Pictures of G-Code And Generator

[ger21]

You can use both inches and mm's by using G20/G21, but it's better to have separate profiles setup for each.

[routalot]

I would never argue that not understanding Gcode is a good thing,but I would suggest that typing it or using an online service to generate some of it isn't really an effective use of time.Particularly when free programs to generate the machine instructions are readily available.I saw the mention of using 18 year old software and thought of a friend who still types in Gcode for a controller that uses DOS.It works for the fairly simple parts he makes and the long days of turning out umpteen thousand parts.In that time he can program and refine for the next part if it is something that hasn't been done before and stored on a floppy (!).

I noticed the reference to single blocking a program for the time taken to run the first part.It works for some people but if I had half a million lines to run I would be very old before the job had come off the machine.A simulation takes a few minutes and can be sped up or slowed down to suit the needs.How long does it take to program ramp entries reliably?How easy to add tabs?We use machines because it is easier than using saws,chisels,files and other hand tools so I don't find it inconsistent to use an electronic machine to produce a toolpath in a shorter time and with a stronger probability of success.

[cubby]

Ger, And All you guys that helped me, I THANK YOU ALL, Great group of guys, I did what Ger told me in the IJ settings and tried another cut,
What was weird is I could NOT SEE anything in the Mach3 toolpath window, But the Tool traveled to the correct coordinates and cut the perfect 2 inch circle, That's all I care about is trying to learn how to do the simple things moving on to more things as I learn, It is hard to teach an old DOG new tricks, Some one mentioned CAD/CAM Software, I have Fusion 360, Also Rhino 6. And of course Lazycam, I have read till my eyes are sore, But I learned long ago, There are NO Dumb Questions if you DON'T Know, Only a DUMB MISTAKE if you DON'T ask the Question, I will Post some Pictures as this build begins to come together,
Thanks Again fellers, I feel I have been taught some basic Things about the G-Code,
cubs

[the_gentlegiant]

Great Cubs! I'm sorry I kept filling your thread. And here's the worst of it. Glad you have it all figured out. And learned some things!

--------------------------------

"How long does it take to program ramp entries reliably?"

Not very.

How easy to add tabs?

Tabs? Who use tabs except woodworkers on vac tables and maybe I don't know... knife makers??

"We use machines because it is easier than using saws,chisels,files and other hand tools"

I totally agree.

"I don't find it inconsistent to use an electronic machine to produce a toolpath in a shorter time and with a stronger probability of success."

Good sentence. Me neither.

Hello routalot,

I think you're misinterpreting the use of older software. (BobCAD V20) I do all programming in 2D, and don't manually type in a single coordinate or movement G code at any time. I do create some tool paths (mostly the helical ones) in a CAM assisted semi-manual workflow. Even HSM tool paths. (These take a little but of time all depending.) My software has great functionality for producing bores and bosses and what have you, with controllable lead ins and outs like typically needed for adding cutter comp. Creating offset paths is super easy too. At most I may type in a simple large face mill flattening or cleanup path at the beginning of a program. Just like you I can highlight drill locations or part outlines or created tool paths, click one button and have hundreds or thousands of lines of code in a second. The only thing I'll then do manually with that is perhaps control the startup of all that code, and fill in the comp calls, start up Look Ahead or configure drill cycles. Maybe set as a subroutine or control the Z height just as examples. The ONLY difference is I'm doing it in the actual machine G code program. You and others do the very same thing in windows and pull down menus. Also what I do manually is writing in multiple subroutines or repeat routines into the main controlling program, or controlling multiple work offsets. I'm seldom programming for one part but for small to medium production runs, sometimes with many tools. A lot of it for a pallet machine where typically running 16 or more offsets and maybe the first op on one pallet followed by a 2nd op on the other.

