A little confused

[Cold Fusion]

Alright, I've been reading all about the software side of cnc and I think I'm confused I know that everything starts as Cad, which I can make myself easily. Then it goes into a cam program? After the cam program does it go to a program like Mach1? Is it possible to skip the cam program and go straight to Mach1 with an Iges or Step? Also, how do I do drill cycles? Here are all the bits needed in one run of chassis. Carbide Cutter, 1/8 drill bit, 5/32 drill bit, 3/16 drill bit, end mill, and countersinking bit. I know that the more expensive Cam programs have all the features to do this, but I can't afford it. Sorry for the ignorance but I need someone to someone to lay it out for me

Similar Threads:

• confused
• A little confused.
• Need Help!- Confused
• Confused with SFP
• Need Help!- I'm confused

[turkgeltz]

After you've drawn it in cad, you use the cam software to toolpath it and convert it in to g- code (machine language). Then you use the software like Mach1 to co-ordinate it with the hardware on the machine(motors and controller).

[Scott_bob]

with a good CAM system, who needs a CAD system.

On the subject of lowest cost approach to CNC mfg:

As far as im concerned the point of a CAD system is to design your part.

But, the CAD data is usless to a CNC machine.
CAM systems talk CNC language, CAD systems do not.
So, Eliminate the CAD system (for entry level Design/Mfg).
You practically can't do much CNC programming without CAM.

There are some very good CAM systems that will prolly do everything you need to do...

Regards,
Scott_bob

[ger21]

Those CAM systems are probably way more than he wants to spend, too. I posted a link the other day to a free CAD program, Intellicad. This will let you save you're files as .dxf's. I believe Mach1/2 can convert you're .dxf files to g-code. You can also use ACE converter to convert .dxf's to g-code, which is also free. Available from http://www/dakeng.com. With all those different size holes, the easiest thing would be to use a cutter no bigger than the smallest hole and mill all the holes with the same tool you're cutting out your parts. It would probably be easiest to countersink the holes on a drill press while the next chassis is being run. Otherwise, you'll be spending more time changing tools than making parts.

Gerry

[Cold Fusion]

Ok, I've got all my Iges chassis files converted to Dxf. I have decided to drill all the holes with a 1/8 bit then go back and enlarge and countersink with my drill press. Now I have three problems left.

1: I need to make a 1/8 deep pocket in the chassis

2: I need to be able to switch the drill bit with a routing bit.

3: I need to have 6-8 chassis per sheet. Can I get that done in G code?

I can not afford to buy a fancy cam system right now. Someone on the boards offered to make my G code for me for a small price. I like the idea of this until I can do it myself. If it doesn't work out with him is there anyone else who wants to make some money generating code for me? I would have at least 7 jobs to start with and probably 10 more later on.

Btw, which do you prefer, Mach2 or Master5?


Here is a picture of the chassis I do.

[buscht]

Hi Cold Fusion, It looks to me that you don't have any problems just possibilities.

First, reread Ger21's post carefully, I think that his idea about using the smallest router bit and routing out the larger holes is a very good way to do this.

You should not have to do any hand work on these parts other than maybe the countersink. I am assuming that they are 3/8" or less in total thickness and you can route the whole thing with a 1/8" diameter cutter.

1. Your pockets are easily cut, just set the Z height correctly and step the cutter over many times.

2. Switching out a drill and router bit. You may want to take a look at www.precisecut.com. They have bits with collars on them. That way when you exchange bits they allways go into the collet the same distance. This is vital for your pocketing. You need to know where your Z depth is located.

I noticed from various postings that you want to make your own machine. One advantage of this is that you can make it the way you want. There is no law saying that you can only have one router in the Z axis. I think that MACH 2 can handle at least 6 axises.

Make two Z axises and mount them right next to each other. One would be the drill and one would be the router. Under program control you can have your cake and eat it too. Let's say that you have 6 inches between the two centerlines of the Z axis's, then you just have to make sure to add an additional 6" to the appropriate X or Y axis. When programing, you just use the fixture offset to set the 6" offset.

This would give you a true production machine for only a couple of hundred more dollars.

3. Can the G code make 6-8 parts per sheet.? No problem! You have to look at this differently. How big is your sheet? Can get that big of a sheet on your router?

You show 4 parts in your photo. Is this the entire variety of production pieces? and are they always produced in these quantities? Meaning kits of one each?

Depending upon your answers to these questions will determine how you go about nesting the parts on a sheet.

Let's assume that you only make these 4 parts, but they are not related. I would then take the geometry for one of them (in CAD) and lay it out on your sheet. Copy this and put as many as you can on the sheet.

Save this whole thing as a DXF and it becomes the program. The router or G code doesn't know or care if this is one part or 10,000 parts. Its totally up to you as how you want to make them.

