Official Luthiers Forum!

I've heard good things about both units and PreciseBits has a precision collet for both. I just need to deal with mounting it. What is yours mounted on, a K2?

Yeah, I've heard they're pretty loud. I'm spoiled with my Foredom for inlay routing as it's pretty darned quiet.

It'll have to do for now until I can afford quiet high speed. I wonder... are the rpm requirements because of a small end mill not being able to clear chips fast enough or the specialized needs of shell?

Of course! I won't even saw it by hand without my custom made bench pin with shop vac attachment.

I already warned the guy who's shop it's in about this. There is NO way I'm spraying his shop with shell dust. And the DC hose needs to be grounded. This controller has a Xylotex board. I hear they're easier to zap. From the little research I did on this years ago that wouldn't have been my first choice. Geckodrives sound much better, even though they're more expensive.

Well hopefully I'll have access to the old PC to use tomorrow and I can install Mach and see if I can get this thing to move! I have no .xml file for it. I'll just have to wing it. I watched Art's setup video tonight so I have a place to start. I'm so excited!

...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

Well, for my first machine, it was simply mounted with Home-Depot pipe straps Hey, it worked!

I am now running a VFD spindle, but that thing was not bad for a beginner....You can see my latest machine on a post here.

If you look at precisebits, they have a good discussion on basically calculating speeds and feeds, and it is a good read overall. If you stay within the limits, no problems, but feeds will be rather slow for that small diameter bits...

Xylotex boards are great beginner boards, but you can zap them very easily, and it looks like you have done your homework. You can't kill Gecko's! They are great.

Take a look also at keling technology, I am currently running their drivers, and those things breathed new life into my steppers (running them at 60v though...)

http://kelinginc.net/StepperMotorDriver.html

_________________
-CyborgCNC
http://www.cncguitar.com
https://www.facebook.com/CncGuitar

Whatever works! I was just thinking today that I had a lot to learn - to me, double-sided tape is a fixture.

I saw your spindle and got very jealous. I wanted one and I didn't even have anything to put it on yet.

I'm learning a lot from their website.


Well I'm not zapping anything because I'm not even talking to it yet. Human issues, not computer ones. It turns out the old machine that was available is not available until the very busy owner gets around to backing it all up. It's my problem that I'm bouncing up and down wanting to get at this mill.

So... enter the craptop! I have an old Systemax laptop, so I decided to roll the dice. It seems like maybe a 50/50 chance it might work, from what I've read. But it sure beats just waiting. I ordered a Startech PCMCIA parallel port card for it as I saw that at least a couple people got those to work. So if I'm lucky, on Weds I can start trying to test the controller. Or having it fail miserably, but at least I'll be doing something with it. Gee, there's no smiley for going out on a limb. ;->

Thanks,
...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

There's no smiley for opening a can of worms either...LOL. You had to mention your machine, cyborg. I just looked at the shot of the inside of your controller box on your website. It might be old, but still... my box looks empty! I took the liberty of comparing them side-by-side, mine is on the left, yours on the right:



Is that a power supply in the upper left? You have a power supply? I have a transformer and 2 capacitors. What does your pwr supply do beyond that? Filtering? Voltage regulation? Mine only has 2 prong A/C plug. That looks dubious to me. It's like a 60's lamp cord. ;->

You have another circuit board in the bottom right. What's that for, if you don't mind me asking?

...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

I got my PCMCIA parallel card today. I installed it and ran the Mach DriverTest utility. I hope it doesn't need to be connected to the controller box for that? It worked great! Almost a smooth line. A little +/- 1 pixel jitter.

Then I took the laptop out to the machine and did the configuration Art discussed in the first video. On the good side, when I turned the controller on the fan came on and no smoke appeared. The steppers did lock. When I tried to jog with the arrow keys however... no movement!

Ok, so maybe I need to start with a .xml file. I hope I'm doing this the right way: I ran something called Mach Loader and selected Mach3Taig2000.xml. It loaded but the screen was white. The usual interface screen was gone. So I loaded Mach2Mill_Taig_Xylotex.xml. That one looked better. I added my ports & pins to that.

Still no movement. Any ideas on where to start debugging? I'll post over in CNCZone and other places too.

Thanks,
...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

Hi Athena,

Good news about the motors locking up, now it's just a matter of trying to figure out what isn't configured properly. My first thought was something that took me a while to figure out, is Mach configured to the correct printer port id?

Bob

Yeah..good news on your motors locking up....but what I do not see in your controller, is a breakout board?

That is what the other circuit board (as you asked) is on my controller.....You really need one, as it also will optically isolate your computer from the controller. Not doing that, you can fry the computer....

Another item, like Bob said, have you made sure your port is the correct one, and that you are using the proper pins?

