Tormach 1100 vs. Haas VF3

[philbur]

How do you crash your tool if the tool path looks good in fusion 360 and you set your zero point?

How do I crash thee? Let me count the ways.

Phil

[philbur]

Seems the key to the CNC machine is the stepper/servo motors and the controller ... the rest is just tig welding a bunch of stainless steel or iron.

You need to do much more reading

Phil

[tmarks11]

Oh is the software onboard the tormach that coverts the gcode to the machine internet based or is it hard programed so that the machine can work totally offline?

PATHPILOT is linux based software, no internet connection desired, required, or possible (unless you do a little hacking).

So another question, is it worth it to buy a CNC lathe for cutting ultra fine threads or can the tormach do them just fine (these would be NPT threads and also straight threads but not in high volumes).

You can threadmill with Tormach. CNC lathe for cutting threads is a bit overkill. Single point threading with a manual lathe isn't very hard, CNC is only really called for for volume production. Depends on the geometry of what you are cutting, of course.

How is the tormach 440? Do you know if the tormach can be upgraded to AC brushless servos?

I would buy the 770 instead of the 440. Only slightly more expensive, and much more capable and accurate. Plus the 770 is a tried and true design, 440 is new and has had some growing pains. Actually, I would buy the 1100 (and did). Depends upon your application, wallet, and available space.

"Upgrading to servo"? You are throwing away 1/3 the cost of the machine doing that. It you really want a servo machine, recommend you look at one that was built from the ground up that way. Wander over to the Novakon forum, and read some threads about the Novakon Pulser.

If I were to build a CNC machine do the AC brushless servos require capacitor impedance matching due to the AC nature of the power...

Recommend asking these questions in the BENCHTOP MACHINES forum, which is about DIY machines. The audience in this forum bought their machines complete so they could spend their time using CNC to build things, instead of spending years trying to build a CNC machine that works.

Seems the key to the CNC machine is the stepper/servo motors and the controller ... the rest is just tig welding a bunch of stainless steel or iron.

What makes a CNC mill useful is its accuracy. Which requires precision parts and precision alignment when you build it. Again, go to the bench top machine forum and read some of the 700 page build logs to find out some of the gotchas.

The stepper/servo and controller is the easiest part that you can hook up in an afternoon. It is "just" everything else that takes forever.

[akmetal]

PATHPILOT is linux based software, no internet connection desired, required, or possible (unless you do a little hacking).

Very nice

You can threadmill with Tormach. CNC lathe for cutting threads is a bit overkill. Single point threading with a manual lathe isn't very hard, CNC is only really called for for volume production. Depends on the geometry of what you are cutting, of course.

That is great news, can a regular lathe do 4 ft rod?

I would buy the 770 instead of the 440. Only slightly more expensive, and much more capable and accurate. Plus the 770 is a tried and true design, 440 is new and has had some growing pains. Actually, I would buy the 1100 (and did). Depends upon your application, wallet, and available space.

I could just buy one but wife wants to get out of condo into a single family and I need the garage space too for stuff like this, plus I am putting together a hangar project for my plane ... but you are right it is only a little more

Right now I have more time than money to leverage, house and hangar is just going to work and saving. All my projects are on hold until I can get a proper cnc machine. I can keep studying for my second PE I suppose lol All my other tech projects are on hold until I have a cnc that can at least cut bulk aluminum.

"Upgrading to servo"? You are throwing away 1/3 the cost of the machine doing that. It you really want a servo machine, recommend you look at one that was built from the ground up that way. Wander over to the Novakon forum, and read some threads about the Novakon Pulser.

How leveraging are AC brushless servos? Tormach has a statement that says AC brushless servos are superiour to what they install but are cost prohibitive, I looked up AC brushless servos online and they are not cheap but not so overly expensive that it couldnt be an add on.

Recommend asking these questions in the BENCHTOP MACHINES forum, which is about DIY machines. The audience in this forum bought their machines complete so they could spend their time using CNC to build things, instead of spending years trying to build a CNC machine that works.

I have years but my worry about spending money on raw materials and having the project be a boondoggle and the money is just wasted. But your right it would take years. If I had a local machine shop that was actually willing to take on work I could just get the few things done I need and move on with my projects but I live in a small fickle market. The shops will do the threading but they wont build the complex flanges I have made up in solid works, they "passed" on it and its only a 3 axis design not like an impeller wheel or anything.

