[forrey]

Hi. I have pretty much decided to get the Torus Pro-S now that Ray's PDB seems to be up and running (come on tool changer).
I seem to be having a problem finding the rapid and max feed rates for the machine, looked on the web site but if they are there I am to tired to see them.
Thanks.
Forrey.

[SCzEngrgGroup]

Both the Pulsar and the Torus Pro servo versions are capable of 500 IPM rapids, but I would not recommend running either that fast. 200-250 is a more reasonable value for a machine this size.

Regards,
Ray L.

[BTP]

Just curious, why would you not recommend it? If the machine is capable, than why not?

[SCzEngrgGroup]

Just curious, why would you not recommend it? If the machine is capable, than why not?

Because:

a) It buys you very little in terms of productivity on most jobs. IME, just a few percent on average.
b) At those speeds, you have no time to react if something goes wrong. The X axis will travel over it's full range in about 3 seconds, Y about 2 seconds, Z about 1.5 seconds.
c) At those speeds, when you have a crash, there is potential for serious damage. There is a LOT of momentum behind a 200# table moving at 500 IPM.

Regards,
Ray L.

[forrey]

So 500ipm is the max rapid rate (G00) what is the max cutting feed rate(G01).
Thanks.
Forrey.

[kvom]

The machine doesn't know the different between a rapid and a feed, so maximum feed rate is the same as the rapid.

[waltpermenter]

Because:

a) It buys you very little in terms of productivity on most jobs. IME, just a few percent on average.
b) At those speeds, you have no time to react if something goes wrong. The X axis will travel over it's full range in about 3 seconds, Y about 2 seconds, Z about 1.5 seconds.
c) At those speeds, when you have a crash, there is potential for serious damage. There is a LOT of momentum behind a 200# table moving at 500 IPM.

Regards,
Ray L.

a) professionals like rlockwood and myself would certainly disagree, if it aint cutting you aint making money as the saying goes.
b) you aren't going to be able to react at 200 ipm either so this is nonsense, a crash is usually from the cutter making wrong direction move or a rapid instead of feed move so the crash is pretty much instantaneous.
c) once the endmill snaps the damage is done
walt

[forrey]

The machine doesn't know the different between a rapid and a feed, so maximum feed rate is the same as the rapid.

Ahh Thank you.
Steve.

[SCzEngrgGroup]

a) professionals like rlockwood and myself would certainly disagree, if it aint cutting you aint making money as the saying goes.
b) you aren't going to be able to react at 200 ipm either so this is nonsense, a crash is usually from the cutter making wrong direction move or a rapid instead of feed move so the crash is pretty much instantaneous.
c) once the endmill snaps the damage is done
walt

a) Few people here are making money with their machines, so this is a specious argument.
b) It certainly is not "nonsense". I've stopped crashes from occurring lots of times, when I could see what was going to happen if I didn't stop it. When running new code, I always have my thumb over the stop button, just in case. Even at 200 IPM, it takes quick action sometimes to prevent a crash. But it is absolutely do-able in many cases.
c) The endmill won't snap if you rapid into a hard stop, or crash the spindle into a vise or other obstruction. At 500 IPM, there's a much greater chance of breaking something expensive. The energy of an object in motion increases with the square of velocity, so at 500 IPM there is more than 6X as much energy to dissipate in a crash.

It does not benefit less experienced users to tell them what a "professional" would do on a "professional" machining center when running volume production. People here are NOT professionals, these are not production-quality machines, and nobody here is running the same parts over and over, day-in, day out with carefully tuned code. And nearly everyone here is running Mach3 which is notorious for doing random, unpredictable things at random times. These machines are worlds apart from a real VMC in terms of capability and performance. Do the math, and you'll see my comments are absolutely correct for these machines. Nobody here is cutting at 300 IPM - more like 20-30 IPM on average. At those speeds, the difference in job time between 200 IPM rapids and 500 IPM rapids is mouse nuts. On the job I'll be running in a few hours, simulation shows the job will run in 59 minutes, 34 seconds with 200 IPM rapids. With 500 IPM rapids, that drops all the way down to 58 minutes and 55 seconds. A whole 39 seconds saved! That's not even 5 minutes per day saved! Woo Hoo!

Regards,
Ray L.

