Beretta 92FS/M9 Frame

Page 1   Page 2

beretta92fs_4.jpg (31135 bytes)

For my next project, I will be doing the Beretta 92FS frame.  For those who aren't familiar with this pistol, the above pictures will give you an idea of what it looks like.  I recently purchased a used (police trade-in) Beretta 92FS pistol to use as a reference while I make the solid model of the frame.  As of right now the only references I have for this project is the pistol I bought.  I also have the drawing below, but it doesn't give near enough detail.  I'm trying to get my hands on any reference information that will help me out while I'm working on the solid model.  So if anybody has any useful information (blueprints, sketches, etc.), please let me know.  I would really appreciate any help you can give me to help this project take its first steps.

beretta_rough_print.gif (37101 bytes)


I started to work on the solid model of this project.  So far I have the rough profile of the pistol.  Below is a pic of what I have done so far.  If you have any information you can share with me on this pistol I would really appreciate it.

beretta-1.jpg (21245 bytes)


I now have a pretty good start on the solidmodel.  I've been using the Co-ordinate Measuring Machine (CMM) at our shop to measure some of the complex hole locations and surfaces.  Below you will see how far along I am with this solidmodel.

beretta92fs_2.jpg (63190 bytes)


I got some more work done on the solidmodel.  About the only thing I have left is the back of the frame where the hammer slides in.  Some of this area is pretty complex, so it's going to take some time to get all the measurements and transfer that data to the solidmodel.  Below is what the model now looks like.

beretta92fs_3.jpg (60327 bytes)


I now have the solidmodel complete!!!  I'm going to go back and triple check all my measurements before I start the machining to insure I didn't make any mistakes.  I created a Solidworks E-drawing of this frame if any of you would like to check it out to see if I've missed any features.  Please let me know if you find anything wrong.  The E-drawing is located under Downloads. Below is what the final model looks like.

beretta92fs_4-.jpg (58744 bytes)


Since I now have the solidmodel finished of this frame, I decided to have it "printed" out on a rapid prototype machine also know as stereolithography.  Doing this will allow me to test fit some of the parts, and insure my solidmodel is correct before I start the complicated machining process.  I've looked the prototype part over and everything seems to be ok so far.  I'm now getting REALLY excited about starting the machining of this frame.  Below you can see a couple pics of what the "printed" frame looks like.  This frame is made of a clear type of material so it was hard to get a good picture, but maybe you can get an idea of what it looks like.  I had to use some "silly putty" to help stand the part up for the pictures.

P2020208.JPG (63391 bytes) P2020209.JPG (61226 bytes)


I cut the aluminum blocks from a 12 foot bar of 6061-T6 aluminum we had at the shop.  The dimension of the block is 5" x 1½" x 7.760"  Below you can see a pic of my reference Beretta 92FS frame setting on top of the aluminum block.  I stripped my Beretta 92FS pistol down to just the bare frame, and I used it to help make the solidmodel.  I hope to be able to turn this 5.7 pound block of aluminum into a working pistol frame.

P2030215.JPG (73572 bytes) P2030217.JPG (73909 bytes)


I got the 1st operation programmed for the CNC machining center.  Here is a picture of the block of material in the machine ready to be machined...

IMG_0102.jpg (128028 bytes) 

The first tool to run was the 1/2" hog end mill.  This tool was used to rough out most of the material.  I stopped the machine mid-way through this tool to show how much material was being removed...

IMG_0104.jpg (156896 bytes)

The next picture I took was after the outside profile had been roughed out.  You can see the block of material is now taking the shape of the pistol frame...

IMG_0107.jpg (130702 bytes)

Below is a pic after the 1/2" hog end mill was finished roughing the frame...

IMG_0110.jpg (129287 bytes)

You can't really see it from the pictures, but the 1/2" hog end mill leaves a rough surface finish on the material.  So I used a 1/4" carbide end mill to make the finish passes around the frame.  I used the flash on the camera for this picture to try and show how clean the outside of the frame now looks...

IMG_0112.jpg (140124 bytes)

The next picture is after all the holes were drilled...