What makes a CNC mill useful is its accuracy. Which requires precision parts and precision alignment when you build it. Again, go to the bench top machine forum and read some of the 700 page build logs to find out some of the gotchas.

I will do that thank you - do you have the link to the 700 page document
I really like the lay out of the Xcarve if it just had stout stainless frame and better drill motor so that harder metals could be machined accuratly.

The stepper/servo and controller is the easiest part that you can hook up in an afternoon. It is "just" everything else that takes forever.

Thank you so much for the help. I can see me standing in my garage with a pile of stainless steel to weld up and then have to find a runner for the motor to actually move on lol. The stainless is not overly expensive but its not cheap enough to toss and write off as a boondoggle either.

[akmetal]

I agree

You need to do much more reading

Phil

[tmarks11]

Tormach has a statement that says AC brushless servos are superiour to what they install but are cost prohibitive, I looked up AC brushless servos online and they are not cheap but not so overly expensive that it couldnt be an add on.

Go to Novakon's website and you can see that they offer servo motors as an option on their machines, so you can see what the upgrade cost is (ignore the novakonsucks website, that guy is nuts).

I think Tormach wants a simple, straightforward product line that fits the price point they have chosen. Adding $3k for servos on the PCNC1100 would start getting close to where you could buy a new Haas TM1 instead.

I will do that thank you - do you have the link to the 700 page document

Hoss has been on the zone forever, and has done several cnc conversions.

http://www.cnczone.com/forums/bencht...sss-g0704.html

CNC Conversion

That is great news, can a regular lathe do 4 ft rod?

yes....but.... you need a setup to keep the rod from whipping around. the forces from spinning will unbalance a long rod, cause it to destroy, then bend, then beat you and your lathe to pieces. For something that long, you would probably need a couple steady rests to support it and thread it in bits. You would also need a very large lathe (14x60), which wouldn't fit in your space constraints. In reality, a long threaded rod like this would typically be done one a very specialized lathe (screw cutting machine).

Most people would just buy threaded rod instead if you needed something that long. Threaded rod is very cheap.

I really like the lay out of the Xcarve if it just had stout stainless frame and better drill motor so that harder metals could be machined accuratly.

You mean, if it had a cast iron frame so it absorbed vibration better.

What you are thinking of is a gantry mill. Lots of build logs on those. Here is a recent one where the builder used a Tormach 770 spindle. This one is lightweight, since it uses aluminum for the frame. Better would be a welded steel frame filled with epoxy composite mix for dampening. Plenty of thread on those too, you just need to browse through the project log forum.

http://www.cnczone.com/forums/vertic...ach-build.html

I can see me standing in my garage with a pile of stainless steel to weld up...stainless is not overly expensive but its not cheap enough to toss and write off as a boondoggle either.

For a decent size mill, 500 lbs of stainless isn't very cheap... and not the most ideal material either. No reason not to just make it out of steel.

[akmetal]

Go to Novakon's website and you can see that they offer servo motors as an option on their machines, so you can see what the upgrade cost is (ignore the novakonsucks website, that guy is nuts).

I think Tormach wants a simple, straightforward product line that fits the price point they have chosen. Adding $3k for servos on the PCNC1100 would start getting close to where you could buy a new Haas TM1 instead.



Hoss has been on the zone forever, and has done several cnc conversions.

http://www.cnczone.com/forums/bencht...sss-g0704.html

CNC Conversion


yes....but.... you need a setup to keep the rod from whipping around. the forces from spinning will unbalance a long rod, cause it to destroy, then bend, then beat you and your lathe to pieces. For something that long, you would probably need a couple steady rests to support it and thread it in bits. You would also need a very large lathe (14x60), which wouldn't fit in your space constraints. In reality, a long threaded rod like this would typically be done one a very specialized lathe (screw cutting machine).

Most people would just buy threaded rod instead if you needed something that long. Threaded rod is very cheap.


You mean, if it had a cast iron frame so it absorbed vibration better.

