[byoung0454]

this is the guid rail system I am looking to do.

[byoung0454]

yes, I understand, and I feel your pain. I have had no less than 6 (or more) plasma cutters so far in my life, and I got into this cnc thing hoping to retrofit one.
Great idea, but I just couldn't justify spending the money at first. I wanted all my money to go into my table, and I didn't want to spend much at that.
The first build of the table wasn't beautiful, but I changed and modified it as time went on, and I have two larger tables in the works already.

If you want to save the money, oxyfuel is a viable option. It does cut slower than plasma, and it does and can warp material. Getting cutting speeds and tip sizes, gas pressure, preheat times, etc. takes a bit of practice, but in the end, it is well worth it. After listening to what you want to do, I think it will suit you fine. You can always try and make a little extra money with it so you can purchase a small plasma down the line, should you come to desire one.

If I were doing this all over again, I would consider
www.xylotex.com driver packages, the 280 ozin motors will be fine with some good gear reduction with belts and pulleys.
or www.Geckodrive.com, the G210 is top notch stuff. I love mine.
With that, www.kelinginc.net has some great motors and power supplies.

Any good CAD program, I use Dcad because the demo is free, and it is easy to use. There is a lot of others out there, look around.www.dcad.com

CAM. Well, I don't think (for low budget flame cutting) that I should bother mentioning anything besides www.sheetcam.com it is inexpensive, and designed for flame and plasma cutting. Les (the designer) has a free demo you can try before you buy. There are ways to get around using a CAM program such as coding by hand, but that can get to be a pain. CAM is one of those things that I never knew how much I wanted it until I got it, then I didn't want to go on without it.

Controller, well, you can go the really nice and professional route of Mach3, www.artofcnc.com
run under DOS like AccurateMike here by using Turbocnc www.dakeng.com (it doesn't show you what the part looks like since it runs under DOS, but it is inexpensive)
I personally use Quickstep4 with a pulse pacer,http://cpwojcik.home.att.net/Qstep/Quickstep_CNCx.html
which is inexpensive and really easy to use.

There are a variety of other programs out there that are good controllers, but those are three off the top of my head.

Does all that help? Or is it a bit confusing?

I think I am catching on, I looked at the quickstep controller program and I am hooked on that, it looks really easey to pick up on. Would I need the pulse pacer? from what I gather it is to gain speed with your stepper motors, but for the setup I plan on building I don't think I need high speed. But then again what do I know im am new to all of this. Sheet cam their is no question about that, I looked at Les's program befor and like you said it is made for my kind of work.

I think I am going to go with exlotex for my driver package, one thing I am not sure of is what do I have to do to set up for 2 drive steppers for my x.
For now that were I am at but I will have more question to ask that is for sure.

Thanks again

[massajamesb]

Well, the pulse pacer takes all the load off of your computers processor for step pulse timing.
Your processor inside the computer is what generates the step pulse timing for the drivers. Well, when you try and process more than a certain amount of step pulses, the processor can no longer keep up, (especially since Windows is always doing stuff in the background you don't know about) and you will "miss steps", sometimes rather vilolently. It will not be able to process the steps right,and your motor will stop spinning for a second, and you will no longer be in the position you are supposed to be. Your parts will not come out the way you want.
The pulse pacer has its own timer for step pulse timing, so you are no longer depending on the computer processor to keep timing right. It already has enough to keep up with, this helps to relieve some of its burden.
If you are running microstep drivers and gear reduction on the motors, then you will have to run more steps per inch, and you will have to run a higher step rate.
I used the program without the pulse pacer for a while, and didn't have too many problems with it missing steps, unless I was jogging or running at high speeds. The issue there was more my crappy PC than the program, I was running Windows 98 with a 500 Mhz processor speed.

I recommend the pulse pacer, but it is not neccesary. If nothing else, try the software without it, and if you have the extra money later, get it.
I don't think you will have any problems without it, but you never know....

It is not neccesarily there to gain speed, but smoother and more constant step timing, which does generally result in a higher top end speed.
Does that make sense?

