Generic Oilfield Engine

[cfellows]

Thanks, Maury.  I hadn't thought about custom threads or pipe threads, but I had considered worms for worm gears.  Grinding a cutter for worm profiles might be a bit of a stretch for me, but certainly do-able.

Thanks, Jo.  The threaded stud in the video and pictures is 3/16" (4.8mm).  I suppose the lower limit to size would be determined by the amount of flex in the rod to be threaded.  The upper limit for my set-up is 1/2" and limited in length to around 1.25".

Thanks to everyone for your comments.  A lot of satisfaction in seeing things like this come together.

PS... Here is the G-Code:

z-0.032           (Set the cutting depth to .032")
x0.375 a9        (Move the x-axis 0.375" while rotating the 4th axis 9 times)
z0.1                (Raise the cutter to .1" above the work)
x0 a0              (Return x- axis and 4th axis to home)

[cfellows]

Spent a lot of time today breaking new ground and experimenting with CNC.  Here's the result... a prototype of the front side-shaft bracket cut from MDF. 



I thought it might be interesting to see how I developed the G-Code for this bracket.  First, I drew the design using Visio.



Then, I drew in the small, red, numbered circles at the different x,y points where a new g-code command would be required.  The number represent the order of the points visited by the cutting tool.  The circles are 1/8" diameter, which is the diameter of end mill used to cut the part.  In the drawing, you can see that circle number 1 is highlighted.  In the upper left corner, you can see where Visio displays the X,Y coordinate of the center of the highlighted feature, in this case circle #1.  By moving the cursor from circle to circle, it becomes an easy task to code the g-code commands for each point.  The only additional consideration is where the cutter has to travel in an arc.  All you have to do is use an arc move command and additionally specify the coordinates of the center of the ARC.  All this code is placed in a subroutine which is looped through n times, dropping the cutter a bit on each loop until the part is cut out.

Chuck

[DavidF]

Thanks, Maury.  I hadn't thought about custom threads or pipe threads, but I had considered worms for worm gears.  Grinding a cutter for worm profiles might be a bit of a stretch for me, but certainly do-able.

Thanks, Jo.  The threaded stud in the video and pictures is 3/16" (4.8mm).  I suppose the lower limit to size would be determined by the amount of flex in the rod to be threaded.  The upper limit for my set-up is 1/2" and limited in length to around 1.25".

Thanks to everyone for your comments.  A lot of satisfaction in seeing things like this come together.

PS... Here is the G-Code:

z-0.032           (Set the cutting depth to .032")
x0.375 a9        (Move the x-axis 0.375" while rotating the 4th axis 9 times)
z0.1                (Raise the cutter to .1" above the work)
x0 a0              (Return x- axis and 4th axis to home)

  Ok im scratching my head...dont you need a speed in there somewhere??  im getting ready to venture into cnc...oh boy!! i see alot of broken cutters in my future LOL

[cfellows]

  Ok im scratching my head...dont you need a speed in there somewhere??  im getting ready to venture into cnc...oh boy!! i see alot of broken cutters in my future LOL

Yeah, you need an Fn where n specifies the speed anytime you change the speed.  Otherwise, the speed remains the same at whatever it was last set to.  I deleted the F codes in the g-code I posted to emphasize the axis movement coding.

Chuck

[DavidF]

chuck are you learning g code from a book or website, if so where can I find the information? i plan on starting my g0704 conversion next week...

[Woodguy]

You can of course mill threads without a 4'th axis or a horizontal adapter - something I'm sure Chuck already knows.  Clamp the stock vertically and have the cutter describe a helix around the stock.


There is a good deal of discussion about this on  cnczone.


Chuck's method is more intuitive though and (ever the experimenter) he did it using the horizontal adapter and 4'th axis - both of which he built.


Thanks to Chuck for posting his method. You really cannot argue with those threads - perfect.

[Hugh Currin]

Chuck:

You're doing better than I. I would make some stupid mistake somewhere in writing the G-code and I scare myself often enough already.

Have you thought about getting a CAM package. This makes generating G-code much easier and less error prone. I'm now using CamBam and it seems to work well. The learning curve is fairly steep if you've not used a CAM system before, but well worth the effort. A computer simulator has helped me also. I use a free Linux bare bones simulator. CamBam offers Cutviewer which I haven't used but looks useful. This can catch some errors before going to the mill.

A CAM package allows some nice options which are difficult to hand code. I use entry and exit arcs which start cutting outside the part and arc in to the part. This helps prevent undercutting where the tool sits stationary against the part. Drill cycles are also handy.