IMG_0115.jpg (101934 bytes)

The next tool to run was the 1/4" ball nose carbide end mill.  This is the tool that can really turn a block of material into a work of art.  This tool profiles across the whole frame by machining square corners into smooth surfaces.  This tool takes a really long time to run because it's machining back and forth across the frame stepping over 0.006" each time.  This tool made a total of 1260 passes back and forth across the profile of the frame.  Below are a few pics showing what the frame looked like after this tool...

IMG_0119.jpg (128161 bytes) IMG_0120.jpg (115689 bytes) IMG_0121.jpg (116157 bytes)

I then ran a couple different smaller end mills to finish up some minor details and to cut the rail for where the slide will fit.  Here is what it looked like after those tools...

IMG_0123.jpg (125712 bytes) IMG_0124.jpg (120585 bytes) IMG_0125.jpg (120768 bytes)

That's the end of the 1st operation.  There are some other cuts I could have made on this operation, but I'm saving those for further down the road.  If I made some of those cuts now, it would have made the frame really weak and hard to hold onto for future operations.  I bead blasted one of the frames to reduce the glare in the next pictures.  Below are some better pics, and the last pic is showing the difference between the freshly machine frame and the one that had been bead blasted...

P2050218.JPG (71290 bytes) P2050223.JPG (70759 bytes) P2050222.JPG (69387 bytes) P2050224.JPG (67239 bytes) P2050227.JPG (73340 bytes)

I hope to find time to program and machine the 2nd operation in the next few weeks.


I found time to test fit some of the parts to the "printed" stereolithography prototype frame.  I didn't add all the parts including some of the springs because the plastic frame wouldn't be able to withstand the forces put forth by some of the springs.  I didn't find any problems assembling the parts to the frame, so this far everything looks good to go.  Below are some pretty cool pics of how the clear plastic frame looks with the parts assembled...  

 P2060230.JPG (73388 bytes) P2060229.JPG (74055 bytes) P2060234.JPG (72908 bytes) P2060233.JPG (72898 bytes)


I got the 2nd operation machined.  I had to make a special fixture plate to hold the frame flat and square while I done the machining.  I designed the fixture plate so I could use it for both the left and right machining processes.  I also made the plate to where it would touch as many surfaces as possible, so it would help keep the frame level during the machining process.  Below you can see what the fixture plate looks like...

P2190237.JPG (69798 bytes)

After I got the fixture plate machined, it was time to start machining the 2nd operation on the frame.  Below you can see what the plate looked like in the vise before I clamped in the pistol frame...

IMG_0141.jpg (107875 bytes)

I then clamped the frame in the vise and used a machinist level to make sure the frame was flat in both the "X" and "Y" directions.  The plate worked perfectly, because everything came out exactly level.  Below you can see what the frame looked like before I started the machining process...

IMG_0143.jpg (96362 bytes)

First I used a 1/2" hog end mill to cut around the outside of the frame to remove the slab of material left over from the 1st operation.  Here you can see what it looked like with the slab removed...

IMG_0145.jpg (88861 bytes)

Next, I used a 3" shell mill to bring the frame to the correct thickness.  I stopped the shell mill while it was cutting to take a picture to show exactly what it was doing.  You can see this picture below and also the picture after the shell mill was finished cutting...

IMG_0146.jpg (113032 bytes) IMG_0147.jpg (76591 bytes)

Then I used the 1/2" hog end mill again to rough most of the material away so I could run the 3-D profiling process at a faster feed rate.  Here's a pic of what it looked like after this tool ran again...

IMG_0151.jpg (83655 bytes)

Now for the fun part, to give the frame some curves and shapes..  To do this, I used a 1/4" ball nose carbide end mill running at 10,000 RPM and 125 IPM (inches per minute) with a 0.006" step over.  This is the same process I used for 3-D profiling the 1st operation.  Below are some pics after this tool ran...

IMG_0159.jpg (89931 bytes) IMG_0162.jpg (79357 bytes)

Click this link or on the picture below to watch a short video of the 3-D profiling process.  You can see from the video that the profiling process can take a long time.  Each time the tool goes across the frame, it steps over 0.006"  Then machines back across the frame, and steps over 0.006" again.  It keeps machining back and forth until it reaches the end of the frame, making a total of about 1,300 passes.  I looked back at the program to see how long it was, and for the whole 2nd operation, the program is 58,976 lines long!!!  Good thing I didn't have to program each line by hand. 