If you want to tie the two X drives together, you would either modify your input DB25 cable to have the same step and direction wires going to those two drives, or if you ran Mach III, there is a axis slave feature that would work even better.

[byoung0454]

Well, the pulse pacer takes all the load off of your computers processor for step pulse timing.
Your processor inside the computer is what generates the step pulse timing for the drivers. Well, when you try and process more than a certain amount of step pulses, the processor can no longer keep up, (especially since Windows is always doing stuff in the background you don't know about) and you will "miss steps", sometimes rather vilolently. It will not be able to process the steps right,and your motor will stop spinning for a second, and you will no longer be in the position you are supposed to be. Your parts will not come out the way you want.
The pulse pacer has its own timer for step pulse timing, so you are no longer depending on the computer processor to keep timing right. It already has enough to keep up with, this helps to relieve some of its burden.
If you are running microstep drivers and gear reduction on the motors, then you will have to run more steps per inch, and you will have to run a higher step rate.
I used the program without the pulse pacer for a while, and didn't have too many problems with it missing steps, unless I was jogging or running at high speeds. The issue there was more my crappy PC than the program, I was running Windows 98 with a 500 Mhz processor speed.

I recommend the pulse pacer, but it is not neccesary. If nothing else, try the software without it, and if you have the extra money later, get it.
I don't think you will have any problems without it, but you never know....

It is not neccesarily there to gain speed, but smoother and more constant step timing, which does generally result in a higher top end speed.
Does that make sense?

If you want to tie the two X drives together, you would either modify your input DB25 cable to have the same step and direction wires going to those two drives, or if you ran Mach III, there is a axis slave feature that would work even better.

If you want to tie the two X drives together, you would either modify your input DB25 cable to have the same step and direction wires going to those two drives, or if you ran Mach III, there is a axis slave feature that would work even better.

So whay would your oppion be, go with machIII or go with quickstep? Would one be better than the other?

[massajamesb]

Comparing the two controllers is much like comparing a Geo Metro and a Corvette. They both will get you where you are going, but one will look better doing it. One is merely a inexpensive means of doing what the industry standard one does.

There are a number of cnc users out there who use programs like Turbocnc, Kcam, etc. because a less expensive program suited them and their needs better.

Mach is a really nice full featured controller, it is probably the nicest controller out there. It does untold numbers of things that Quickstep doesn't.

I suggest Mach if you are wanting the best of the best, and a full featured controller that has all kinds of features you will probably never use.
There are a number of plug in devices for Mach like ATHC's, pendants, etc. that will only work with Mach. Quickstep does not have this kind of outside support, it is merely an inexpensive hobby program. It has limitations that Mach does not.
Mach also has a couple of features that would directly benefit someone who burns a lot with oxyfuel, such as "cut from here" option that allows you to restart your cut file on any Gcode line, in case of blow outs or losing cut.

All I can give you is my opinion, and that is Mach is the nicer controller, it does more, it "looks" nicer, and supports just about any kind of cnc ANYTHING you could imagine.
Even though Mach is nicer, and has all the bells and whistles, I still use Quickstep. It just fit what I wanted to do at first for less money.
They both have their place, and they both work.

I use quickstep because it :
Is inexpensive
is very easy to use
runs on a old crappy computer
works for me.

Will I use Mach on my next cnc plasma? Probably, because I want to use a Candcnc ATHC, and it only works with Mach.

IMHO, if you just want to burn a few parts here and there, you don't want all the custom screens, all the extra stuff, pick up Quickstep and give it a try. You might contact the manufacturer and see if he has a full feature demo available.
Mach has a full feature demo on their website, it works with parts up to 1000 lines of code. Try it as well.

again, IMO, Quickstep might be more what you are looking for, but I could be wrong. That is known to happen sometimes
Give them both a resonable chance, don't discount one or the other because of price. There are a lot of really happy Mach users, and I have heard no real complaints about Quickstep either.