You said that you cannot afford a fancy CAM machine. Again, no problem. http://www.artofcnc.ca/Mach1.html

He states that MACH1 has a direct DXF import feature. I don't know how it works, but it might just be a matter of calling out layer names in your DXF file.

Another cheap option is to simply write the G code by hand. Your parts are very simple and if you learn the G code you can simply type the instructions in NOTEPAD and save the file as a TXT file.

This method is very error prone and I wouldn't want to do it all the time, but the cost is right.

Having someone else right the code might work, but I think that you both would have to be pretty good at reading each other's mind. Either that or your directions to him would have to be so specific and your understanding of your needs so good as to give clear directions that you will have the skills to do this yourself.

T

[Cold Fusion]

First, I am building my own cnc router. A couple of people have told me to start small on my first machine. That's good advice for a casual hobbiest who might not finish a complicated machine. But I view this as something I have to have. I spend so much on machining these parts that if I had started earlier I could have payed for the whole router. Here are the the specs I want right now:

All Aluminum gauntry and frame
1.25 Thomson Linear Rails and Bearings
Porter Cable Router
250oz or higher steppers
7amp 3 axis controller (maybe a 4 axis 6 amp)
6 inch aluminum z axis
42x24x workbed


With this size workbed I could get about 13 chassis per sheet.



Is the only advantage of dual Z axis speed? I don't really care if if it takes 20 minutes longer. After all, it will be in my garage and probably be working only 10 hours a month.

[balsaman]

How are you doing the pockets right now?

Those parts will be easy to program. Looks like simple 2D drawings are all you need, then convert with Ace Converter.

I would use a 1/8" endmill and have it cut the holes to whatever diameter you want. No need to redrill.

Turbocnc does support tool changes if you must change tools. It's best (a must) to have fixed tool lengths then tho.

Mach2 is better. It is the latest release.

Turbocnc version 4 is almost ready and will be VERY good. I am a beta tester. Cheaper too.

Eric

[steveald]

Hi,

In regards to changing from drill bit to router bit...why? I just use a 1/8" endmill for holes smaller than 1/4". For plastic, you could use one 1/8" bit for everything except countersinking (you were going to do that on a drill press anyway) so no toolchange required. I think most controller software will import DXF, just remember that you need to create a DXF toolpath, not a DXF outline of the part. (use CAD to offset the part outline by 0.0625" for a 1/8" bit, and save the offset lines to import into controller) It looks like you want to cut the same part many times, so it won't be a real time waster to spend the extra time to create the G-Code, you could even do it manually and use something like the Flashcut CNC demo program to verify it. You can even use the Flashcut demo to create G-Code from DXF, edit it and verify (you just can't control your machine with it without the signal generator.) This route might violate the Flashcut demo software license, you'll have to look int that yourself, but my point is there are ways to avoid purchasing expensive CAM software. My parents bought a house for less than a MasterCAM license, it was a long time ago mind you

Good luck, and Happy New Year,

Steve

[ger21]

I think the best thing to do is spend more time reading and learning how how the different controllers and software packages work. Learn how to write g-code manually, so that you understand what's going on when you're parts are'nt being cut write. Download the Mach 2 manual from http://www.artofcnc.ca and read every page in it. Then read it again. It's VERY informative. Once you have a good handle on the processes involved, download the free CAD software I posted here:

http://www.arch.adelaide.edu.au/~twyeld/ftp/intellicad/

Then learn how to draw the toolpaths you're going to need for you're parts and use ACE, or the Flashcut DEmo, or whatever else is out there to write your g-code.

If you spend some time up front learning exactly how everything works, it will save you from making a lot of scrap when you're machine is done.

Gerry

[Cold Fusion]

The manual doesn't work. Could you email it?

[HuFlungDung]

Cold Fusion,

Yes, don't spend a fortune on software to cut parts like those. Bobcad (or equivalent) would suffice, and is not free, but at least it has the basics of a cadcam system. It will help get you on the right footing to branch out into other fancier stuff later on, should you need to.

I've never tried the route that a lot of the guys are recommending with dxf file convertors, but you'd still need to edit in a lot of stuff to make a program (I guess). At least something like Bobcad has the toolup/tooldown, and various other gcode macros handy to insert.

[ger21]

Did you try to download it here?:

http://www.artofcnc.ca/M2MA2-7.pdf

Gerry

[ger21]

But isn't Bobcad still like $500 or so?

I've never tried the route that a lot of the guys are recommending with dxf file convertors, but you'd still need to edit in a lot of stuff to make a program (I guess).

Cutting wood and plastics like this, all that's usually needed are some G1's, 2's and 3's. Not all that other crap you metal guys use.