[rlockwood]

a) Few people here are making money with their machines, so this is a specious argument.
b) It certainly is not "nonsense". I've stopped crashes from occurring lots of times, when I could see what was going to happen if I didn't stop it. When running new code, I always have my thumb over the stop button, just in case. Even at 200 IPM, it takes quick action sometimes to prevent a crash. But it is absolutely do-able in many cases.
c) The endmill won't snap if you rapid into a hard stop, or crash the spindle into a vise or other obstruction. At 500 IPM, there's a much greater chance of breaking something expensive. The energy of an object in motion increases with the square of velocity, so at 500 IPM there is more than 6X as much energy to dissipate in a crash.

It does not benefit less experienced users to tell them what a "professional" would do on a "professional" machining center when running volume production. People here are NOT professionals, these are not production-quality machines, and nobody here is running the same parts over and over, day-in, day out with carefully tuned code. And nearly everyone here is running Mach3 which is notorious for doing random, unpredictable things at random times. These machines are worlds apart from a real VMC in terms of capability and performance. Do the math, and you'll see my comments are absolutely correct for these machines. Nobody here is cutting at 300 IPM - more like 20-30 IPM on average. At those speeds, the difference in job time between 200 IPM rapids and 500 IPM rapids is mouse nuts. On the job I'll be running in a few hours, simulation shows the job will run in 59 minutes, 34 seconds with 200 IPM rapids. With 500 IPM rapids, that drops all the way down to 58 minutes and 55 seconds. A whole 39 seconds saved! That's not even 5 minutes per day saved! Woo Hoo!

Regards,
Ray L.

The effect of rapid speeds on cycle time is dependent on too many factors to use isolated examples. For instance, I have a program I'll be running tomorrow which at 200ipm rapids would take 3 hours. At 500ipm rapids, it will take 1.2 hours. All it does is rapid back and forth for 36,000 inches, but is equally valid in comparison to your case sample, which likely is horribly optimized to make good use of rapid motion.

If you have fast rapids, you can take advantage of them and it will have a dramatic effect on output. Reccomending people not take advantage of the full rapid speed available simply makes people assume it is not truly capable. If it is capable, you are not doing Novakon any favors.

Have you ever considered running unproven code at reduced rapid speeds? I regularly run machines with 1200ipm rapids, slow by modern standards. The last guy you suggested 500ipm rapids were not appropriate to, runs a Brother with 1900ipm rapids in roughly the same envelope as a Torus Pro. I bet he chose that machine because the increased rapids have minimal impact on throughput.. Do you think 500ipm will seem too fast to him?

When you take your personal experience and extrapolate it to everyone else, you start sounding like a certain Tormach fan.

Do you not have a thread here titled "Torus Pro as a production machine"? Rightly or not, there are NUMEROUS companies running Tormachs 40 hours a week. Perhaps the same cant be said for Novakon, yet.. But don't you think they would like some of that business?

I assure you, many peoples ears perk up when their 500ipm rapids are mentioned, and the interest quickly goes away when you hear "yeah but YOU probably shouldn't. You'll crash." Do Ferrari sales men constantly remind their customers they're more likely to wreck by driving too quickly? That's not their concern. Is the machine capable of a certain performance level, or not? Quit dodging the damn question, or tempering the answer. If YOU choose not to run it that fast, that's your decision. It has no bearing, however, on what anyone ELSE should do.

[waltpermenter]

a) Few people here are making money with their machines, so this is a specious argument.
b) It certainly is not "nonsense". I've stopped crashes from occurring lots of times, when I could see what was going to happen if I didn't stop it. When running new code, I always have my thumb over the stop button, just in case. Even at 200 IPM, it takes quick action sometimes to prevent a crash. But it is absolutely do-able in many cases.
c) The endmill won't snap if you rapid into a hard stop, or crash the spindle into a vise or other obstruction. At 500 IPM, there's a much greater chance of breaking something expensive. The energy of an object in motion increases with the square of velocity, so at 500 IPM there is more than 6X as much energy to dissipate in a crash.