In addition, laptops do not really make good CNC controllers (although they can work). A lot of them use settings in the BIOS, that will shut down the speed, have "speed-step" technologies etc....basically items that make their battery last longer....they are supposed to do that, they are designed for that, However......

CNC machines and controllers, require an uninterrupted pulse train, in order to maintain proper stepper positioning. The things I mentioned above in laptops, can interrupt the pulse train, and can result in a stalled motor, or missed steps. This is also the reason, that if you read the installation of mach3, they guide you to basically striping down completely ALL services windows might use, because again, this can result in a stall. Can you imagine you cutting a few thousand line program, and the computer decides to pop up a screensaver, stalling the motors?

I hope you can see where I am going with this. You need to make sure the computer is properly configured, is attached to a breakout port, and the pins are maped corectly to the right port for mach3.

_________________
-CyborgCNC
http://www.cncguitar.com
https://www.facebook.com/CncGuitar

Dude, I wrote this kind of stuff back in the 80's. I cannot imagine running ANYTHING realtime under Windoze! Props to Art for pulling that off. I knew about the laptop issues going in to this. But it's what I can get my hands on now - and the driver test was great! The laptop was stripped down a bit a while ago by me. I went in and did the things above the "DO NOT do this unless nothing else works" line. I wouldn't be surprised if I had to do further things to it though. The breakout board issue bothers me. More on this later. Glad I used my computer and not someone else's!

Someone on another forum suggested that I have an old DeepGroove controller box.
So I got the DeepGroove.xml file. There was an option to select LPT# I didn't see before. It was set to 3, I changed it to 1. I suspect Art does not address the port this way, however... it did cause me to notice the address. It was FCD8, not FC98. Hey I typed what I wrote down. I just wrote down the wrong number. Several things in the DeepGroove file are different than I had set anyway.

So now I'm talking to the box! The Y axis sounds okish. I was just so happy to see it move. The X axis has issues. I thought it made an ugly screech at first, then I noticed that it was the handle on the end of the encoder vibrating. No real apparent movement of the table but it sounded bad enough that I didn't play with it a lot. I think I might have seen the other end of the screw lurch once or twice.

When I looked at the motor tuning section it had wacky (to me) numbers in there. Steps per unit had some long number after the decimal. I changed it to 32000. I set max speed to 20 ipm. The Y is working with accel of 0.5. I tried that and 0.8 on X and it didn't seem to be a difference.

I'm getting the sinking feeling it might not be a Mach parameter issue... but then what do I know?

I had to cut the session short because I got a rush request, so I thought I'd post my results here before flogging that poor X axis anymore. Or trying the Z with motor/spindle hanging on it.

BTW I can move the X axis by hand ok with the controller off. It takes a lot of turns to go anywhere compared to the old cross-slide table I had on my drill press, but that was a different animal and had larger wheels. I don't have anything else to compare it with though.

Thanks,
...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

It wasn't. And I probably needed a different .xml. I could figure it out... I just couldn't read my own darned handwriting.

...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

Results of the the morning test session:

5 ipm on X didn't help it. It's mostly not moving. Sometimes it will lurch a little forward or backward. I did try the Z axis and it moved quietly and smoothly. I have no idea if they're accurate at all, but at least the Y and Z sound ok.

I realized that I needed a smaller acceleration value in order to get a longer ramp up time. I have it down to 0.3 right now to be ultra easy on the thing but it doesn't help the X.

I don't want people to think my laptop could be the culprit, so I decided to do even more pruning. I thought it was pretty lean to begin with and I'd already defragged the drive. I hadn't turned ACPI off though. That was, um, interesting. It went through and "found" the drivers again for everything. I had to run and get the CD for my new parallel port. It never found one for "Intel Extreme Graphics" 82852/82855 GM/GME graphics controller. It seemed to be doing all the graphics I needed just fine so I killed it. If it becomes an issue I'll find the driver online, but Mach ran ok without.

Did this help? No, it was a step backward because it changed my parallel port address...LOL. What IS it with me and that @#$%ing port address. Now it's FED8. It's moving around faster than my X axis..LOL! Once I fixed that in Mach I was back to were I was. X still sits there and vibrates.

Could it be the motor? The X axis driver on the Xylo board? The X axis turns fine by hand, in fact it almost feels a little easier than the Y. BTW, I'm a girl - not a guy who bench presses 250lbs. ;-> I have twigs for arms. The best I can do now is surf for people with similar problems and see what they did.

What is the G's field? Am I supposed to enter that or is it calculated? The doc I have ignores this field. Searching hits too many things on G code.