What you are thinking of is a gantry mill. Lots of build logs on those. Here is a recent one where the builder used a Tormach 770 spindle. This one is lightweight, since it uses aluminum for the frame. Better would be a welded steel frame filled with epoxy composite mix for dampening. Plenty of thread on those too, you just need to browse through the project log forum.

http://www.cnczone.com/forums/vertic...ach-build.html


For a decent size mill, 500 lbs of stainless isn't very cheap... and not the most ideal material either. No reason not to just make it out of steel.

Good point on the stainless (I could probably get regular steel locally too which would be easier, possibly scrap off a big project). I looked at Novakon's website and that looks like a nice machine

So what are the advantages of a brick crap house built gantry CNC vs the Tormach style CNC machines? The super small spaces on the tormach style is not the greatest but do you loose acuracy on the gantry style because of the span? For instance if it was like 3' wide so that I could also do small sheet metal cuts.

BTW these forums are awesome, with the vast knowlage on here I think I could really do this. Do the AC brushless servos pin into the path pilot and then my lap top feeds the gcode into the path pilot?

I am actually a little excited to try my hand at tig welding this thing together.

[Steve Seebold]

Go to Novakon's website and you can see that they offer servo motors as an option on their machines, so you can see what the upgrade cost is (ignore the novakonsucks website, that guy is nuts).

I think Tormach wants a simple, straightforward product line that fits the price point they have chosen. Adding $3k for servos on the PCNC1100 would start getting close to where you could buy a new Haas TM1 instead.



Hoss has been on the zone forever, and has done several cnc conversions.

http://www.cnczone.com/forums/bencht...sss-g0704.html

CNC Conversion


yes....but.... you need a setup to keep the rod from whipping around. the forces from spinning will unbalance a long rod, cause it to destroy, then bend, then beat you and your lathe to pieces. For something that long, you would probably need a couple steady rests to support it and thread it in bits. You would also need a very large lathe (14x60), which wouldn't fit in your space constraints. In reality, a long threaded rod like this would typically be done one a very specialized lathe (screw cutting machine).

Most people would just buy threaded rod instead if you needed something that long. Threaded rod is very cheap.


You mean, if it had a cast iron frame so it absorbed vibration better.

What you are thinking of is a gantry mill. Lots of build logs on those. Here is a recent one where the builder used a Tormach 770 spindle. This one is lightweight, since it uses aluminum for the frame. Better would be a welded steel frame filled with epoxy composite mix for dampening. Plenty of thread on those too, you just need to browse through the project log forum.

http://www.cnczone.com/forums/vertic...ach-build.html


For a decent size mill, 500 lbs of stainless isn't very cheap... and not the most ideal material either. No reason not to just make it out of steel.

I had the above mentioned Haas TM1 when I had my shop. It was equipped with a 10 station tool changer, rigid tapping and a 5C 4th axis.

I have told my wife, if I had it to do over again, I would have taken some of the money I got when I sold my FADAL and paid off the Haas and brought it home. The TM1 would run on single phase power, or I could have put it on a phase converter.

It had rapids of 200 IPM, and a spindle speed of 4,000 RPM, and 7.5 horsepower.

[tmarks11]

The super small spaces on the tormach style is not the greatest but do you loose acuracy on the gantry style because of the span? For instance if it was like 3' wide so that I could also do small sheet metal cuts.

You will get more flex, which corresponds to loosing accuracy. In machining, 0.002" can be critical amount of flex.

Want a gantry to cut sheet metal? You should build/buy a gantry-style cnc plasma cutter. A lot faster. The accuracy requirement to cut sheet metal is such that a gantry built out of rectangular steel tubing is good enough, especially since there is no cutting forces generated by the plasma torch. A commercially built cnc plasma cutter is MUCH cheaper than a cnc mill, and you can get them sized to cut full size and half size sheets of sheet metal.

http://www.cnczone.com/forums/genera...ting-machines/

Trying to build a gantry mill to both do machining and cut sheet metal is going to end up with something that does neither job very well.

[akmetal]

the sheet metal cutting is a secondary project, the primary work is smaller dense blocks of aluminum, copper and possibley Inconel. So it sounds like I need to stick with a standard cutting mill rather than a gantry style and maybe buy a plasma cutter later for sheet metal.

You will get more flex, which corresponds to loosing accuracy. In machining, 0.002" can be critical amount of flex.

