[Scott_bob]

We had Fadal Engineering do a test cut for us using their new Augusta control, otherwise known as the 104d.

Its pretty fast, has good compatibility with legacy Fadal routines, is supposed to be their new control on all the CNC line by early 2004.

Fadal has finally decided to address the serious deficiencies in their control designed in the 1980’s. That is 15 years ago!

The CNC industry should not tolerate any longer poor performance and high repair and component prices!

Why should we have to pay hundreds of $s for memory, keyboards, even the PC in an MP32 control option is 3 times what a PC would cost… And how about the price they ask for a floppy drive, you’ve gotta be kidding!

Is there any interest out there in discussing the limitations and solutions for this aging CNC control?

I’d be glad to help…

Scott_bob

[Klox]

Welcome to CNCzone Scott_bob!
Personally i don't know anything about the Fadal Control. It's an interesting point you make on spares, repairs to CNC controls.

Do you have knowledge on other controls? I think this can get a very interesting topic.....

Klox

[cadman]

About 1 1/2 years ago I was told Fadal was coming out with a new controller and it was supposed to be at WESTEC this past year, so its a good thing they are finally going to release it.

I would like to see them support the NURBS G code (G62.2 ?). My cam system will output NURBS and right now roughly 90% of my work coming in is entirely NURBS surfaces. For the most part my controller (-5) works fairly good, but when the toolpaths start to contain lots of simultaneous XYZ points on steep surfaces (especially "L" shaped), the machine starts to jerk and can't keep up with the programmed feedrate.

cadman

[Scott_bob]

I have some experience...
28 years, ugh!
25 years CNC...

Mostly, Fanuc, Fadal, Allen Bradley, Bandit, Tree, Yasnak, oh a bunch more I guess...

But enough about me.
HSM is a revolution in our industry. It is being promoted not so much by machine tool builders at large, but by other high tech players.
End Mills or Cutting Tools have evolved to a state that currently they can go much faster than most CNC machines can go smoothly.
Let me stop here with this. What is meant by smooth motion control?
Automatic Feed reduction?
Fadals ansewer to this is to slow the feed at the end of every block of code. But this is only good when a sharp change in direction is next. Slowing the feed rate when there is not a sharp change in direction is just a waste of time. And in 3D machining, Hugh amounts of time...
In 3D, usually the code is point to point or linear, (but not always) and Fadal has done some work here to improve performance, but that should be another point.

I like the point that one guy on this site says:
1st you get it right, then you go faster...

[Rekd]

Hi bob, err.. scott.

I didn't vote because the option I would have voted for is not there. (I'll refrain from stating what that option would be). I've used fadals since the early '80s and I'm here to tell you that the controller is nearly the only thing I liked about them. The 'real' keyboard, the functions and utils etc.

'Rekd

[Scott_bob]

what is the factor that:

"most" significantly affects High Speed Machining on a Fadal
HSM is most commonly defined as feed rates above what the Fadal is capable of delivering.

Feeds lower than 300 IPM would be not be considered HSM by some industry definitions. HSM is somewhat relative, because we all refer to processes that are faster than what has been possible to date on our own machinery. But I think that using the phase HSM when we are using feed rates in this lower range of the spectrum <300 IMP, deminishes the term HSM.

The term High Velocity Machining should definetly be reserved for describing processes higher than 1000 IPM. At this level there is much less discussion going on.

There are CNCs out there accurately feeding at "programmed" feeds from 1000 to 3000 IPM. Not on a Fadal...

I say "programmed feeds" because a good control with intelligent look ahead and feed optimization can be programmed at an optimum feed rate for the tool in use, and the control slows down relative to the angular change in dirrection. Like the Fadal does in point to point, only with a much higher degree of accuracy.

Using a .187 dia. Carbide End Mill. How fast can your Fadal mill a slot .250" x 1.000" and hold +/-.001 in 6061-T6 Aluminum say .2 dp?

Feed = ?

[Rekd]

Using a .187 dia. Carbide End Mill. How fast can your Fadal mill a slot .250" x 1.000" and hold +/-.001 in 6061-T6 Aluminum say .2 dp?