Much of my programming is for a fast, high horsepower 40 taper running up to 3/4" carbide and large insert drills as fast as I can run them. Handling this beast of a machine is something to contend with. The simulation portion of both of my software versions mine as well be non existent for how bad they are. So me single blocking is two fold. One is I haven't seen the tool movements prove out on the machine. Also some of these drills and milling routines are removing a lot of material quickly. A person can use experience to decide what will fly and what won't, but with this much power and torque to deal with, it would take nerves of steel to push the button and let it go first time through on every tool. Let's say a big 1 3/4" insert drill is hurling itself to within 0.100" of the top of the part at 1260 ipm, and then using lots of horsepower drilling through a few inches of steel or aluminum. Trust me, until you've seen the machine do it at least once, you're not going push the green button and walk away or let the program run it's course without complete, albeit single block control over the situation. When I say single blocking, that's mostly starting up tools for the first time. I do not single block thru a whole program. I might run some tool paths above the part for verification of motion and get a sense of the feed rate written. All these things are typical I think for a lot of people. I may also single block the first time through between certain offset or subroutine changes. I write super clean programs with little to no wasted movement. A lot of that is keeping the Z axis down as much as possible, even between vise and offset changes. Also there may be a fair amount of rapid movements deep into bores. So some of those situations I may single block the first time through. That or at least hold for a second (using single block) and re-read at the control what's about to happen. I'm almost certain there's not a post processor out there that can write highly controlled multiple subroutine, work offset and Z height controlled programs like I do. I've seen numerous post processor posts on this and other forums that are junky as hell. I'm sure they're not indicative of high end output. But still crap is mostly what I've seen.

I mentioned earlier that I do not do any 5 axis type 3D contour work. So millions of lines of code is not seen here. Some of my longest programs have been for engravings of logos into plastic. That type of programming has few fears and is not prone to errors. So my longest programs have had the least amount of oversight. Odd but true.

Honestly I really do wish I had simulation. After using it for awhile and gaining trust in it, I would be happy to bypass controlling or single blocking certain parts of programs.

I also wish I had 20-40K to invest in true high end CAD/CAM software with simulation. I'm not so stupid as not to realize that some parts of my programming and at the control processes could be sped up, (not necessarily improved output) by some modern, high end software help.

Okay this turned out to be a bomber. My thanks, or is that apologies, to the few that may read all the way through. I've got to quit screwing around here and get back to work!

[handlewanker]

Wouldn't it be cool to have a 3D simulation video that you could run on your PC using the program you wrote or got with a CAM compiler.

It would appear as a virtual anime milling machine with a job in a vice etc and the simulated cutter cutting away at the work piece just like a real machine did......some of us like to watch machines work so why not a video that you control by writing the program then running it on a simulated milling machine?

If you made a mess of the G code sequence Mach3 would soon tell you where it went wrong, and after the piece was finished you could view it in 3D and turn it around like you do with Fusion 360...…….also you could measure the piece to see if you applied the various offsets to get the sizes you anticipated.

When all things are good you could if you wanted to then run the program on a regular mill and be happy, but mainly the outcome would be the learning curve for G code compiling and learning.

I do that now for learning practice on my CNC mill by writing the code for a hypothetical work piece and running it with a wooden stick dummy cutter in the chuck and a Polyfoam dummy work piece in the vice.....nothing gets dented that way and there's no messy clean up afterwards.
Ian.

[routalot]

I suggest some of you fellows might take a look at Camotics for simulation.Its free but somewhat slow and it does help.I could allay some of the fears of the-gentlegiant about the cost of CAM software with simulation,I use Freecad's Path workbench in my home workshop it costs absolutely nothing and simulates toolpaths quite well,if a little slowly.It takes a little getting used to but does a great job,particularly if you spend an hour or two watching the tutorial videos on youtube.It is a little short of post processors for industrial controls as most of the development has been done by hobbyists.Good post processors for top level systems are out there but they may cost quite a bit at first glance,you recoup the cost with the efficiencies they generate.

[cyclestart]

edit/ This old man wasn't thinking lol. Camotics expects G20 even if file is imported as imperial. This concept came up a few posts back. Camotics works as expected if G20 is added.

I suggest some of you fellows might take a look at Camotics for simulation

Camotics is great for back plotting files like depthmap gcode, I often use it to check output of dmap2gcode.
For those using imperial measurement it can produce strange results under some conditions.
An example being the gcode that started this thread:
G00 Z0.1
X5 Y10
G02 I1 F100
G00 z.1
M02
results in strange simulation if imported as imperial units.