Next I used a 1/4" carbide end mill to mill the slot for the slide release and to contour around the grip screw bosses.  I also contoured the relief hole for the hammer pin hole.  Here's what it looks like now...

IMG_0163.jpg (93348 bytes)

Then I used a 3/32" carbide end mill to cut the slide rails and to cut the relief for the trigger pin hole...

IMG_0164.jpg (88492 bytes)

The last thing to do was to drill the hole for the slide release lever.  Below are the pics after the 2nd operation was finished...

IMG_0165.jpg (83640 bytes) IMG_0166.jpg (92412 bytes)

I bead blasted the frame and took some better pics to show exactly what has been done this far.  Below are a couple pics to show how the fixture plate can support the pistol frame on both the left and right side...

 P2190239.JPG (69932 bytes) P2190238.JPG (71340 bytes)

Here are a couple close up pics of the frame... 

 P2190243.JPG (65946 bytes) P2190240.JPG (71454 bytes)

Below you can see how I am able to slide the slide on the pistol frame.  Of course I can't slide it all the way back yet, because the top of the frame hasn't been machined.  My plan for the next operation is to cut the top of the frame.  Hopefully, this will allow me to test fit the slide with the barrel in place.

P2190246.JPG (68596 bytes)


I got the fixture plates made for the 3rd operation.  I will be able to use it for the 3rd operation as well as the operation where I stand the frame on end to machine down inside the dust cover.  The fixture plates have 2 bosses on each side to align the frame on the fixture.  This will insure the frame is setting exactly square in the vise when I start the machining.  Below are a few pics of what the fixture plates look like...

 P2270250.JPG (72082 bytes) P2270251.JPG (66028 bytes) P2270247.JPG (66164 bytes)


I got the 3rd operation machined.  For the 3rd operation I had to do something I've never done before.  I had to make a special broaching tool and try to broach on the CNC machining center.  I needed to use a broach to cut the corners square where the trigger falls in the frame.  Those corners aren't that big of a deal, since I could just file them square with a hand file like I've done in the past with other frames, so the real reason why I needed the broach was to cut the ramp inside the frame where the bottom of the barrel locking block rides up.  Since I was going to make a broach tool, I wanted to make sure I could use it in more than one place.  So for this operation I used it to cut the ramp, and to also cut the corners where the trigger falls through.  I will also be using it on one more operation, it'll be the operation where I make the relief cuts for the hammer slot.  Before you try to use a broach on a CNC machining center, you have to make sure the spindle will lock in position, so there is zero rotation of the tool in the machine.  Come to find out, broaching is no easy task on a CNC machining center.  Not only do you have to make your own special broach, you have to make a program with a tool that doesn't necessarily cut on center.  By this I mean, when you grind the tool, you try your best to make sure the tool is on center, but it can be off a little bit because of how many times you have to flip and rotate the broach tool on the surface grinder.  And something that didn't help the situation was the fact that I'm not a tool & die maker, and I've never made a broach tool before.  I started out with a piece of 3/8" square A2 tool steel blank.  I first set the blank up on the surface grinder to grind the end round, so I could have a round shank to hold on to in the CNC machining center.  Then I set the blank up to grind a 15 degree angle on the end of the tool, this will be the main cutting edge.  Next I ground a 7 degree relief cut on the front and sides of the tool.  And finally I ground the back of the broach flat so it would have clearance inside the trigger hole in the frame.  I have no idea if those are the correct angles for broaching aluminum, but they looked good when I modeled it up in Solidworks, and it ended up broaching perfectly.  I left a 0.125" step where the square part of the broach meets the round part.  I done this so the tool wouldn't be trying to slide up inside the holder while I was broaching.  After I had the broach made, I wanted to heat treat it so it would last a long time.  Before heat treat, the steel was something like a hardness of Rockwell C30.  After I finished heat treating the broach, it was Rockwell C62.  Below are a few pics of what the broach tool looked like after it was finished...

IMG_0208.jpg (80746 bytes) IMG_0207.jpg (72407 bytes) IMG_0206.jpg (73678 bytes)

Now that I had the hard part finished, it was time to start cutting some aluminum.  I used some 4" tall steel jaws to help hold the fixture plates and the frame in the vise.  I also used a c-clamp to hold the top of the vise tight, since using tall jaws usually makes the top of the jaws spring open when you try to clamp on a part.  Below you can see what the setup looked like at the start of this operation...

