. Jason… have a re-look at the bevel gears before the design progresses any further.  Once I get over these two potential hurdles I’m pretty sure the design will be ok.

Starting with the sprag clutch.  The internal design of the clutch is to have ramps that “jamb” rollers in place, forcing the entire mechanism to spin in one direction and allow the rollers to clear the ramps and allow the internal rotor to spin freely in the other.  Getting the ramps machined correctly is critical to the success of this clutch.

Below you see the layout for one of the ramps.  The given measurements come out of the design tool.  The ramp is, in reality, an arc of a circle with the center offset from the center of the clutch.  Each of the six ramps gets the same x,y offset with respect to its placement on the disk circle.  Here I’m showing just one of the layouts for clarity.



To machine the ramps I’m using my rotary milling head.  This allows me to mount the disk in my milling vice and use the x,y coordinates on the table to set the offset for the center of the arc, and then mill the arc using the rotary milling head.  One thing I didn’t plan for was the need for a long reach end mill, since the depth of cut is 13/16th inch.  I ordered three and broke only one.



Below is a better view of how the milling operation was performed. 



And a photo of the completed clutch shell with the six ramps finished.




With the outside of the clutch complete I started on the internal rotor.  This piece mounts inside the above part and is fixed to the main shaft with a key.  Here I’ve sized the piece, drilled and reamed the center hole, and machined a lip where a bearing will interface. 



Here’s a photo of the completed rotor with the keyway broached.


Next I’ll be looking at the sides of the clutch housing and the bearings.

[crueby]

Great work on the clutch! 

Thats the first time I've seen a milling head like that, it must be an interesting mechanism inside it to drive the spindle at an offset. A rotary table/adjustable offset in the mill head, that would open up all sorts of operations. Is that a common accessory for larger mills?
Chris

[Craig DeShong]

Chris, I don't believe these are that common.  Several companies made "attachments" for the Bridgeports and Bridgeport clones to extend their capabilities and Volstro was one of them that made this milling head for the Bridgeport and several other mills I believe.

I have a machinist friend who has one and some years ago showed me how it worked.  When I got my Bridgeport some time back, getting one of these milling heads was on the "to do" list. 

Today, of course, they are all obsoleted by CNC machining.  You can pick them up from used equipment dealers for a reasonable price if you are patient.  There were two for sale at Cabin Fever last January, priced at about the top of the price range.   

[Craig DeShong]

Thanks for those of you following along.

Today I finished the work on the Sprag Clutch and I couldn’t be more pleased.  The clutch functions well, freewheeling in the one direction and instantly and positively grabbing in the other.  There is also no tendency to “stick” when the direction is reversed after engaging.

With this post I’m documenting the machining of the remaining component parts… Here is a photo of the sides of the clutch.  The inboard clutch side attaches to the drive gear, thus the four drilled and recessed holes in the one cover.  The other four holes present in both sides are for assembling the clutch body.



Since the outside of the clutch freewheels on the shaft the clutch needs bearings to support the clutch body (the part with the ramps) and center it over the rotor.  Here is one of the bearings.  The narrow diameter presses into the center hole in the cover while the larger end fits in the recessed side of the rotor.



This is the onboard bearing.  It is longer because the drive gear (the gear that meshes with the rack) rides on this bearing and is attached to the clutch body.  I haven’t made the drive gear as of yet.



Now a photo of the clutch under assembly with the rotor and rollers installed within the clutch body.  It’s sitting on one of the covers and the other cover is sitting alongside, waiting to be attached.



Proof is in the pudding as they say… I made a “dummy” shaft, just to test the functioning of the clutch and clamped it in my vice, keyed the shaft to the internal rotor as will be done on the real assembly, and assembled the clutch; here is the result. 
<a href="https://www.youtube.com/watch?v=tod6Rjr11wQ" target="_blank">http://www.youtube.com/watch?v=tod6Rjr11wQ</a>


Now to revisit the bevel gears and once I work them out I can think about completing the design and get into some serious building.

[Brian Rupnow]

Craig--I've always known how those clutches work, but I never seen anyone make one before. Great stuff!!!---Brian

[Craig DeShong]

Brian: Thanks for your comment.  Thanks also to those stopping by to see my progress.

Here is a photo of the bevel gears on the full size Crossley I’m attempting to model.


As I stated previously and as Jason so kindly pointed out up thread; I needed to re-visit the design/construction of the bevel gears that will drive the fly-ball governor on this model. 

Jason’s comment prompted me to do a little research and the research lead me to this URL https://daycounter.com/Calculators/Bevel-Gear-Calculator.phtml which is an on-line bevel gear calculator  I found on the net.  If you input the diametric pitch and the relative number of teeth you wish the two gears to have the calculator supplies you with all the pertinent information you need to design/cut the appropriate gears.  Put in a DP of 32 and 10 teeth on the pinion and 30 on the gear and you’ll get the numbers with which I’m working.

I had a few false starts that reduced the length of a bar of ½ diameter aluminum tubing I’m using to form the pinion gear by a few inches and added some heft to my scrap bin, but in the end I got things sorted out.

At the risk of being overly pedantic, I’m going to go through the steps to make one of these bevel gears.  No special equipment is needed (like Chuck Fellow’s helical gear cutter I used on my previous project); just a few gear cutters and a dividing head.  Ivan Law fully describes the process in the book I mention up thread. 

Forming a bevel gear required three passes for each gear tooth. After completion of the 1st pass the gear blank looks like the photo below.


You should notice that each of the gear teeth form a wedge.  The next two passes will remove additional material so that the individual gear teeth are formed correctly.  This is done by rotating the dividing hear ¼ of the travel normally done to advance to the next gear tooth position and then adjusting the height of the mill table so that the cutter enters the gear blank from the center of the gear, in the center of the previous cut, and shaves material off the bottom of the gear wedge, forming the top of a gear tooth.  In this photo I’m about to start pass two.



Here is another look from the other side of the same operation.  It’s hard to tell exactly what’s happening in these pictures (and hard to tell when milling the gears also).  Sound more than anything is an indicator.  The cutter enters to blank with relatively little sound and the sound of the cutter removing material intensifies as it cuts deeper into the wedge.



With the second pass complete, the bottom of all the gear teeth has been formed.  Now the gear blank is rotated backwards ½ the rotation you would require to advance to the adjacent gear tooth, the mill table is lowered so that the gear cutter again enters the gear blank from the center of the gear, in the center of the previous cut, and material is shaved off the top of the gear wedge, forming the completed gear tooth.  In this photo I’m about to start the 3rd and last pass.



Here is a photo of the complete gear



The pinion is formed using the same three pass technique.  Lastly I give you a short video of the two gears.  Both these gears were cut from aluminum as a “trial run” to make sure I have the correct design and process.  Now I’ll repeat the process using steel.
<a href="https://www.youtube.com/watch?v=jjA6WzDQXzE" target="_blank">http://www.youtube.com/watch?v=jjA6WzDQXzE</a>

[Brian Rupnow]

Keith--if you are trying to achieve a 2:1 ratio, a 10 tooth and 30 tooth won't do it. Maybe I missed something.---Brian