This problem may be specific to my install of Camotics which is a bit old (v1.1.1).
Mentioning it just in case Cubs tests this code with Camotics.

[cubby]

I suggest some of you fellows might take a look at Camotics for simulation.Its free but somewhat slow and it does help.I could allay some of the fears of the-gentlegiant about the cost of CAM software with simulation,I use Freecad's Path workbench in my home workshop it costs absolutely nothing and simulates toolpaths quite well,if a little slowly.It takes a little getting used to but does a great job,particularly if you spend an hour or two watching the tutorial videos on youtube.It is a little short of post processors for industrial controls as most of the development has been done by hobbyists.Good post processors for top level systems are out there but they may cost quite a bit at first glance,you recoup the cost with the efficiencies they generate.

Is this Camotics any better than Just running the Simulation in Mach3?
I been just testing the code in Mach3 before ever loading it to a machine, Even then I will cut air until I'm sure this is what I am looking for,
Of Course I'm Just learning this,
I did complete my very First Motor plate today and cut my first REAL 50mm Hole, Started out nd it burned up n end mill before I could shut her down,
And I figured out how to go into and edit my code to drop my cut from .050 to .025. Slowed the Spindle down some and added some cutting fluid,
And Wha La!! Worked great, That might not be much for you guys, But was an ACCOMPLISHMENT For this old man,, :-)
cubs

[handlewanker]

LOL....now that is what I call having a great day when you make or "improve" on a G code sequence to do what you want...…...that is an achievement by no stretch of the imagination...…….like learning to ride a bike for the first time.

BTW.....I'm a noob at CNC but I'm learning as I go....the CNC cookbook entries by Bob Warfield has been a huge help in my fathoming out what the code does...….. and when I write a code sequence for a dummy workpiece I run it with a wooden dowel in the chuck to see if it does touch down on the top of the work and do all the other things I coded etc.
Ian.

[cyclestart]

The back plot in Mach3 will work fine.
Camotics is like the back plot and simulation found in CAM software (and it also does some other interesting things)
except:
afaik it can't single block lines of code
afaik it doesn't display the g-code text file while tool path is being drawn.
There is no way to select a drawn line in the tool path display and highlight the line of gcode text that produced that move.
I'm not familiar with the back plot in Mach3 but would wager it does all the above.

What Camotics does do it does well but it's not the right tool for every job.
Be aware the developer models Camotics on linuxcnc g-codes but Mach3 g-codes will be very similar.
Sounds like you're making headway and finding the right fun-2-frustration ratio. Attitude is half the battle most times.

[Kummin]

Routalot - I'm new to the game. Have been considering Fusion 360, have heard mostly good things about it, only thing I don't care for is its reliance on 'the cloud'. Anyway, I had not heard of Freecad or Path before. Do you do your design work in Freecad, then generate the tool paths with Path? Thanx.

[routalot]

I design in Freecad and then use the Path toolbench to create the toolpath.It has some great benefits and one or two quirks.If you are used to creating lines in 2D it might be a bit of a stretch to convert to parametric 3D modelling and the documentation doesn't progress as quickly as the developers can create new features.The big benefit is that if you modify one or two of the features using the model tree,a click on the recompute feature sees the model change shape and the toolpath lines reflect the different trajectories.

There are some great tutorials on youtube and I highly recommend watching sliptonic's series on using the program-he literally wrote the book on Freecad.The forum at https://forum.freecadweb.org/ contains a lot of good information too.Best part is the files can be stored anywhere you find it convenient and there won't be a subscription model.There seems to be some consternation in some quarters about forthcoming Autodesk terms of business which won't be happening here.

[handlewanker]

I think there is a real need for free high end CAD/CAM programs that are past their use by date...….by that I mean when a program like Fusion360....whatever...…. first comes out it is at cost to use it but as the next "improvement" arrives the last version should automatically slip down the cost ladder.

Eventually by version 20 or so the cost for everyday DIY hobbyists.....who want to use it to make a small amount of money commercially...... would be so cheap it is almost give away.