It does not benefit less experienced users to tell them what a "professional" would do on a "professional" machining center when running volume production. People here are NOT professionals, these are not production-quality machines, and nobody here is running the same parts over and over, day-in, day out with carefully tuned code. And nearly everyone here is running Mach3 which is notorious for doing random, unpredictable things at random times. These machines are worlds apart from a real VMC in terms of capability and performance. Do the math, and you'll see my comments are absolutely correct for these machines. Nobody here is cutting at 300 IPM - more like 20-30 IPM on average. At those speeds, the difference in job time between 200 IPM rapids and 500 IPM rapids is mouse nuts. On the job I'll be running in a few hours, simulation shows the job will run in 59 minutes, 34 seconds with 200 IPM rapids. With 500 IPM rapids, that drops all the way down to 58 minutes and 55 seconds. A whole 39 seconds saved! That's not even 5 minutes per day saved! Woo Hoo!

Regards,
Ray L.

Your comments don't make a lot of sense for one people that spend this kind of money for a hobby machine more often than not are doing it to make parts they sell, yourself included. read more posts, i read plenty of people talking about making 500 of this or 1000 of that. also i've already mentioned 2 professionals and there are plenty more who get these machines when they retire or semi retire from the profession. seems rather silly to market a machine with 500 ipm rapids and suggest that users hinder that but maybe there are reliability issues with the machine at those speeds so the reduction is necessary and 500 is deceptive advertising.
walt

[SCzEngrgGroup]

I assure you, many peoples ears perk up when their 500ipm rapids are mentioned, and the interest quickly goes away when you hear "yeah but YOU probably shouldn't. You'll crash." Do Ferrari sales men constantly remind their customers they're more likely to wreck by driving too quickly? That's not their concern. Is the machine capable of a certain performance level, or not? Quit dodging the damn question, or tempering the answer. If YOU choose not to run it that fast, that's your decision. It has no bearing, however, on what anyone ELSE should do.

I do soooooo wish people would respond to what is actually said, rather than making up garbage. I am not "dodging the damn question". I made it VERY clear that both machines are perfectly capable of running 500 IPM rapids. If you bother to READ my post, far from saying "yeah but YOU probably shouldn't. You'll crash.", you'll see my exact words were "I would not recommend running either that fast." How is that telling anyone what they "should" or "should not" do? Please get your facts straight.

Sure, I could concoct some pathological job to support almost any crazy argument. But, again, READ what I actually wrote: "It buys you very little in terms of productivity on most jobs. IME, just a few percent on average." This is based on years of experience, doing a mix of small production runs, and one-offs. Nothing you've said refutes my point. And that's not running "horribly optimized" code. It's straight out of SheetCAM, with zero optimization. I would argue that any code that does 36,000 inches of rapids in an hour IS horribly UN-optmized, and could almost certainly benefit FAR more from simple optimizations to reduce the amount of rapid travel than from simply speeding up the machine.

Personally, I'm going to exit this thread, since it seems some want to turn it into a pi$$ing match, rather than a useful discussion.

Regards,
Ray L.

[rlockwood]

By horribly optimized I intended "not optimized", meaning it likely made poor use of rapid moves. The theoretical program I discussed ONLY travels back and forth in rapid, and serves no other purpose but to present an equally ridiculous counter example to yours.

Regardless of your intentions, the fact that you (whether you like it or not) now serve as a representative of Novakon and are reccomending people to not make use of the full speed the machine is "capable" of, reads like a disclaimer, or warning that it is not reliably able. Couple that with an incredibly weak argument given as logic. Consider that anyone considering the servo version is likely doing it on some basis that they believe the increase in speeds or accuracy will justify the additional cost.

These will likely be folks who are looking to buy it for financial gain, not simply as a hobby. A few percentage gains when extrapolated over the 2080 hours in a standard work year, in MANY scenarios, pay for the whole damn machine.

P.S. I can piss pretty damn far!

[Hirudin]

Once in a blue moon, when I actually run my Torus Pro-S, I run the X and Y with 12000 mm/min rapids (IIRC). I think I have the Z rapiding at 9000 mm/min. It's debatable whether that's smart or efficient, but it's been working for me.

Most of my crashes have been the result of my own errors when jogging or using MDI. The majority of the rest I blame on Mach3 doing something unexpected. As a non-expert, non-professional, non-blowhard I wouldn't suggest amateurs put too much stock in their ability to hit the E-stop button *before* a crash happens.

Also... seriously: it would be nice if everyone participating in these discussions would learn to read before spouting off.