On the Fadals I've run... about 15 IPM, and even then I wouldn't guarentee +-.001



Of course, things may have changed in the last 5 or 10 years since I've used them.

'Rekd

[Scott_bob]

ref:

Posted by Rekd on 12-09-2003 07:47 PM:
--------------------------------------------------------------------------------
Using a .187 dia. Carbide End Mill. How fast can your Fadal mill a slot .250" x 1.000" and hold +/-.001 in 6061-T6 Aluminum say .2 dp?
--------------------------------------------------------------------------------

On the Fadals I've run... about 15 IPM, and even then I wouldn't guarantee +-.001
************************************************** *

That sounds just about right... On a Fadal the slot width could be held +/-.001 going NO faster than F15.
--------------------------------------------------------------------------------
I should have described the end of the slot for this discussion better, it would have R.125 on both ends, so it would be oval, (G03 at both ends)...
--------------------------------------------------------------------------------
This Slot length would NOT be +/-.001

Why?
Because the Fadal Control is not accurate within +/-.001 in circular interpolation at feeds higher than F40. Not even the New Augusta or 104d...

A CNC machine uses what is called the "feedback loop" to control the current position in a move as constantly as it is capable of processing. The CNC with the tightest feedback loop will be the most accurate; the loosest feedback loop will be the least accurate.

Let’s face the facts here; we all get what we pay for, right?
The low end priced CNC will not be as tight as the high end priced CNC.

Would any of the Lower priced CNC users out there disagree with this observation:

Accuracy of +/-.001 interpolated circles smaller than .500 in diameter or R.250
Feed NO faster than F15.
Accuracy of +/-.001 interpolated circles bigger than 1.000 diameter or R.500
Feed NO faster than F40.

Anyone perform Renishaw Ball Bar tests on your CNC to see why?
Renishaw identifies and rates the factors as to their affect on circularity…

By the way Rekd, what was that “other” contributing factor that you wanted to include in the poll?

Scott

[Rekd]

Originally posted by Scott_bob
ref:

Posted by Rekd on 12-09-2003 07:47 PM:
--------------------------------------------------------------------------------
Using a .187 dia. Carbide End Mill. How fast can your Fadal mill a slot .250" x 1.000" and hold +/-.001 in 6061-T6 Aluminum say .2 dp?
--------------------------------------------------------------------------------

On the Fadals I've run... about 15 IPM, and even then I wouldn't guarantee +-.001
************************************************** *

That sounds just about right... On a Fadal the slot width could be held +/-.001 going NO faster than F15.
--------------------------------------------------------------------------------
I should have described the end of the slot for this discussion better, it would have R.125 on both ends, so it would be oval, (G03 at both ends)...
--------------------------------------------------------------------------------
This Slot length would NOT be +/-.001

Why?
Because the Fadal Control is not accurate within +/-.001 in circular interpolation at feeds higher than F40. Not even the New Augusta or 104d...

A CNC machine uses what is called the "feedback loop" to control the current position in a move as constantly as it is capable of processing. The CNC with the tightest feedback loop will be the most accurate; the loosest feedback loop will be the least accurate.

Let’s face the facts here; we all get what we pay for, right?
The low end priced CNC will not be as tight as the high end priced CNC.

Would any of the Lower priced CNC users out there disagree with this observation:

Accuracy of +/-.001 interpolated circles smaller than .500 in diameter or R.250
Feed NO faster than F15.
Accuracy of +/-.001 interpolated circles bigger than 1.000 diameter or R.500
Feed NO faster than F40.

Anyone perform Renishaw Ball Bar tests on your CNC to see why?
Renishaw identifies and rates the factors as to their affect on circularity…

By the way Rekd, what was that “other” contributing factor that you wanted to include in the poll?

Scott

You're absolutely right, Scott. To do a circle THAT small with a Fadal would take an act of congress..

Have another look at the poll, in which I have now voted.. (WoW, I tied it up!) My second choice would be the programmed tool path, as it would directly affect the functionality via Feeds.

'Rekd

[Scott_bob]

Ref. (new entry in the Poll)…

A Different Controller/Machine

Response by,

Matt
CNCZone Community Director
San Diego, Ca

Your problem is my problem. Unless I have a problem with your problem. In which case we have a problem.