Imagine a million DIY'ers paying a dollar a month on a yearly subscription to use the earlier versions that are still quite powerful but not as full of the bells and whistles that industry users or a heavy DIY home worker would need or go for.

That would be 12 million smackeroos yearly for something like old rope.

As the program strata rose in the versions the cost across the structure would also be pro rata......you could update to a newer version as time goes by when you need more oomph.

I don't deny anyone a fair bit of the cake but if it gets too pricey you start to go off the boil quick.

I like Fusion 360 and have a couple of tutorial books and UTUBE videos too to do the necessary and it being free for hobbyists is good as well.

I have two versions of Turbocad...one version 16 the other version 21...….trying to sell them on now is almost at give away for free to a good home.

As I have the discs in hand I might once again have a look at it to see how user friendly it is...…..Fusion would be more useful as it has a CAM facility too.
Ian.

[routalot]

I think there is a real need for free high end CAD/CAM programs that are past their use by date...….by that I mean when a program like Fusion360....whatever...…. first comes out it is at cost to use it but as the next "improvement" arrives the last version should automatically slip down the cost ladder.

Eventually by version 20 or so the cost for everyday DIY hobbyists.....who want to use it to make a small amount of money commercially...... would be so cheap it is almost give away.

Imagine a million DIY'ers paying a dollar a month on a yearly subscription to use the earlier versions that are still quite powerful but not as full of the bells and whistles that industry users or a heavy DIY home worker would need or go for.

That would be 12 million smackeroos yearly for something like old rope.

As the program strata rose in the versions the cost across the structure would also be pro rata......you could update to a newer version as time goes by when you need more oomph.

I don't deny anyone a fair bit of the cake but if it gets too pricey you start to go off the boil quick.

I like Fusion 360 and have a couple of tutorial books and UTUBE videos too to do the necessary and it being free for hobbyists is good as well.

I have two versions of Turbocad...one version 16 the other version 21...….trying to sell them on now is almost at give away for free to a good home.

As I have the discs in hand I might once again have a look at it to see how user friendly it is...…..Fusion would be more useful as it has a CAM facility too.
Ian.

You have described a situation that would benefit hobbyists and which would destroy the companies that develop the software.Who would buy the commercial product when much of the same functionality is being given away?Thats always assuming than no piracy goes on.

As it is we get posts on this forum week in and week out asking for help with commercial packages and which are so basic in nature that you have to question why any reseller would skip such basic training.Occasionally such posts come from somebody who has taken up a new job and is having to learn a new package because the previous fellow dropped dead or something but a surprising number of hobbyists seem to have the same high end software and no training.

I know this is the Gcode forum and I might be stretching the hospitality a bit but in reply to the question about using Freecad to create toolpaths I bashed out a quick example and took a screenshot-which should be attached.Keep in mind that the dimensions of any feature can be changed by calling up the relevant item in the model tree and typing in a new value and the model regenerates,together with the toolpath.All at zero cost to the user and legitimately.

[handlewanker]

Yes?......who in a serious manufacturing business would even dream of using an old version of a program that they rely on to earn their crust of bread.....they would be the main buyers of that software because they are the ones who buy the latest software to make money with or go out of business.

A customer who presents a work program to a manufacturer for a quote would not be impressed if the quote produced rubbish due to an old software version being used by the manufacturer..

On the other hand, a hobby person who now uses the free version, because they aren't making money etc, could have access to a more powerful but older version and still be able to afford the upgrade as and when necessary......I have to wonder how many hobbyists are using the free Fusion 360.

As I said, a $ a month for a million hobby users would generate quite a bit of revenue that normally would not be making any.

If a software company relies on the free software for hobby use to fund their expansion into higher versions....that is getting nothing for something they could be earning a million $ a year from.

Supposing Fusion360 was no longer free to use for anyone and cost a $ a month for the older version and rising up the scale to a full version...…...would that mean nobody would be using it?...…..I would think that even a hobby person with access to more powerful software that still did the basic job, would be keen to get their hands on it for such little cost.

BTW, if you are in a competitive business and are really serious about being in the running for money making....you would be buying the latest software or leasing it....whatever...…...that is how the software companies make their money as service providers.
Ian.

