Richard's Latest Project

This forum is dedicated to those Hobbyists Interested in CNC machining & 3D Printing in their home shops. (Digital Read Outs are also topical, as is CAD/CAM as it relates to CNC)

Moderator: Harold_V

Post Reply
Carrdo
Posts: 1444
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Richard's Latest Project

Post by Carrdo »

Was over at Richard's shop this week and took some photos of his latest project after he made some modifications to stiffen his shop constructed Super Sherline CNC mill.

Here he is making a 1-1/2 " scale ( 7-1/4" gauge) side frames in 6061 aluminum for a riding or freight car for one of his locomotives using an existing 1" scale casting as his prototype.

The first photo shows his overall setup to do the milling.

The second photo is a closer look at the work. Look at the length of the end mill being used (a 1/4" dia. cutter having a LOC of 1-3/4") as the mill has to machine down to a depth greater than 1-1/2" in places.

One thing which I do not like about the Sherline milling head is its use of an extension holder to hold the end mill when one is using this diameter cutter. I always prefer to have my end mills held in a collet up within the spindle to minimize cutter overhang. I do this even with my 1700lb. manual mill and never extend the cutter any further than absolutely needed.

I actually caused the problem with Richard's CNC mill previously when he wrote a program to do some compound curve milling for me in 01 flat stock. It was discovered that the head of the machine was just not stiff enough when executing the program without the cutter chattering and vibrating even with a minute depth of cut cut and feed. Several end mills were destroyed as a result.

Richard made the modifications to stiffen the head assembly considerably and also discovered in the process that one of the Thompson shaft bearings was loose. Fortunately, all of these bearings have a take up mechanism so this problem was easily corrected.

Even with the modifications now made, when this long cutter enters a corner, the end mill doesn't really like it (you can hear it squeal and start to vibrate) but since the cutter resides there only momentarily, one can get away with it (I am certain Richard will have more to say on this).

The third and fourth not very good photos photos are of the 1" scale prototype casting.

Photo 5 is a 3D rendering of the part created by a program he uses called Synergy. The transition from 2D to 3D CAD is quite a steep learning curve initially.
Attachments
Photo 1.jpg
Photo 2.jpg
Photo 3.jpg
Photo 4.jpg
Photo 5.jpg
Last edited by Carrdo on Fri Aug 28, 2015 3:24 pm, edited 1 time in total.
Carrdo
Posts: 1444
Joined: Thu Mar 06, 2003 2:20 pm
Location: Toronto, Canada

Re: Richard's Latest Project

Post by Carrdo »

Photo 6 is a poor photo of the program Richard wrote to execute the milling.

Photo 7 is also a poor photo of the paper print out from the CAD program used. I believe it is a 2D drafting program.

Photos 8,9 and10 are details of the modifications made to stiffen the head assembly.

I am certain that Richard would be happy to answer any questions you may have as I just took the pictures.

Richard is just so far ahead of me on anything related to CAD/CAM and CNC machining.
Attachments
Photo 6.jpg
Photo 7.jpg
Photo 8.jpg
Photo 9.jpg
Photo 10.jpg
RET
Posts: 960
Joined: Wed Jun 07, 2006 8:36 am
Location: Toronto, Canada

Re: Richard's Latest Project

Post by RET »

Hi,

Don started this; I wasn't going to say anything, but it might give you a better idea if I added a bit more information.

To give you an idea of the size of what you are looking at, the piece of aluminum bar stock (6061-T6) is 2" thick by 4" wide by 13" long. Because of the length of the end mill, I'm only taking .020" per pass so its a slow process. I'm using 1.35" per minute for a feed rate and .1" per minute as a sink rate to the cutting level in each of the "pockets." With a shorter end mill, heavier cuts and faster feeds would be easy but I'm actually surprised that the machine works as well as it does with that length of cutter. I started this project after finishing the changes to get an idea just how much better the machine might be. Because of the long endmill length, I still don't know how much the machine will take, but I do know its a lot better.