Very nice build you have going. I like the thread milling also. I need to find the time to build a 4th axis for my mill. (too many projects) Thank you for putting up a build log, I am enjoying it.

Thanks.

Hugh

[cfellows]

chuck are you learning g code from a book or website, if so where can I find the information? i plan on starting my g0704 conversion next week...

I read through the entire tutorial on Bob Warfield's website:

http://www.cnccookbook.com/CCCNCGCodeCourse.htm

Well presented and easy o digest.  Unfortunately, Mach3 doesn't support all the code covered in the tutorial so you'll need to figure that out as you go.  Mach3 also has g-code and m-code reference information available from the main screen.

Chuck

[cfellows]

You can of course mill threads without a 4'th axis or a horizontal adapter - something I'm sure Chuck already knows.  Clamp the stock vertically and have the cutter describe a helix around the stock.


There is a good deal of discussion about this on  cnczone.


Chuck's method is more intuitive though and (ever the experimenter) he did it using the horizontal adapter and 4'th axis - both of which he built.


Thanks to Chuck for posting his method. You really cannot argue with those threads - perfect.

Yeah, I read about that and watched a video by Tormach.  However, it seemed to me that since the cutter is not set an angle that the threads might be a little bit sloppy.  Also, the 4th Axis solution is a lot easier to program if you are doing your own g-code.

Chuck

[cfellows]

Chuck:

You're doing better than I. I would make some stupid mistake somewhere in writing the G-code and I scare myself often enough already.

Have you thought about getting a CAM package. This makes generating G-code much easier and less error prone. I'm now using CamBam and it seems to work well. The learning curve is fairly steep if you've not used a CAM system before, but well worth the effort. A computer simulator has helped me also. I use a free Linux bare bones simulator. CamBam offers Cutviewer which I haven't used but looks useful. This can catch some errors before going to the mill.

A CAM package allows some nice options which are difficult to hand code. I use entry and exit arcs which start cutting outside the part and arc in to the part. This helps prevent undercutting where the tool sits stationary against the part. Drill cycles are also handy.

Very nice build you have going. I like the thread milling also. I need to find the time to build a 4th axis for my mill. (too many projects) Thank you for putting up a build log, I am enjoying it.

Thanks.

Hugh

Thanks, Hugh.  I had to do a number of practice runs to get the final result.  I actually made 4 videos before got it right.  The code was pretty easy, but I forgot to tighten the er-collet securing the blank the 4th axis in the first video.  On the second video, I had forgotten to tighten up clamping bolts on my mill / drill head.  I think the third video I had the cutter set too deep (because of the first and second problem...  )

I had considered CamBam as my first choice based on what I'd read on this and the HMEM forum.  I will probably buy it eventually, but for now I'm having fun with just using g-code.  I guess once a programmer, always a programmer...

Chuck

[cfellows]

Progress has been a bit slow this past week.  These side shaft engine designs are a lot more complicated when it comes to the valve actuating assemblies, cams, etc.   And, since I'm making this up as I go along, it requires a bit more thought in getting everything to work right.  I've now nearly finished and fitted the rocker arm support posts, which are made from 1/4" square steel rod...





And here are the lifter arms, made from 3/16" square steel rod. 



The solid end will pivot on a post which screws into the head and the slotted end will carry a small roller, 1/4" diameter, which  rolls on the cam.  Next I have to make the post and fit it to the head.

Chuck

[cfellows]

Another day working on fiddly bits.  I'm in the middle of doing the valve assemblies and today I worked on the exhaust lifter, roller, and push rod. 



The lifter bar is 3/16" square steel rod about half an inch long.  The push rod is 3/32" drill rod and the screw that holds it all together is 3-48.  The roller is made from 1/4" drill rod.  The pushrod end is 3/16" cold rolled steel, drilled and tapped 3-48 to accept the push rod.

Tomorrow I'll work on the intake assembly.

Chuck

[b.lindsey]

That is some great "small" work Chuck...I can relate from the current project I am working on. Sometime the little fiddly bits can be fun though

Bill

[cfellows]

Thanks, Bill.  I've got the valve assembly, rocker arms, push rods, lifters and cam rollers pretty well done.









This whole assembly was the hardest to design and the most worrisome about how it would turn out, not to mention making these little pieces and getting them to look pretty decent.  I'm pleased with the results and with a little fine tuning and some proper fasteners, it's going to work just fine.

Chuck

[Tennessee Whiskey]

Chuck that's a very nice looking valve train. Are the two holes in the right rocker arm post for another attachment?

Whiskey