Want a gantry to cut sheet metal? You should build/buy a gantry-style cnc plasma cutter. A lot faster. The accuracy requirement to cut sheet metal is such that a gantry built out of rectangular steel tubing is good enough, especially since there is no cutting forces generated by the plasma torch. A commercially built cnc plasma cutter is MUCH cheaper than a cnc mill, and you can get them sized to cut full size and half size sheets of sheet metal.

http://www.cnczone.com/forums/genera...ting-machines/

Trying to build a gantry mill to both do machining and cut sheet metal is going to end up with something that does neither job very well.

[Steve Seebold]

the sheet metal cutting is a secondary project, the primary work is smaller dense blocks of aluminum, copper and possibley Inconel. So it sounds like I need to stick with a standard cutting mill rather than a gantry style and maybe buy a plasma cutter later for sheet metal.

I cut a lot of Inconel 718 on my 1100, and I hold +/-.001 all day long.

I have one customer who gives me Inconel, a I do pretty well with it when he orders 50 to 75 pieces at a time.

[akmetal]

How leveraging is the tormach stepper motors over an AC brushless system?

I think I will just buy a working system once I save up and I can find other uses for my tig welder later on.

I cut a lot of Inconel 718 on my 1100, and I hold +/-.001 all day long.

I have one customer who gives me Inconel, a I do pretty well with it when he orders 50 to 75 pieces at a time.

[akmetal]

So I was just reading an article - We've all been seeing ads for Tormach Machines, here in PM. - Page 2

They are talking about VMC machines (which it sounds like are the house down payment type machines), but they were stating that machines like tormach wont get the acuracy that a VMC will get. I am looking at doing small scale engineering/science R&D prototyping so I wont be one of these guys trying to run the machine to death (and I picture some nut job screaming and yelling in a shop to get parts out the door because his buisness plan is crap), but I digress.

However I do need the parts to be acurate and repetable even if it takes longer, they have to be right, It is a MAJOR hassle to get raw materials to my location so if I have to scrap something its not cheap. The machine just has to do EXACTLY what is in the solid works/fusion model and the bit cant girate or skip steps and loose its zero point etc. I have to be able to hold the part up to the solid works model and it has to be the same.

As a super newbie its hard to tell if these guys in the above link are just grouchy machinists who think that a sub 10k machine should be doing production work or if machines like the tormach really do go haywire and chew up expensive stock for no reason.

I will be making scientific apparatus so I wont need it fast or even be using the machine every day (since most of the time spent designing and building the part out in solid works or on paper) but when I do run it it better be spot on and mate up with the rest of the design like a glove.

I cant afford a VMC so if that is the true answer then I will have to endevor to build one and it sounds like there are guys that do it (as eluded to in the attached discussion) so its not impossible. I am not sure if these guys are litterally doing their own copper coil windings for the spindle or how far they are taking the builds ...

Are there classes or VERY comprehensive text books on this stuff. For the price of a house down payment I can spend alot of sleepless nights figuring stuff out, I did it for my engineering degree, PE and other endevors lol.

If a VMC is not really necessary for prototype work then that is great news ... it is just hard to desipher crochety intellectual dishonesty from lagitimate machine short commings since i am so new at this but it has become necessary to learn CNC machining if I am to build any scientific apparatus of any significance.

Also if I am spending almost 10 grand I dont want to do my own tool changes becasue you invariably almost alway loose your zero point and something goes arwy when you start the machine again with the new tool.

[tmarks11]

So I was just reading an article - We've all been seeing ads for Tormach Machines, here in PM. - Page 2

Practical Machinist has an enormous chip on their shoulder concerning any machines built in asia. They are indeed "grouchy" concerning this topic. Post a question about Grizzly and they will snap at you like a cornered wolf.

With a commercial VMC, you can get accuracy in the tenths of thousands (i.e., +/- 0.0002"). With Tormach, you can hold +/- 0.001". How accurate do you need your parts to be?

Classes? You can go to your local tech college and take a 1-2 year course on machining, which helps... if you have the time.

Or you can start by downloading the (free) copy of Fusion 360 from Autodesk and learn how to do CAD/CAM, which is about 40% of the learning curve. Buy a copy of HSM Advisor from zero divide to handle all the Feeds and Speeds calculations for you. Go to youtube and watch some instructive videos on cnc machining. Go to NYCCNC and watch some of his early videos.