[accuratemike]

One of the good things about TurboCNC is that it runs under DOS. With no multitasking, DOS has full command of the processor and generating steps. Step generators are for Windows based software, solely to eliminate the evil multitasking errors. When I did mine, TurboCNC was a free download that you could try and ~$60 to buy. I haven't checked in years, but I suspect it is still along these lines. And, with DOS, any old POS processor (even 386's!) will generate more than enough pulse speed, way more than enough. I use the free FlashCut demo to view tool path (see screen shot I posted earlier). There are some codes for TCNC that FlashCut doesn't recognize, so I get the toolpath straight and then add the code for pauses etc. before I run it. (After adding this stuff FlashCut balks on the extraneous code, no biggie really. Just an order to do things) For me it goes: Draw the part in AutoCAD-export .dxf, Convert to g-code using FlashCut demo and play with tool path, Add code for torch on-off and preheat etc. in NotePad, Change the file extension to .cnc, Run it in TCNC. Not as hard as it sounds and all free or cheap. I have two machines. One is screw driven (direct mounted steppers) and rapids around 100ipm. This is the 30" machine and I don't really need the speed for the torch (usually around 30ipm). It works very well. The other is 4'x6', rack driven (4:1 reduction) and rapids as fast as 1200ipm. I built it for a friend to use as a plasma machine and bought it for chump change when he lost interest. I have slowed it down to 6-700ipm because 1200ipm can be scary (and un-necessary). The little Y gantry is ~70#, the big must be more than 100#. Both are single motor on the X. The little has 187 oz motors, the big is powered by ~900 oz motors. I don't think you really need a dual drive X unless racking on the guides is a problem. I did the big motors on the big one because I might put them on my bench mill someday and the machine is a router right now (I knew I might get it someday ). As a torch, 900oz is WAY overkill. There is no tool load to overcome. They did give me unreal accelleration and rapids though (fun to show your friends,100+# going 0-1200ipm in .2sec and an inch or so is COOL. Too much fun and you'll break a gear or strip it off the shaft though). The little is driven by Xylotex, the big by Gecko's. And, since my Plain Jane hand torch will cut 5" steel (I believe it, it goes through 1" like butter), a 3-hose system may not be necessary either. I just use a solenoid triggered air cyl. to push the torch's button. No other valves. I have used a trim router and Z slide on the little one too. It works, so you really don't need a ton of power to push 70# around, even with some tool load. Don't be drawn in to over-engineering the thing, it is amazing how far simplicity will take you. You will have plenty of effort involved in making the machine, even a simple one, if you plan on doing it well. Mine has had ZERO adjustments to the design since I started using it. I have knocked out a ton (maybe literally) of cool stuff with it. I guess it has been at least a couple of years. I wish I had made it at least 4' wide in one direction, I have to cut sheets to get them in it. My dimensions were based on a 6' acme screw, cut in half. I'd have spent the few more beans for more rod had I thought about it.
I somehow got un-subscribed from this thread so I missed some of the exchange. I'm re-subscribing this time so I can watch your progress, POST PICTURES. Good luck, MIKE

[massajamesb]

Yes, Turbocnc is still that same way, and is a very viable option. There are posts in Sheetcam for Turbocnc, so it will generate the code that Turbo will recognize.
As far as that goes, it just depends on how much you really want to see the design being cut on screen. For me, the price was about the same either way.

I agree with your advice. If I were just building a small torch table, I would just use a hand torch. Only thing I would do is make the torch height adjustable somehow.
I wish I had just built my first table the way I wanted it to begin with, but it was a good learning process. I kept changing and changing until I finally decided to leave it alone and just build another.
Maybe one of these days I will actually add some paint or the sheetmetal sides and ducting I have built for it, who knows?
I am envious of your machine, Mike, always have been. Shoot, I may build a small torch table just (mostly) like it to have around for the occasional thick stuff I burn.