[ger21]

So old software works perfectly fine in the hands of hobbyists, but produces garbage in the hands of professionals?

There are plenty of business owners that will always try to spend as little as possible on software, especially when software costs can be $10,000 or more per year.

The bottom line is that selling old software at a discount costs them more in lost sales of their more expensive new software. The goal of all companies is to make money, and I think they know more about what it takes to make money than you do.

[cyclestart]

Mastercam 8 is now 20 years old. Legitimate potential customers for Mastercam's new offerings would choose v8 to save a few bucks/day ? I'm talking seriously geriatric software that is still capable of reliable output. Making v8 available free/cheap doesn't fit CNC Software inc's business model and there's little upside for doing so.

re FreeCad: This project needs a benefactor(s) so developers get some reward for doing the drudge work of debugging, documentation, improving the UI etc. Like Blender or Mozilla or Python etc. As it is now FreeCad can be very frustrating.

[handlewanker]

Well, they wouldn't be selling old software at a discount to a hobbyist who is using the free version because he's not making money etc.....so where is the lost profit or lost sales on freebies?

At the same time the state of the industry is not good if people who want to compete against foreign trade are attempting to do so by "investing" in outdated old software.....where is the sense in that when the Chinese are making inroads on your industries?

Is this part of the "Make America great again" plan?.....LOL, you would have to be pulling my leg.

BTW....on the topic of older versions of top class software, I have Photoshop 4 complete with the manual which I bought from a registered dealer for $40.....Adobe has lost out with that lost sale but they could have sold it to me if it was available from them.

I also have a "copy" of Photoshop 7 and the difference for general purpose work is minimal......that is not to say that if I was in the graphics business the latest version would not give me a better chance to compete against the competition and I would have to be a very poor accountant to consider buying the older version as part of a business plan.

As an amateur photographer Photoshop 4 is totally adequate for my minimal needs so, why go to the top of the ladder and spend mega bucks?
Ian.

[handlewanker]

Mastercam 8 is now 20 years old. Legitimate potential customers for Mastercam's new offerings would choose v8 to save a few bucks/day ? I'm talking seriously geriatric software that is still capable of reliable output. Making v8 available free/cheap doesn't fit CNC Software inc's business model and there's little upside for doing so.

re FreeCad: This project needs a benefactor(s) so developers get some reward for doing the drudge work of debugging, documentation, improving the UI etc. Like Blender or Mozilla or Python etc. As it is now FreeCad can be very frustrating.

Huh?....if Mastercam 8 is still good at version No.1 that it started out at, why spend money on improving it for little gain?.....you may as well just copy a million copies of the original CD program and sit back while the money pours in......no overheads to worry about and no staff to maintain trying to improve the program that is working OK.

That is to say...…..who would buy into a new version for a couple of grand when the older version you bought 20 years ago was still completely useable?
Ian.

[cyclestart]

Huh?....if Mastercam 8 is still good at version No.1 that it started out at, why spend money on improving it for little gain?.....you may as well just copy a million copies of the original CD program and sit back while the money pours in......no overheads to worry about and no staff to maintain trying to improve the program that is working OK.

That is to say...…..who would buy into a new version for a couple of grand when the older version you bought 20 years ago was still completely useable?
Ian.

Huh...what? "still capable of reliable output" is high praise ? Yes it produces reliable toolpaths. It's also 20 year old software written for 20 year old computer hardware running w2000 and later XP. They release software on a 2 year cycle, or something near that.
The CAD included (Mastercam Design) wasn't cutting edge even 20years ago. At it's heart it's DOS software modified to run on Windows. Still I'd love to have it on hand in case Autodesk pulls the rug from underneath us free license Fusion users. I took Mastercam training back than and still have the manual. I don't remember saying a Fortune 500 company would want this software, in fact I suggested the opposite. Maybe you missed the question mark in that sentence.

Mastercam considers themselves a prestige company with the support, maintenance, training and customer service that goes with that. I believe they would rather shred their old software than sell it to someone like me for a what I'd be willing to pay.
btw "a couple of grand" will buy something from a Mastercam dealer but it won't buy any software.