If you already have solid works (which is what I use), then download HSM Express plugin (which is the free version of HSM Works) as your CAM software.

Tool changing? You number all your tools, and set their zeroes before your start. When your g-code calls for a tool change, it will reference the tool table, and change the height offset to what you already have in there. A little bit of organization, and all you have to do is set the work offset when you start machining each piece.

[SoCalPlaneDoc]

So I was just reading an article - We've all been seeing ads for Tormach Machines, here in PM. - Page 2.

I shopped for a couple of years for a CNC mill.. had my eyes on a HAAS mini. My work has one along with a few other HAAS machines of various sizes. Of course, they all poo-poo'd Tormach machines as toys.. anyone that has unlimited access to a big-buck VMC will do that..
The reality is, they cost so much unless you're going to make money with it, it doesn't make sense Plus, (and the killer for me) they wont fit through a standard garage door.

Enter the PCNC1100...

I "settled" for the Tormach. Had 25K budgeted and was ready to pull the trigger and got lucky enough to find a used Series II machine within a reasonable distance.. 5.5K. after the trip to fetch it and upgrades to Series III level.. 7.2K thus began my CNC experience.

I design automated machines for a living and we have a decent sized and capable machine shop to make what I design. One thing the company insists on is that dowel pin holes have a .0005 tolerance between them. It's tighter than really needed but has become the standard in the company. I was told by all the tormach people that I'd never get than with the PCNC. well guess what?
I adjusted the gibs and bearings in accordance with Tormach's procedure, then prepared G-code to go *past* the first hole, come back slowly and drill, then move in the same direction to the next hole. BINGO, CMM machine verified, Dead on. In fact, the CMM guy told me the HAAS machines don't get that accuracy, but that's because they just bulldoze through and are totally dependent on the machine... technique means everything.
The big ol pricey VMCs are built to be accurate and efficient and can be operated by monkeys. that makes them profitable. But for R&D work, small runs, etc... the Tormach is big bang for the buck. anyone that knows how to be accurate hand-cranking a Bridgeport, Lagun, Tree, etc... can do beeuuutiful work on a tormach.
oh, btw.. after doing the Tormach adjustments I ended up with .0007 backlash on the Y, and .0004 on the X... I call that winning. on my endmills, I enter .0005 smaller than actual (.4995 instead of .5000 for instance) and get awesome results.
the support and available of parts and information cant be beat. Tormach been fabulous help even knowing that I have a used machine. There's a huge community with all kinds of advice and info too..

In the end, the CNC guys where I work love the machine for what it is (and they really like having an engineer in the house that understands their environment)

[WOTDesigns]

Your terminology of 'spot on' 'mate up like a glove' 'small scale' 'accurate' 'repeatable' 'EXACTLY what's in the solid model' all SCREAM that you won't know the difference between the two machines. If you don't know the tolerances you have no idea how to actually HIT the accuracy variance from HAAS to Tormach. I had both running side by side. Haas is faster, much better surface finish. That's it.

Reliability? Same. No lie. Haas crash cost me $3,000 for a new spindle. Tormach crash cost me couple hundred to rush new bearings. Haas control gets stuck every now and then so I have to power cycle. Chips regularly get onto the Y home and the machine alarms (WHICH WON'T BEEP EVEN IF BEEP IS ON dumbasses made it so alarm causes machine to sit till u happen to notice the light which is hard to see from my OFFICE)

I'm actually considering in another tormach this time a 770 with ATC because I need a 10K spindle. I really don't want TTS tooling again, but I think it's worth flying to test these parts on someone's 770 to find out in the spring. Its either that or $35K for a used (newer) Haas Super Mini mill to get the same spindle speed and travels. I DON'T want a 10K lb 10 foot long machine to make 1/4 square parts with 0.003 tooling... That's just dumb. I can get a Fadal 40x20 with 12K spindle that runs 3 phase (I have 3 phase) for much less than the Tormach but it takes a LOT of space.

