Reply with QuoteI just wanted to say nice job.
And ya I a, surprised this hasn't been done before too.
Chris M
Running a SpectraLight 0200 from the parallel port
Typically the SpectraLight 0200 mill uses a custom ISA or PCI card to interface the computer with the controller. These cards are not supported under EMC2.
I've successfully interfaced the factory controller with a standard parallel port, eliminating the need to use the OEM interface card. Currently I have control of all three axis, as well as spindle on/off control. I have cut a few pieces from foam and the results were perfect.
Details are available on this page.
I plan to extend the cable to support ESTOP (keyboard and mushroom switch) as well as spindle speed control. Currently speed is controlled with the manual switch.
Similar Threads:
I just wanted to say nice job.
And ya I a, surprised this hasn't been done before too.
Chris M
Have you tried the mod yourself? I'm anxious to gather feedback.
Hopefully soon I will be able to give some feedback about the effectivness of this process, as I am working on trying to work out weather the X axis stepper is dead during D&T.
Well, good luck and I hope it goes well. If you need any readings from a working system, let me know.Originally Posted by rexhunt
Circuit board
I've received feedback from several people. On the whole the mod seems very reliable. There is one issue, and it poses a very significant safety risk. The parallel port, when not being controlled by controller software, often defaults to a state that causes the spindle to activate.
Put another way, the spindle may turn on as the system boots, or at any time the controller software isn't running.
This has significant potential to injure someone. Especially if you turn your computer on, and decide to change tools while waiting for the computer to boot.
I am testing a circuit board which will eliminate this possibility. In the meantime, turn the controller box off when changing tools.
that is usually why the interface (or a simple break out board) has what is call a charge pump - or watch dog. Unless it is getting constant signal from the control software - the board will not enable. (like a 10khz signal or such)
sam
Sam,
Thanks for that insight. I planned to place a microprocessor (attiny2313) on the breakout board. However, I didn't realize the significance of the charge pump. This bit of info makes things MUCH easier. Thanks again.
Shannon
When are you going to teach the rest of us how to interface with say a Dyna 2400? Maybe so I could drip feed it? Actually I would rather run Mach 3 directly into it but I dont know what is possible. I do know my machine works but I dont have a way of communicating with it in its present arrangement. Tips? I had a Spectralight and would have been glad to try your mods but I converted it over before your work.
hi,
My name is Rashmi,
I have a cnc of fanuc om of Technics.I have a repetative problem that my cnc give troble while starting . It gives the error 401 vrdy off servo problem with the alarm light blinking . I am very much suffered from this problem.My consultant also can't solvw the problem.can anybody help me?
I had a huge advantage, the mill and the legacy control card and software were all working perfectly. Reverse engineering it was trivial because all I had to do was attach an oscilloscope and observe the signal changes. Without a working set-up, I'm not sure this would have been possible.
The irony is that this project came about as I was starting a conversion over to a commercial controller. I had no intention of attempting to build a cross-over cable for parallel port operation. I was of the opinion that if it was possible someone would have already done it. In fact I located several posts here where people stated they planned to do just that and then nothing. In my mind it must be impossible, or well beyond my feeble skills, or it would have been solved already.
I had my eyes set on a gecko (can't go wrong with them from what I can tell) controller and was comparing steppers. I was curious if I could salvage the power supply and save a little $$$. So I attached the oscilloscope and after a few minutes of observing signals on the cable it became apparent that a parallel port could reproduce the signals.
That Dyna looks like a nice little mill. If it uses steppers a conversion shouldn't be too difficult. But you know that alreadyI'm new to machining, and purchased mine to drill PCB boards and cut openings in enclosures. Now I've been bitten by the bug.
Just a few minutes ago the FedEx guy handed me a box of components from the mail order electronics company (Jameco). Time to build a circuit with minimal safety features like charge pump (awful name) support and the all important blinking lights.
Of course I'll release the design for others to use / improve.
Good luck with the Dyna, update me with progress. That looks like a great little mill.
Hi Shannon!! Me and a couple of friends are doing the exact thing as you. So far we can control the speed of the spindle trough PWM but we haven't been able to move the steppers(
We built a board (the breakout board) that's looking pretty good so far.
The features we're trying to implement are:
PWM to DC control of the spindle .............................. OK
10 3mm red leds indicating speed of spinlde................. OK
Leds indicating direction and step ........................ OK
Only problem is that we cant seem to get the steppers to move. Any chance you could post a picture of your controller board? We've noticed a few jumpers missing on the board and we were wondering if that could be the problem
Here are some pics of our work so far pic1 pic2
There's one with your name on it Shannon
Difficult is a relative thing!Either way you did a great job and I am sure others will appreciate your work. Even better you shared it. Cool place this is. We will all be watching for future updates.
I had a huge advantage, the mill and the legacy control card and software were all working perfectly. Reverse engineering it was trivial because all I had to do was attach an oscilloscope and observe the signal changes. Without a working set-up, I'm not sure this would have been possible.