The motors hiss a bit. Is this ok? It seemed the X was louder and almost crackled a little... but it could be psychosomatic. ;-> Something's wrong in that axis so I'm more suspicious of it. I'd replace the motor if I had a spare. Wait a minute... do I? I need to check what kind of motor is connected to the rotary table. EDIT: Shoot! It's a shorty. Not the same size as the X axis motor. Oh well. I guess if I really wanted to start taking things apart I could possibly swap the motors on the X and Y axis? Or do you guys think it's still more likely a loose nut behind the keyboard issue. ;->

My Kernal speed is still the slowest (25k?) Is this a problem?

...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

WOW..so many questions being fired....where to start?

Kernel Speed: Whatever value you choose, depends on a couple of factors: How good is the computer you are running it on, and how fast you want your machine to be? The two are related. A higher kernel speed, given a particular step on your motors (like 1/8, 1/16 etc) will be able to drive the steppers at a given speed. If your kernel speed is not fast enough, then you will be limited to how fast your steppers can move or rapid, at a given step resolution. There are a couple of good threads explaining this on the forums (like cnc zone).

-You MUST get the port address down. But you already know that....if you can not map it in mach3, you will not be able to get things done. In addition, I have heard bad things with folks running parallel ports on PCMCIA cards etc. Some, do NOT have the proper voltage levels that the controller expects to see, and hence, many issues arise. Again, search on cnczone, and you will see some of the cards people have used and worked, while others who do not...(Another reason why a laptop is a bad choice....).

-Yes, it is OK for the motors to hiss...that is not an issue. It is current running through their coils, so that is normal.

-Yes, switch the x and y, and see if they behave the same, and vice versa. In other words, if you switch the x motor, and it is now on your Y, and it works normally, the motor is OK!

-yes, use the slowest kernel speed at first. What step resolution have you configured the xylotex board for? (It is on the jumpers).

-Do not worry to much about the G number at first...you can fine tune later....

_________________
-CyborgCNC
http://www.cncguitar.com
https://www.facebook.com/CncGuitar

Hi Athena,

May I suggest that you get these 2 PDF files on Mach3. I think they may help point you in the right direction.
MK

http://www.machsupport.com/docs/Mach3Mill_Install_Config.pdf


http://www.machsupport.com/docs/Mach3Mill_1.84.pdf

_________________
I'm outside looking in, just farther from the window than most.

Andrew Werby from ComputerSculpture.com e-mailed me with a great idea. Just swap the connectors! Much easier. So I did this and found the problem shifted to the Y axis and the X as now ok.

Next step is to try to use the 4th axis to drive the X. I'm trying to figure out what I need to do in Mach. I need to assign the Step Pin # and Dir Pin # from the A(4th) axis to X in my Ports & Pins config? Both my Z and my A have DirLowActive checked. I haven't figured out if I select that on the X axis or not. EDIT- did this and nothing bad happened. Still not sure about which way DirLow should be though. It seems to move with either setting but then I'm just jogging it with the cursor arrows.

As far as "step resolution on Xylotex board"... Doc page says:
Driver is set at 8 microstep mode
Steppers are 200 steps per revolution.
Steps per unit is 32000

I know the laptop was not recommended. The Startech port was one of the more expensive ones and at least 2 other people got them to work with laptops and Mach3. It was recommended as the most like a real parallel port.

Not having a breakout board makes me leery of using a borrowed PC. I'm not saying I think the laptop's great or that it won't cause a raft of problems in the future... I just didn't think it was the cause of my X axis problem. So it looks like I have a blown X axis driver. Knowing my life, the laptop-related issues will be more subtle and harder to debug. ;->

Thanks so much for dealing with all of my questions!
...Ath

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html

Thanks for all your help guys. 'm doing much better now. I have the 4th axis running X for now. I'll get in there to get VREF values when the guys aren't working in that shop torturing wood. Too many sudden extremely loud noises behind my back.

I also discovered that the arrow keys were slow jog and shift-arrows was fast. I had that reversed in my notes. No wonder I thought this thing was so darned slow. I typed G0 X1 Y1 Z1 today and to my surprise the thing took off like a dirty shirt... well, at least compared to yesterday.

Then I got down to some motor tuning. X will take 20 ipm and it sounds happy. Y and Z both ramp up to speed, move at 20 ipm for a bit then "stall", if I'm using the correct term. They make a loud buzzing noise then stop. I reduced the speed gradually and found that 15 ipm is the fastest they'll do. They're short axes so that's a fine rapid speed for me for now. If I ever get running a lot of parts on it I imagine I'll get more impatient. ;->

I'm concerned about the lack of limit or home switches though. All my previous experiences with anything like this were in R&D and due to "early" hardware or software something was always taking off in the wrong direction and trying to injure itself or you. It was very entertaining...LOL.

Thanks again,
...Athena

_________________
_________________________________________
http://home.comcast.net/~ath3na3/Portfolio/Port.html