This is absolutely my point!
So, to ask again the Poll Question:

What is the most significant factor in High Speed Machining on a “Fadal”

Your answer is door #1 or CNC control right?

Remember the question is “on a Fadal”… You are absolutely right, that without either changing the control, in which case the CNC would no longer be a Fadal or, change the whole CNC, true HSM is not possible!

Ref.
HSM is most commonly defined as feed rates above what the Fadal is capable of delivering.

Feeds lower than 300 IPM would be not be considered HSM by some industry definitions. HSM is somewhat relative, because we all refer to processes that are faster than what has been possible to date on our own machinery. But I think that using the phase HSM when we are using feed rates in this lower range of the spectrum <300 IMP, deminishes the term HSM.


Is any body out there interested in a solution to this problem of doing HSM on a Fadal CNC VMC or HMC, or any other low end CNC machine?
There are solutions!

Come on, chime in folks…

The only way you’re going to get your favorite Machine Tool Builder to listen to your demands is by leveraging your opinion!
Is your CNC builder listening to you?

Best Regards,
Scott

[Rekd]

I'm not sure where you're going with this, but I've got an idea.

Mayhaps you should re-phrase the question. As I read it, it is limiting me to a choice of what's already available on the Fadal. Perhaps a question like "What is the most significant change Fadal could make to improve/add performance for HSM" Or "What is the most significant 'limiting' factor in High Speed Machining on a Fadal"

Just a thought.

BTW, are you a Fadal user, rep, engineer, lawyer or what? You mention you 'can help', in what ways?

'Rekd

[Scott_bob]

Ref.
__________________
Mayhaps you should re-phrase the question. As I read it, it is limiting me to a choice of what's already available on the Fadal. Perhaps a question like "What is the most significant change Fadal could make to improve/add performance for HSM" Or "What is the most significant 'limiting' factor in High Speed Machining on a Fadal"
__________________


I'll try to improve the Poll Question, (If there is an admin looking, go for it, I tried but I don't have access even though I started the thread)
I like your suggestion:
__________________
What is the most significant 'limiting' factor in High Speed Machining on a Fadal
__________________

To your BTW, Qu.
Yes, no, yes, no
And I can only help by passing on my experience.

Would you or anyone else be interested in making your CNC Milling machine accurate in 2D or 3D contouring?
Matt, you are correct, you cannot get there with a Fadal control, old or new...
"There" means to me; true HSM or smooth accurate feed rates ABOVE 300 IPM for Aluminum, and smooth accurate feeds for harder materials 3-5 times the feeds possible on the lower end CNC machines out there...

Keep in mind that HSM means:
lighter depth of cuts at very high feeds, with better cutting tools and more RPM...
(very generally defined)
HSM is more a process, than a destination.

I'll give as much info as you want to hear.
Is anyone else interested besides Matt?

Regards,
Scott

[Rekd]

I modified the question for you.

How long have you been, um... blessed? with using Fadals?

'Rekd

[Rekd]

BTW, I also modifed the poll selections, and updated my choice accordingly.

'Rekd

[Scott_bob]

I finished my school training in 1976 at believe it or not, from San Diego City College under Mr Bell and Mr Dan... something or another.

It wasn't until 1987-88 that I started using Fadals, so 16 years...
They have changed very little since then... It was Fanuc, Yasnak, Tree etc bofore that...

How about that, I am a working graduate of the San Diego City College machineing technology system in the late 1970's, wow, small world... No regrets either, this trade has been good to me.

Hey thanks for updating the Poll question. Should be an interesting discussion as it goes on...

Regards,
Scott

check out the comparison charts over in HSM
http://www.cnczone.com/showthread.ph...0&pagenumber=4

[hardmill]

Scott_bob,
I'm in, send the file my way.
[email protected]
We have a Yasda w/ a fanuc 16i and hpcc (high precision
contour control).

PEACE

[Rekd]

I'd like to check it out also, although I don't have a HSM machine, I've got a 2002 VF3 and 4 2000 VF3's, I'm curious how they'll do.