The irony is that this project came about as I was starting a conversion over to a commercial controller. I had no intention of attempting to build a cross-over cable for parallel port operation. I was of the opinion that if it was possible someone would have already done it. In fact I located several posts here where people stated they planned to do just that and then nothing. In my mind it must be impossible, or well beyond my feeble skills, or it would have been solved already.
I had my eyes set on a gecko (can't go wrong with them from what I can tell) controller and was comparing steppers. I was curious if I could salvage the power supply and save a little $$$. So I attached the oscilloscope and after a few minutes of observing signals on the cable it became apparent that a parallel port could reproduce the signals.
That Dyna looks like a nice little mill. If it uses steppers a conversion shouldn't be too difficult. But you know that already
Just a few minutes ago the FedEx guy handed me a box of components from the mail order electronics company (Jameco). Time to build a circuit with minimal safety features like charge pump (awful name) support and the all important blinking lights.
Of course I'll release the design for others to use / improve.
Good luck with the Dyna, update me with progress. That looks like a great little mill.
PCB Board
I assume you have visited my site (http://build.spaceopera.org) already.
I just put the finishing touches on this section: http://build.spaceopera.org/site/ada...ard/index.html It provides an overview of the PCB board that I have been working on.
Curiously, I do not have spindle speed control working. I have not even started on spindle control. The current cable supports a full range of speeds (0-30+ ipm) on my machine on all axis. But no spindle control. I am VERY eager to get my hands on a full schematic for your motor control circuit.
Do you get ANY movement on ANY of the axis? Even a bit of chatter without any movement? Also, does the controller box "hum" when you press F2 in EMC2? I would like to trade schematics. I see several DIP packages on your design, I assume they are optical couplers and a microprocessor of some sort. Please provide details!
Hopefully you will knock this little hurdle out of the way and have the mill running in short order. Feel free send e-mail (shannon dot haworth @ gmail [no spam please] dot com).
Cheers,
Shannon
Last edited by shannon_haworth; 08-04-2010 at 03:52 AM. Reason: Grammer corrections
Hey Shannon!
We finally got it to work! Turns out an inverter gate was burnt, no big deal! Now we've found that one of the axis is no working due to a burnt L298 but we'll fix that in no time.
About the spindle control it's really easy to implement! EMC2 gives you the option to output a PWM signal just for that. Then, what you need to do is transform that digital output into a analog one that goes from 0 to 10volts (I think it's a standard range).
Here's the schematic of what you need
http://www.picbasic.co.uk/forum/showthread.php?t=245
You're gonna need an external power supply of 12v if you wanna go up 10v.
Let us know if you have any questions!
Cheers,
Fede
Good job hunting the gate down.
I've found that after the conversion the reliable top speed for rapids went from 30 to 42. What speeds are you seeing?
Please post a "it works" comment on my site. It reassures users who are on the fence.
I knew about the technique of converting a PWM stream into a analogue voltage range, however being a software guy I got stuck pretty quick. I'll give that link a look, with luck I'll be able to wrap my head around it.
I'll be sure to let you know when/if I'm stuck.
Shannon
Does anyone know the answer to the question I have posed in this thread: http://www.linuxcnc.org/component/op.../lang,english/
At some point I forgot to update the site with new EMC2 instructions. Thanks for pointing this out. I have updated the site with complete instructions for EMC2. The instructions are here: http://build.spaceopera.org/site/emc...ion/index.html.
Let me know if I can be of any assistance. And PLEASE be careful using the adapter cable. There is a serious safety issue where the spindle will turn on when linux is booting, and after leaving EMC2 (if it crashes for instance). I am working on a logic board which will eliminate this issue. Please do NOT allow students to use the mill with the mod in place.
Thanks,
Shannon
PCB
Is it possible for me to get a copy of the schematic so that I can make a circuit up that will allow me to cut a PCB with the mill?
I have a feeling that the port on my pc is only putting out 3v or similar...would that affect the circut?
The PCB board that I'm currently using is a simple affair that replicates the adapter cable, pin by pin. On the site I mention that I've been building a board that incorporates charge pump support. I even posted a picture of my first failed attempt.
However, I may never build the board. I've recently discovered this off the shelf board (a charge pump detection board). Combine it with a pair of parallel port breakout boards from ebay (the ebay store store MDFlyStore and a complete solution can be created for about $40.00 USD.
Note, you will need to purchase separate male and female breakout boards. They sell a breakout board that combine both connectors, however the ports are directly wired together, it won't work. Although it is VERY nice for sniffing the signals with the oscope. I learned this the hard way.
The components are on order and I will update the site once I confirm everything works.
Concerning voltage, 3.3 volts could very well be a problem. The controller box expects 5 volts. That is what the controller card creates, I would not expect 3.3 volts to work.