Before I stiffened up the machine, it would never have performed this well. The stiffening arms are made from 2" square mechanical tubing in various wall thicknesses. The heaviest is 1/4" wall for the bottom tubes so I could tap them for 5/16" cap screws. The upright sections are 1/8" wall and the plates that bolt onto the overarm are 3/8" thick. It took a while to fit each arm to the machine, welding each section as I went and to make the square plates and drill & tap the 5/16" holes in the overarm (both the column and overarm are made from 2" by 5" steel bar stock). Of course, each weld required removing the stiffening arm parts from the machine, taking them outside to weld, then reassembling the arm to the machine so I could fit the next piece. Fun & games, but now its done and the result is certainly worth all the hard work.

The mill has 16" travel on the "X" axis, 9" travel on the "Y" axis and 8" of travel on the "Z" axis, so you can see what I'm doing is comfortably within the limits of the machine. Those travels are almost exactly twice what the Sherline mill is capable of. The Sherline software can also handle an "A" rotary axis for which I built a rotary headstock. That headstock also works quite well.

The ball bushings on each axis are sealed (with grease fittings) and run on 1" dia. Thomson shafts which are on support rails (you need solid support with no give of any kind if you are going to be successful in building your own machine). You can see this construction easily in the last picture Don took. I didn't realize that the ball bushings were shipped with some play until I was most of the way through adding the arm braces. Once I tightened the ball bushings up, all the remaining play disappeared.

The ball screws are 5/8" dia., 5 turns per inch (machine tool quality) and have a 4 to 1 reduction with timing belts to the Sherline stepping motors (same ones Sherline uses). This way, the computer doesn't know the difference between this mill and the Sherline one.

As Don mentioned, I use a 3D CAD program called "Synergy" to do the drawing. That program has Solid modelling capability as well as parametric solids. It also will generate "G" code for the little mill, although for simple projects like this one I write the code by hand using data from a 2D CAD program (those are the drawings in one of Don's pictures).

Perhaps that will be enough for the moment.

Richard Trounce.
Last edited by RET on Sun Dec 27, 2015 10:19 pm, edited 1 time in total.
RET
Posts: 960
Joined: Wed Jun 07, 2006 8:36 am
Location: Toronto, Canada

Re: Richard's Latest Project

Post by RET »

Hi,

This is turning out to be a bigger project than I thought.

First off, the mill itself is much improved. Using the long 1/4" dia. cutters definitely limits the depth of cut and feed rate that can be used. With the 1 1/4" cutter, I can take .030" at a feed rate of 1.5" per minute; with the 1 3/4" cutter, that drops to .020" depth of cut at the same feed. With both cutters, the limit is the deflection of the cutter, not the machine.

As you can see in one of the pictures, I switched to a 3/8" end mill and with that the machine is happy with .050" depth of cut at 2" per minute which I think is quite respectable for a home built tabletop machine. With the same cutter, the machine will actually handle .060" depth of cut at the 2" feed rate, but limiting it to .050" is easier on the machine.

One strange thing I found out by accident. On two separate occasions, using the 1/4" dia. 1 1/4" cutter, I was vacuuming the chips away and got too close to the cutter. It pulled the vacuum nozzle in and jammed the cutter in the work so neither the cutter nor the steppers could move. My stupidity, right?

Anyway, it took me several seconds to realize what happened and click on "pause" to stop the feed. Then I clicked on "Abort" to kill the program so I could switch to Manual. After freeing the cutter, I used the manual incremental feed to zero out the "Z" axis then the "X" and "Y" axes. I expected that 0,0,0 and "Home" would now be very different because the steppers had been jammed and couldn't turn, but the program hadn't lost a step! Home and 0,0,0 were still identical. I can tell they are by looking at the dials on the hand wheels attached to the steppers.

Somehow the software "knows" when the steppers don't move. The Sherline steppers have six wires connecting them to the stepper control box which is actually powered from and resides in the computer case. This control box connects to the computer through the parallel port. Not that I'm complaining one bit (this is great), but perhaps one of you can tell me why this is so.