With a good converter, you shouldn't have to edit the resulting files. Problem is, good converters aren't usually the free ones. However, with a little practice, you should be able to create .dxf's that ACE can convert with NO editing needed, except adding feedrates, for the parts that you want to cut. I'm sure there are a lot of people here that use ACE on a daily basis that don't do any editing of their files. It does have a lot of limitations (that's why I'm writing my own AutoCAD exporter), but for what you want to do, you should have no problems once you take the time to learn to use what's available.

Gerry

[Scott_bob]

Cold Fusion,

I use a great little program that is easy to use, not expensive, download it, use for free for 30 days, then for <$35.00 you get a licence (if you like it).

http://www.macroexpress.com/

I imagine you could get the G code needed from some these guys ideas from your autocad.dxf options...
Then you could use an editor to put together a complete program like huflung was talkin about.

But here's where the macro utility comes in:
With this program you can automate any windows task into playable macros.
This has saved me hundreds of hours every year. Made my job more interesting. Eliminated human errors, yes I make um. But the computor never does! I use nearly 100 macros. It's not Visual Basic but it's not as difficult to use either.

Try it, then let us know what you think?
Be warned though, I'm known at work as the macro mainiak, it could happen to you too.

Automation pays!
Scott_bob

[HuFlungDung]

Originally posted by ger21
But isn't Bobcad still like $500 or so? ~snip~

Gerry

Heaven's Gerry, that would be list price for version 17. Never buy Bobcad at list price

BTW, good luck with your convertor program

[ger21]

So your saying grab an older version? What do those usually run?

Gerry

[balsaman]

Send me your dxf file, I will make a gcode for you in no time.

Eric

[buscht]

Cold Fusion, I see you have lots of replies concerning low cost G code generators so I won't add to that.

Let's talk about your goals. Excuse me if I am generic in some areas because I don't know all the specifics.

1. You are wanting to produce a product, this is not a hobby application. Even though its low volume it has to be profitable.

2. You need a reliable CNC machine sized properly with the appropriate tooling.

3. You need a way to create the part geometry and tool paths easily.

I agree with you comments about how you want to go about sizing the machine.

But what I see as the important point when making a product instead of something for your own use is the cost of your time.

I am not talking about the router cut time, but your time spend putting the material on the table, putting bits into the collets, countersinking holes, deburring parts, cleaning up, and about a million other things.

Since, you already stated that your total production volume would be about 10 hours per month on the machine, that tells me that if the machine time was increased it wouldn't be a problem.

This is a key point that people get hung up on and don't understand.

As long as your router can keep up with your production requirements, it doesn't matter if it takes 10 seconds or 10 hours to complete the parts. (Only if you aren't standing there watching it.)

I recommend you analyze how much time you are going to spend on the manual processes. This will tell you what is time consuming and what isn't and where you might want to add a feature to the machine or not.

For example, if countersinking the holes takes you 20 minutes of manual labor, it might be worth adding a countersink tool on the machine.

Or if you use my theory about router time not being important (Again, the main point here is that you have enough time to keep up with production and the machine will cut unattended.) You can generate the countersinks with the 1/8" router bit that cuts the hole. You simply have to move the bit up in Z while simultaneously increasing the circles that you cut with the bit.

You can keep the steps small and you would get a very nice countersink.

Back to your question about the second Z axis and what would the benefit be. There is only one benefit in your application and that would be to save your time (manual labor). I would consider a second Z axis as a last resort. It makes the machine and programming more involved. But if you absolutelly cannot make the product without two tools, its an option.

Figure about how long it takes to swap out cutters. Let's say 3 minutes. You have to do this twice because after the program is over, you have to swap back to the original cutter. Unless you write two programs with the drilling and cutting sequence reversed. Anyway let's say 6 minutes. Add in time to leave your other work (or TV) to go over to the machine and back, say another 6 minutes. So we have 12 minutes per sheet, 6 parts per sheet. That means you have to add 2 minutes per part for the tool change. If your shop rate is $30 per hour then you have added $1.00 per part in cost.

Sometimes, two Z axis are used simultaneously to get two parts cut at a time, but this is for high volume production and isn't appropriate here.

So with all that being said and reading the recommendations from the other posters, you should look really hard at one Z axis and see if you can get a 1/8" router bit to make your components complete 100%.

You never said if you have milling requirements on both side of the parts or if its all top down. This also has to be taken into consideration.

Another issue you have to consider is how you are going to hold the parts in place on the table. Keeping in mind that this is production, a vacuum table would be the best. Custom made clamps would be a close second. Otherwise you are going to be wasting time drilling and shooting screws into your router tabletop.

As with all free advice, listen, learn, and do what you want.
T