Hello Guest it is May 20, 2024, 09:13:47 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - adprinter

Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 »
I won't be able to use the MDF with Bronze, too hot. May try plaster or make a wood positive and cast it in sand.
All of my aluminum casting attempts have been done in sand. Although, the very first time I tried pouring some molten- I had a couple of pieces of plywood that were left over scraps from cutting some wooden gears out on the CNC Router table. I will never forget that experience! I knew that the molten would probably burn the plywood, but I THOUGHT that it would at least solidify before too much burning occurred. WRONG! That plywood did not even slow the molten's descent down, on it's way to the ground! (It literally VAPORIZED the plywood on contact!).
Since learning from that experience, I have been making sand molds. And being retired, (and unable to afford Petrobond casting sand), I learned how to make my own foundry sand. After trying many different recipes, the one that finally worked with good results was: 80% sand, mixed with 15% Bentonite clay (CLUMPING type Kitty Litter), and 5% Charcoal ashes. I had to construct a special machine to use for grinding the kitty litter into powder, which consisted of a steel turn-table, with two grind stones (like those on a bench grinder) which ROLL against the surface of the steel turn-table. A variable-speed DC motor was used to drive the turn-table, at the slowest possible speed, while friction drives the grind stones.

The slow speed is needed to prevent centrifugal force from simply slinging the kitty litter to the outer rim of the turn-table, (and thereby OUT of the path of the grind stones). Only small amounts can be ground at a time, as it is after all, Bentonite clay, and will begin to clump onto the grind stones. A knife-edge was later added to the stones, to keep the clay scraped off during the grinding process. It is also helpful, to first DRY the kitty litter, by placing it in a pot, and into the foundry furnace, stirring constantly until it is completely dry. Once the kitty litter clay has been ground into the finest possible powder consistency, then pour it into the sand, add the charcoal ashes, and begin mixing by hand, until all three elements are evenly mixed.

 I use one of those plastic tubs, with a snap-on lid to store the foundry sand in, to prevent (at least as much as possible) the sand mix from becoming too dry, or too wet. When using the sand to make the sand molds, I use a spray bottle of water, and lightly moisten the sand, stirring constantly by hand, until it becomes wet enough to form cakes when squeezed in my hand. At that point, it is ready to pour into the flasks to make the sand mold. The sand mix is re-usable, however (because of the kitty litter) has to be re-ground to break up the clumps back into powder form. I have learned, that this step is only necessary for a small amount. (The sand that actually contacts against the pattern being molded). The rest of the sand acts as just a "filler" of the flask. And when rammed up (packed into the flask) properly, it will ALL become very similar to concrete. Especially after a pour, and must again be broken up, and ground. A five gallon plastic bucket, with a few rocks thrown in, and turned tumbler-style (again, S-L-O-W-L-Y) will speed up the process.

Sorry to ramble, but I know that others are reading this with interest, and I just wanted to warn folks that metal casting is VERY DANGEROUS. I am actually amazed at the results you were able to achieve with the MDF, that it did not Vaporize on contact with the molten metal! I would hate to hear that someone tried this, and lost a foot or something, as the molten metal poured right through the MDF or other non-fire proof mold material. Metal Casting is very dangerous, and must be done with respect, paying attention at all times to what you are doing. Serious injury can occur!

Show"N"Tell ( Your Machines) / Re: Scratch build Small mill
« on: September 13, 2013, 01:13:15 AM »
Can't tell from the photos, what size motor are you using to drive the Z axis? (From the looks of what you have so far, it's gonna need a hefty motor to lift all that weight without hesitation). Also, is that a draw bar spindle you have built? I'd like to see the specs on that, since I have considered doing something similar to my machine. What type of motor are you using to drive the spindle? Servo, stepper, or standard drill press variety? Looks like you've got a fairly hefty design so far!

Share Your GCode / Re: Clock
« on: September 04, 2013, 12:16:16 AM »
I don't have any G code files to share, but can tell you what I did for a clock project. I searched the internet for wooden clocks, and found a site which featured free plans to build a wooden clock. The drawings were contained in a PDF file, which I used as a pattern (gleaning the gear tooth counts, diameter info from the drawings in the PDF). I then used a program called Wood Gears, which was only a $20 download at the time, to create the profile drawings for the various gears needed. Exported the drawings as HPGL plot files, opened them from LazyCAM, and created tool offsets in LazyCAM (then deleted the actual drawing). Then exported the G code generated to Mach3. And cut the various gears. The PDF file lacked some of the drawings (specifically, the Yoke piece for the Pendulum), so some creative drawing, cutting, and experimentation was necessary to produce the missing piece. After several months, I actually did succeed in building the clock. However, I learned that 1/4" plywood is a poor choice of material to use for constructing the gears. Because it WILL warp. And it doesn't take much warpage at all, for the gears to become unmeshed! I plan to try again, using some thicker material, but just haven't gotten around to it yet. As for the PDF file I mentioned, do a search for the Law Wooden Clock. The site is somewhere in the UK, and they do also sell clock plans which include the DXF files. Good luck with your project. And be patient, it won't happen overnight. But it is a good project to begin learning about your machine.