As you can also see from the second picture, all the extra bits are now off the part and it is close to its final shape. In the picture, I'm using a 1/4" dia, 1 3/4" long ball nosed cutter to get a radius where it is needed.

Richard Trounce.
Attachments
Using 3/8" cutter to bring the top to its finished height and shape
Using 3/8" cutter to bring the top to its finished height and shape
Shows the part with most of the excess material removed
Shows the part with most of the excess material removed
RET
Posts: 960
Joined: Wed Jun 07, 2006 8:36 am
Location: Toronto, Canada

Re: Richard's Latest Project

Post by RET »

Hi,

The part is now finished and the pictures show what it looks like.

I used the Bridgeport to cut the "axleboxes" into the towers at each end. If you look at the 4 3/4" size you can see the resemblance since that is what i used to make the CAD drawing for the larger part. I just scaled it up and then wrote the code for the 3 different program sections.

Now all I have to do is make 3 more just like the first one. For each one, I'll use the Bridgeport to do the initial machining that removes as much extraneous material as possible before I transfer the part to the little mill for the CNC portion.

When the parts are finished, I'll have to draw and make two bolsters and frame crosspieces to complete the truck frame. I plan on using 3/8" dia. ground shafting for the non-rotating axles and put needle bearings in the hollow wheel axles that the wheels are pressed onto. Those hollow axles will each have a grease fitting in the centre for lubrication. This will make a car that will roll at a touch.

Because of the width limitations, I had to crop the picture to get it to load. I cropped the right end on the first picture and the left end on the second so you could get an idea of what the full picture looks like.

Richard Trounce.
Attachments
Initial view showing the 3 different sizes.
Initial view showing the 3 different sizes.
View showing the completed part
View showing the completed part
Slant view of the same part.
Slant view of the same part.
RET
Posts: 960
Joined: Wed Jun 07, 2006 8:36 am
Location: Toronto, Canada

Re: Richard's Latest Project

Post by RET »

Hi,

I'm now working on the second piece. This time I removed as much material as possible with the Bridgeport and bandsaw so there are no "extra" pieces to hacksaw off part way through. I also put in the three 1/4" dia. locating holes (the center one is "Home" for all 3 programs but it disappears when the center upsidedown "T" is cut through). The 3 separate programs perform 3 different operations. The first program uses the 3/8" dia. 2 flute cutter to bring the "top" (outside face) to its final height and shape. The second program uses 1/4" dia. 4 flute cutters (1 1/4" length of cut and 1 3/4" length of cut) to do the outside profile and to cut some of the triangular and "T" shaped holes in the piece. Most of the holes go right through. The third program also uses the same 1/4"dia. cutters to finish the outside of the axleboxes and complete the holes in the part.

While it is still a very slow process, I have been fine tuning each program to improve its performance. In program #1, the 3/8" dia. cutter now takes .050" depth of cut at 2" per min. In program #2, the 1 1/4" length of cut end mill now takes .040" depth of cut at 5 1/2" per minute while the 1 3/4" length of cut cutter takes .020" depth of cut, also at 5 1/2" per minute. I think it might tolerate a .030" depth of cut, but I haven't tried it yet. I don't want to push things too much.

None of the cutters like to "sink" (they squawk a lot) but they seem to like the faster speeds much better.

For all the programs I've increased the spindle speed by a third which was necessary to get the higher feed rates. I think that the 4 flute cutters allow for the higher feed rates in the second and third programs. With 4 flutes the next flute starts to cut before the previous one has finished resulting in no vibration. The cut is very smooth except at the corners where there is a sharp change in direction.

Its still "fun & games" but I'm learning a lot. As I said before, without the extra bracing to make the machine much stiffer, there is no way that I could have even dreamed of the speeds and feeds I'm able to do now.

Hope this inspires others.

Richard Trounce.
Post Reply