'Rekd

[tschuett]

Scott:

Your posts have a certain commercial tone to them as if you're selling something, but trying to get folks to egg you on rather than just coming out with it. How come?

Glancing at the test and the data you've posted, I don't see any constants but the program. Thus, speed and accuracy may both vary. That doesn't seem very scientific.

You may say that we are only testing BPT and that doesn't have accuracy, but the issue is then how the CNC uses that BPT. If, as many do, processing data does not require the accurate execution of that data, fast BPT may be quite misleading.

I once saw a high speed CNC that cut a 2" square so fast that the square turned circular. Your test is a circle, so there is no accuracy issue to it, but that overlooks the need for accuracy in the controls that are tested, potentially rendering the results pretty useless.

In my experience, results vary. High speed is relative to your perspective, your materials, your cutters, your machine, and perhaps most important, your budget. What is right for one individual or application may not be right for another. What is best might not be affordable or might strain the budget or financial resources available.

Fadal has long built a great machine value. Old and new Fadals can run faster and more accurately with alternate controls, including some retrofits. Reliability and service-ability may also improve or diminish with alternate controls. Needless to say, Fadal will continue to do their very best to make the best machine value possible in today's very competitive marketplace.

[Scott_bob]

i like a good debate, just not with myself.

I don't have anything to sell, unfortunately, else a guy could make some extra dough.

I don't hold out much hope of getting anybody in the machining community to agree with me, as we are some of the toughest people to agree on anything. There has got to be easier ways to make a living than by selling technology. In this industry I'm sure you know smart guys who are just getting by nowdays.

I do want the machining industry to do well in our country, and that is what sometimes motivates me to spread some good news around, I think a lot of us these days could use some more of that.

When I get home I'll go into more detail per your comments...

Regards,
Scott

[Scott_bob]

regarding statement from:
tschuett
__________________
Glancing at the test and the data you've posted, I don't see any constants but the program. Thus, speed and accuracy may both vary. That doesn't seem very scientific.
__________________

You're right about one thing, fast BPT doesn't mean anything without accuracy. The BPT process created by my friend does address this issue, by making his offer to bring his Renishaw ball bar, to measure accuracy. That would satisfy the scientific requirement. This also is why the BPT program runs a circle using X, Y points... Also by moving in a circle you can see backlash, and with the ball bar instrument you can measure velocity.

Did you know the Fadal control has no velocity feedback? That's a tachometer for a servo system. You cannot get good smooth motion if you don't control your velocity, you can't control what you don't measure...

BTW, I see our Fadals make shapes different from what is programmed every single day. That is exactly why I tried in vain for years to get our company to buy different machines, better ones. That is not going to happen, so what is another option?
Fix what is broken. The control on a Fadal is what is responsible for it's poor accuracy. I'll post a typical plot of a Renishaw BallBar analysis on one of our Fadal CNCs.

My friend, believe me, I have not overlooked the need for accuracy, in fact that is what has driven me to research any option to fix our problems. We make hundreds of molds each year, and besides the slow feeds we're stuck with, the accuracy is always at risk because we try to go as fast as we can, doesn't everyone?

tschuett, you mention seeing a machine cut a 2" square so fast that it turned circular. You must have been watching a Fadal, or a low end CNC like it. This reinforces my point. A good control would have given you a 2" square. If a control is programmed to move in a square, then it better make a square.

I gotta say concerning your last 2 paragraphs, I don't believe a shop like mine can afford to wait around for solutions to our challenges. We are facing cost downs from our automotive customers in the range of 5 to 15 percent every year. How are you going to keep automotive customers from going elsewhere. And it's no longer just the automotive customers, it's the aerospace customers who are dictating these cost downs as well. The good ol days are gone, I'm sorry to say.

Believe me when I say, if you wait for a low end CNC machine tool builder like Fadal to solve these challenges, you are going to go out of business. I had a chance to meet with Fadal Engineerings, Augusta R & D team recently. They were very frustrated with the absence of upper managements sustanined support for their development project. Funding cuts, shifted deployment dates, they had to endure canabalized R & D equipment as parts were needed to be used on a machine to ship to a customer. Well, as a customer we decided not to wait for them any longer.

Best regards,
Scott