[forrey]

I am a professional (as much as anybody can be ), served a five year apprenticeship in the 70's in the UK and have been around CNC since the mid 80"s. I know this machine will not have the capabilities of a full pro machine, I do on the other hand want to make some part time money with this as well as doing a few projects for my self.
I have been following most of the posts and thought this machine is the best way to go. Now something Ray said, QUOTE And nearly everyone here is running Mach3 which is notorious for doing random, unpredictable things at random times. . has me some what concerned if say I got the coming tool changer I would not be able to leave the machine for fear of Mach3 making a boner and crashing, why have the tool changer if you have to hover over the machines big red. What are the reliable alternatives to the software?.
Forrey.

[rlockwood]

Hirudin, I concur entirely. I don't think someone new to CNC or machining in general should be running unproven code at potentially dangerous speeds. I also cannot understand why, if mach3 is such a liability, Novakon or anyone else would be supporting it and/or bundling it with their hardware.

Its early, more thoughts when I wake up..

Torrey, Linux CNC has never done anything other than what I've commanded it to. I also entirely agree with you; if you can't leave your machine unattended for fear it will stuff the tool into the table, what's the point in an ATC?

[kvom]

My shop instructor insisted that the first run of a new program on the VMC be done cutting air at 10% feed rate with a finger on the stop button. I try to follow this to some degree. Almost all of my mistakes have been jogging/MDI, or mis-measuring clearance on workholding.

If I were doing HSM toolpaths, then it's likely that having the rapids set to the max would be a big plus.

So I can see both sides of this discussion. Actually on my 4 y/o NM200 my rapids are set to 75.

[rlockwood]

I agree, your shop instructors directions were entirely logical. In fact, it entirely backs up my argument that it's simply nonsense to limit the over-all maximum travel speed on the basis that you may crash the machine. Its as simple as turning the rapid dial down in situations where you are unsure of what the machine will do.

The sole purpose of a CNC machine is to reliably repeat a set of instructions to a high level of accuracy.

[SCzEngrgGroup]

I have been following most of the posts and thought this machine is the best way to go. Now something Ray said, QUOTE And nearly everyone here is running Mach3 which is notorious for doing random, unpredictable things at random times. . has me some what concerned if say I got the coming tool changer I would not be able to leave the machine for fear of Mach3 making a boner and crashing, why have the tool changer if you have to hover over the machines big red. What are the reliable alternatives to the software?.
Forrey.

There are few attractive alternatives, and none that would be widely acceptable to the average user. Probably the second-place holder is LinuxCNC, which is a non-starter for most user, due to unfamiliarity with Linux, and because none of the typical users other applications (CAD, CAM, etc.) are available under Linux. There are some excellent, VERY robust solutions running under Windows, but they require hundreds, or thousands, of $$ in additional hardware. I run a KFlop on all of my machines, and it has proven absolutely bullet-proof, but the configuration is beyond the capabilities of the average CNC hobbyist, as it requires a fair amount of C programming. For me, it has been a God=send, as it has given me a dead-reliable machine, and even enabled me to write my own CNC controller application, that let's me work the way I want to.

Mach3 works fine for many people. Most only rarely have a problem. Some never do. Others (I was one of these....) have frequent problems with it. It is trying to be a real-time application running on a VERY non-real-time operating system. As a result, its performance is very much dependent on the configuration and characteristics of the specific PC on which is it running, and the hardware it is controlling. Some configurations work well, some very poorly.

If the benchmark of acceptable performance is 100% reliability, 100% of the time, then the ONLY acceptable solution will be a dedicated, commercial machine controller from one of the major CNC vendors (Fanuc, etc.) - i.e. many THOUSANDS of $$ for the controller alone. A Mach3 license is under $200, and runs on inexpensive commodity PC hardware, so it's completely unrealistic to expect the same level of performance and reliability from it as you would from a purpose-built, dedicated, CLOSED hardware/software system. Are you going to NOT buy any Mach3-based machine with a toolchanger for fear Mach3 may cause it to misbehave a few times a year?

Regards,
Ray L.

[waltpermenter]

funny how mach 3 suddenly isn't so terrible if it means costing potential atc customers. just be sure to set your rapids at 2 ipm so you have time to run across the shop and hit the e-stop.
walt