To improve the finish cuts, what I have done is to increase the resolution (stepover). MeshCam Art has an option to use Adaptive Stepover, which does result in a better finish cut. However, it also results in an increase in cutting time. I just finished a piece, which is Old Glory flapping in the breeze. The total cut time for that piece was almost 18 hours. But it did turn it pretty nice. I will post a picture later.

I know that this is maybe not the correct place to put this but as there are a lot of people here who are using routers I thought that I might just get a better response.
Anyway I have a problem to which I do not seem to be able to resolve with my router when machining either soft or hard wood and it is now getting very annoying if you look at the picture you will see what I mean I am getting lines which you can clearly see and no matter what I do to try and cure this problem nothing seems to work, I have tried all variations of feeds and speeds but nothing seems to work, I have a solid machine not a small machine that I have built it does not move or vibrate in any way, as you can see by the picture of the machine it is 5 axis and as I say it is all very solid so I am just hoping that someone just might know the answer to my problems  and please do not tell me to just sand these lines, these lines are around 0.5 to 0.75mm deep and I have tried machining at all different angles with the grain against the grain etc.
So please help me out here I have not got much hair left to pull out
I use MeshCam Art, and have experienced similar results in the finishing tool cuts. This was prior to installation of the 2010 Screeen Set in Mach3, and was the result of setting the Z Zero after tool change (from the roughing tool, to the finishing tool) by using a sheet of paper as a "feeler gauge". However, as I said, this problem was prior to the 2010 Screen Set in Mach3. Consistent Z Zero settings have been one of the best aspects of the 2010 Screen Set, and accompanying macros. Believe it or not, Depending on the piece, you may actually have occassions, when it is actually desirable to do the finishing cuts "Just a hair shallow", which does result in finished pieces with textures similar to those you have pictured.

Hi Joe,
I am no expert in CNC, but I can tell you a little about what I have been able to do which has worked for me. I initially built a 3 axis router table, using Mach3 as the control software. I later expanded the table's capabilities to include a 4th axis. After looking around the internet for software which would work on a 4th axis, I tried CNC Wrapper. Basically, it takes a 3 axis G-code file, and substitutes either the X axis, OR the Y axis movements into 4th axis movements (depending on your specifications when you initially setup the software). The CNC Wrapper program also has stand-alone capabilities, which will allow you to do basics similar to what you have described. (I.E. turn square stock into round stock). However, just as it would be done manually in preparing a piece of stock for turning on a wood lathe, it is best to cut miters along the length of the stock, thereby changing it's beginning shape from a rectangular, or square shape to an octagon using a table saw. I have since experimented with 3D G-code files which were created using MeshCAM Art, then converted using CNC Wrapper, to generate G-code which was then loaded into Mach3 and "Wrapped around" round stock pieces on the 4th axis of the machine.

General Mach Discussion / Re: Home problems
« on: July 03, 2013, 01:42:36 AM »
Ok so I guess it was some kind of noise problem because I had the home switches wired to normaly closed and now I wired them to normaly open and the machine will run all the way until it makes the switch and then stop in all 3 axis,,,,,,,,, Yea !!!!!!!! The only thing I don't understand is before when it was wired to normaly closed when it made the switch it would back off just until the switch was no made now when it makes the switch it stops there and that is it. That is OK just differant. The only other thing is the auto zero no longer works when the home switch is hit and I still have it checked in the config page,,,,, what is up with that ?
It sounds like a problem I too experienced. And like you, I changed the limit switch logic to normally open. However, the problem actually turned out to be a sticking switch (one time), and another time it was actually a broken wire. Still another time, it was a broken switch. You can actually save yourself a lot of head aches, by powering down everything, and just checking the limit switch circuits with an ohm meter. Cycle the switches on and off, to make certain the contacts are not sticking, check the wiring for continuity, etc. before taking the step of changing circuit logic, or Mach3 logic and pulling hair!