[accuratemike]

I deal with torch height from the bottom. It's no production machine so I can deal with clamping the work piece to bars that I raise and lower with shims. From 1/8-~3/8, it goes right on the bed, maybe shimming up the thinnest stuff. Heavy stuff I go bars/shims/clamps and shim down as necessary. You can see some of that in the Rest pics. Once I arrived at the optimal distance for each tip, I just leave the shims I used around the machine. I've thunk up a screw raiser/lowerer, I just don't use it enough to expend the effort for the convienence. Torch height control is kind of moot for me. I may have had one job curl up enough to affect the cut. Mostly I clamp or tack anyhow.
I enclosed the bottom of my machine with my stock of plate. I buy drops from the surplus guys. I store them by leaning them on the legs, making a poor man's enclosure. Keeps the slag in one spot anyhow. No water table, no ventilation (machine is by the overhead door, in the summer I'll put a fan nearby to direct the smoke out the door). Once I started containg the spray of sparks, the shop doesn't get that dirty. The dirt flying every which way from grinding was 100x worse anyhow. (Love the clean cuts)
I'm not denegrating anyone's advise, just throwing my experiences into the mix. When I started mine I mostly got "It'll never work". (Too slow rapids, can't use a hand torch, screws over the top won't work, screws at all won't work, ball transfers won't work, can't push a torch button with air using code, torch tables in general are too dangerous for the home shop, plasma is the ONLY way to go, gotta have a water table, cheap/free software sucks, machine HAS to be BIG, yada yada. All saged advice that I ignored. I'm so satisfied with the machine I could pop. Good thing I have my own brain) At least byoung0454 isn't getting blown away by a buch of ill advised nay sayers. Enjoy, MIKE

[byoung0454]

What do you guys find as a ideal table height? I have been doing some mock ups with some scraps and think that 30" seem ok but what are your guys built to. My other question is about my gantry, is their a ideal wheight I should stick to? I was thinking about using aluminum 2"x6" .25 wall rec tube, would that hold up or do I need to look at steel for this?

[massajamesb]

Mine is about waist height (around 3'), and I like it there. I can reach all the way across the table if needed, and sheets are pretty easy to load that way.
The next one I build will have the cutting bed at the same height as my trucks tailgate when it is lowered, so I can slide sheets straight out.

The lighter your gantry, the easier it is to push around. I wouldn't worry about making it too light, I would say that you want it rigid and strong before you want it light. Just my opinon.
My gantry is two lengths of 3/16x2x3 box tubing stacked on each other, and the gantry is a 3/8 thick piece of 4x6 angle steel. They are heavy, and they move around just fine.

A lot of builders use aluminum, what you have listed will work just fine, IMO.
I just used steel because it is cheaper, and the owner of the scrapyard next to me gave it all to me.

[byoung0454]

well I have changed my mind I am going to go with mach 3, due to the guy that cuts my material now that is what he uses and he said he would help get set up and show me how to operate it. He also told me were to get a 3 hose torch at a good price, it is a esab torch and it is about half the cost of a harris, do any of you guys have any thoughts on this brand.

[2fst4u]

What gear reduction are most people using? I went with a 20 DP 20 deg. rack and was wondering what size of pinion would be best? Also, what size of belt reduction (mxl, xl, l, or ???) and how many teeth for each pulley. I was thinking xl series and around 10 and 35 for 3.5 to 1 reduction.

[accuratemike]

Try to find a torque curve for your steppers. Most steppers make good power 'till around 1000rpm. Some less, some more. You want to do your work with the stepper in the meaty part of the powerband. Rapid speed will end up being what it is. Bigger motors may help with faster rapids as there is more "left over" power as the power band tails off. I just keep pushing it faster 'till it stalls and set my rapid speed below this threshold. Knowing the ipm that you need to cut, gear the system so the stepper is in the meaty part of the powerband. Remember, the gear covers the distance of it's circumference per revolution. I ended up using small gears (~5/8") and 4:1 reduction on the big machine. This gives me good power at cut speeds (does double duty as a router). The 900+oz motors still have enough power at high rpm to allow really fast rapids (up 1200ipm). Do a little math with stepper rpm, gear circumference, reduction ratio and cutting ipm and vary the circumference and ratio 'till the stepper rpm is good. Hope I helped more than confused . MIKE