IMG_0209.jpg (83563 bytes)

I first used a 1/4" carbide end mill to mill out the dust cover area, the trigger area, the slide release area, and also the hammer slot in the back of the frame.  Here's what it looked like after that tool ran...

IMG_0212.jpg (94658 bytes)

Next I used a 1/4" ball nose carbide end mill to 3-d profile the dust cover and the rest of the frame...

IMG_0214.jpg (93525 bytes) IMG_0215.jpg (92237 bytes)

Then I used a 1/8" carbide end mill to mill the slot where the hammer release lever and firing pin catch lever falls into place.  I also used this tool to clean up the corners down inside the frame where the trigger will fall in.  This will allow me to only broach out 1/16" worth of material instead of 1/8" worth of material that was left behind by the 1/4" end mill.  Here's what it looked like after those areas were cut...

IMG_0216.jpg (103170 bytes)

I also used the 1/8" carbide end mill to mill two ramps on each side of the frame.  In the pic below you can see the arrow pointing to the area that was cut.  After the 1/8" end mill, I used a 5/16" x 0.060" woodruff cutter to cut the undercut for where the lugs on the barrel locking block slides into the frame...

IMG_0218.jpg (41562 bytes) 

Now it was time for the fun part.  It was time to see if my broach tool would work.  After I had the broach in a tool holder, I had to make sure it was square with the machine.  To do this, I locked the spindle in position so the tool wouldn't turn.   Then I took the tool down to right in front of the vise jaw.  At this point the spindle was locked, and I had the broach loose in the tool holder so I could turn it.  I kept stepping the tool forward and wiggling the broach until the broach was within about a thousandth of the vise jaw.  Then I tightened the broach in the tool holder.  So now the broach was exactly square with the vise jaw which meant it was square with the machine.  When I programmed the broaching tool path on the computer, I programmed it from the corner of the tool.  It would be hard to find the exact center of the broach compared to the tool holder, so it was easier to just program it from the corner.   To find the corner, I jogged the machine to the front and then the side of the vise jaw.  I knew where the exact corner was of the vise jaw, so after I touched off on the front and the side of the broach on these surfaces, I could then calculate the where the corner of the tool was in relation to the tool holder center.  Getting all of this setup was a lot of work, so after this operation I left the broach in the tool holder so I could use it on the operation where I'll be broaching the area for the hammer slot.  I broached the left side of the trigger hole and the barrel locking block ramp, then I told the machine to rotate the tool 180 degrees and lock the spindle in again so I could broach the right side of the trigger hole.  To broach the trigger hole, the broach only had to make a up and down movement.  But to broach the angle, I had to not only program the broach where it would go up and down, I had to also program it where the broach would come down as it was also moving left to right.  Below are a couple pics of what the broach looked like before it started broaching.  In these pics, the broach is set to cut the left side of the trigger hole and the ramp...

IMG_0219.jpg (89247 bytes) IMG_0221.jpg (74856 bytes)

I shot a short video to show the exact movements the broach was taking as it was cutting the barrel locking ramp.  Click on this link or on the picture below to view the video...

The broach tool ended up working perfectly.  I was really happy with the results of the first time I tried to make a broaching tool and broaching on a CNC machining center.  

I took some better pics to show what the frame now looks like.  I also took a couple close up shots to show exactly what this operation done...

P3050254.JPG (68686 bytes) P3050253.JPG (69227 bytes) P3050256.JPG (78833 bytes) IMG_0028.jpg (40552 bytes)

I was excited to see how the slide and barrel fit in the frame, so I got some parts together and tested it out.  Below you can see what it looks like...

P3060322.JPG (200199 bytes) P3060314.jpg (51149 bytes) P3060318.JPG (185867 bytes) P3060317.JPG (190055 bytes)

So far everything is fitting perfectly.  I can't wait to get this thing finished and take it out for the first test shots.  That's it for the 3rd operation.  The next operation I'll be standing the frame on end, and machining down inside the dust cover so the guide rod and spring can slide in.  

Continued on page 2...

Page 1   Page 2