Feature Requests / Re: Print to Cut functionality & Mach3 on router
« on: June 04, 2013, 11:31:03 AM »
I worked for over 35 years in the printing industry, and saw many changes in technology over the years. (From letterpress printing, to web offset printing of telephone directories). The most advance system I personally worked with, was a computer-controlled registration system. (CCR for short). It involves the use of registration dots which were printed onto the web travelling at speeds up to 2500 feet per minute, for each of the four process colors used. CMYK, a television camera was mounted onto a linear rail system, which allowed the camera to travel across the web, driven by a servo motor and toothed belt. The camera would make a scan across the moving web, under computer control and photograph the programmed location of the registration dots on the web itself. A specific pattern arrangement of the dots (I.E. the distance from Cyan to the Magenta, Cyan to the Yellow, Cyan to the Black, and so on, relative to each of the 4 colors) was the "Target" to be maintained by the CCR. This was accomplished, by electric servos, which manipulated the registration of the plate cylinders for each color on the press itself. To be sure, the system was a multi-million $ system. And further improvements which included a computer controlled ink density scanner and closed loop control (to control dot gain) made that particular printing press the most advanced system I ever worked on. So, I can see where it could be possible to develop a similar application for what you are trying to do. However, I don't think that Mach3 (as it is) could accommodate such a system, since it would require constant re-writing of the G code during program execution.

This was created in acrylic from GCode produced with 'DotG' using an impact magnet.
Is the 'Impact Magnet' you are referring to, an "Electric Engraver" tool? (The brand I have is a Wis, made in China).
The length of stroke is so small on this tool, that it appears to just vibrate. Regardless of the various stroke settings' position. I tried to mount it on my machine to engrave some acrylic. Turned into a 3 day project, just trying to get a successful design implemented on the tool holder. However, even using the surface of a freshly planed piece of oak stock, as the mounting surface for the plexiglass, the results were inconsistent. The tool would plunge so deeply into the plexiglass, that it could not complete it's strokes in some areas, while in other areas the tool tip would not even contact the surface!
 I have already proven The acuracy of my machine's table, by doing engravings using the router tool into solid oak, which involves multiple passes to achieve the desired depth of cut with good results. So the inconsistencies I am experiencing with the Engraver tool MUST BE in the Engraver tool itself. What is the brand name, or supplier info of the 'Impact Magnet' tool you are using?

Looks like that the best way is to machine the top. I will complete my leveling and if the results are not good then I will machine the top.

As for squaring the machine, I did a 10"x10" square and the last leg of the square did not hit the starting point by about 1/8". I adjust the home switches on the Y&A axes a few times and finally I got a perfect square (but the depth of cut was not equal). So as far as I know this proves that the Y&X axes are square, unless I'm wrong
Now REPEAT the cut, and set the Z axis a little deeper. Does the machine EXACTLY follow the first cut? This is the true test of the machine's accuracy, the ability to return to an exact point each and every time. There are many factors at work in "leveling the table", if the Z axis travel is not perpendicular to the table, in both the X and Y directions, then even machining the table top will not make it truly flat. Also the router's actual mounting block or clamp can effect this. Because the cutter will cut deeper into the MDF along one side of it's path, creating a "Shingled" set of tool paths. Actually, this will occur without the need to machine a whole table top, only to discover that the Z travel is not at 90 degrees with the table's surface. By simply cutting a pocket into a smaller piece, fastened to the table top. Only when you are able to do this, (with NO Shingled paths, but a smooth, flat surface as a result), should you consider machining the table top to make it flat. Even then, you are talking wood. It changes with the weather. It all depends on what you want to use your machine for. I recently tried mounting an electric engraving tool onto my machine, to engrave some plexiglass. And learned that the table "flatness" which had worked so well for my wood carving projects, was FAR from being "flat" enough to achieve the consistency needed for the engraver's micro-strokes. Some areas, the tool would plunge so deep into the plexiglass, that the tool could not even complete it's strokes! While other areas, the tool would not even contact the surface with the tip of the tool. I actually obtained a granite counter top slab, to use for calibrating my table top. With intentions of using dual dial indicators mounted in the router spindle to set the table as close to perfect as possible. But perfection is simply a target to aim at, never actually hitting. The best anyone can hope for, is to get it as close as possible. The best results I have achieved for things which require a very tight tolerance, invloves cutting a pocket into a larger piece of stock mounted on the table, and (while still bolted down to the table), attaching the workpiece to that! (As long as I am ready to start cutting right away). Even this method does not work very well, if the planed stock has been left bolted down overnight.  Wood never truly dies..... it is constantly squirming!

Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 »