If you don't know what those terms I listed first translate to in tenths or thousandths of an inch your running in circles splitting hairs. You are trying to decide if you need a Formula 1 car or a NASCAR to go get groceries more than likely. For part to press fit together you only need .003 accuracy and you can get that on either machine with a skilled operator or by accident and you can totally screw it up on both machines a bad operator or an error. It all comes down to learning a machine and learnig to machine. Expecting "perfect" parts by someone learning on either machine is erronious. Tormach takes a little more work to make the part beautiful, but it will make the same part just a little slower. Without explaining exactly what the parts are, how they fit for us to help you figure out the actual required accuracy asking about the accuracy of the two is kind of pointless.

I push out parts that hit .0005 accuracy all year long. Just finished 4,000 parts that only need at BEST .005" accuracy and some are .01" off but they don't matter. NO ONE can tell the difference unless they look REAL hard and it doens't effect the performance. My other product line I have to hit .0003 accuracy on EVERY SINGLE PART. The Haas can't "do" that. (ME, I ) do that.

The Haas spindle and tooling grow about .002" over the day... so a machine with .0003 "accuracy" isn't accurate to .0003" without constant adjustment. I have a 23 year old Miyano lathe next to it that is FASTER, weighs more an is more accurate than my new Haas.



Sent from my SM-G935V using Tapatalk

[WOTDesigns]

Just remembered the time a customer sent prints for milled part (4 sides/set up) that I run similar ones off all the time. Those prints called 0.0005" instead of the normal 0.005" tolerance on all features. I called to ask if it was a typo and he said "No, these are going on McLaren F1 and need to be perfect". I asked if he purchased a CMM since the last time I was there and he said "What's a CMM"? I literally laughed out loud and said "The only way you are going to be able to measure the $600 extra PER PART this is going to cost you". There are 3 or 4 features on the entire part that matter the rest mates to forgings. 0.0005" for a clearance hole for a bolt with .005" tolerance is just stupid. 20 parts @ $12,000 more than his normal cost... he decided to stick with .005" which 90% of the features end up at 0.0025 or better. Made those on the Tormach FYI. Still run them on the Haas.

[akmetal]

So this is one of the parts that I will want to make which will be made out of a 7" aluminum disk and some of the stubs will be threaded (NPT male threads) but I am thinking about just doing the threads on a lathe.

I am hoping I can talk Scott at Tormach into adding an AC brushless servo option for the PCNC 1100.

[WOTDesigns]

Pic is worth a thousand words! You can go ahead and get the "doing it on the lathe" idea off your front burner. Those are off center which means you need a 4 jaw chuck and have to position the part off center for the lathe to cut the threads. That takes skill and a lot of time. As in 5,000% more time than just thread milling them on the mill. On the mill it grabs the tool and makes them. Take a few passes total time to make taper/non-taper threads that size less than 2 minutes. It will take you longer than that to dial in the angle of the cross slide (I am thinking manual lathe) and at least 3-5 minutes per part to set up each offset on the 4 jaw until you get really good... then it will only take 4 or 5 times longer to do on the lathe than on the mill.

What is that pin on the lower left? Is that a locating pin? pressing into a bored pocket on something? Lets say you needed that pin to be +/- 0.002" which is TIGHT. You just program so that you leave .005 material when the part is done. Then create another program that just takes a finish pass on that feature. Measure the feature and it is actually only over by 0.003 (.002 error) you adjust your tool offset -.002 and run the follow up program. Part will be within 0.002 (likely better). It's easy to do on short run parts. Long run (100 parts at a time) it gets annoying.

For cutting a part that size out of AL you gain nothing having AC Servo over steppers. Servos accel and decel FAST and you can push them to the limit without losing a step. Program that part intelligently and you will never lose a step and it wont matter. If I still had an 1100 sitting next to my haas I wouldn't bother pulling a fixture off to make that on the Haas I would run it on the 1100.

You will need to schedule monthly maintenance time to adjust your backlash, gibs, and maintenance, but

[kstrauss]

Why do you want to cut the NPT threads on a lathe? Consider that: you would have to centre each spigot (mount the plate in a 4-jaw chuck?) before cutting the threads. To single point cut NPT you need a taper attachment on the lathe. It would be far easier to thread mill the male threads. From a manufacturing perspective it would seem better to make the raised features as separate pieces. Consider that if the NPT are to be repeatedly screwed brass would be far better than aluminum to reduce galling.

Have you ever made similar parts on a manual lathe? I have using machinist's buttons but it is certainly far quicker on my Tormach!

I still don't understand the obsession with BLDC motors.