[Nth]

I’ve been working on backwards engineering a Prolight 1000’s control box so I wouldn’t have to hunt down a controller card for my mill or be forced to use the older software that comes with it. I’ve gotten to the point of a fully functional mill and it turns out to be a surprisingly easy hack once the circuits were traced. Here is what I've found out.

Disclaimer. Use what is posted here at your own risk! Your control box maybe different than mine.

All that is needed for basic functionality is a parallel port cable, knife, and some time. The port has to give out strong 5 volt signals to work. Most of the PCI cards I’ve tested struggle to get 3.5 volts. They simply can't drive the optoisolators on the inputs. Also the cheap parallel port cards do not give out good signals with strong peaks and floors. One I have runs 1.8 volts off and 3.1 on. The problem can be rectified by using transistors, hex buffers or any number of other circuits to amplify the weak signals and properly drive the control box. I simply used the parallel port on the motherboard . All outputs from the control box (lid open, e-stop, home, limit) NEED to be filtered. There is over 18 volts of PK-PK noise in them from the chopper in the stepper drivers and no amount of debounce can help it. If left unfiltered the feedback is enough to heat up the stepper drivers and cause the motors to scream. 0.1-0.2 µf caps are optimal. Anymore and it will just delay signals. The following table is the pinout for the computer connection on the control box. I/O marks if the pin is for input or output. Active notes if the pin is signaling when high, low or both in the case of directions.

Pin  I/O   Desc      Active    Remarks

   1. I    Acc 1        H    Normally high, pull low to disable the Acc1 plug.
   2. I    Acc 2        H    Normally high, pull low to disable the Acc2 plug.
   3. ?    ???          ?
   4. ?    ???          ?
   5. I    Z Direction  B
   6. I    X Direction  B
   7. -    Ground       -
   8. I    ???          ?
   9. I    ???          ?
  10. ?    ???          ?
  11. O    Lid Open     L    Best to wire this in (or to estop). The mill will kill the spindle on low regardless of the control box or computer’s output. Great way to snap bits.
  12. ?    ???          ?
  13. O    X/Y/Z Limit  L
  14. I    ???          ?
  15. I    TTL Output 1 H     Robotic output 1
  16. I    TTL Output 2 H     Robotic output 2
  17. I    Full/Half Step H    Default high is half steps. Pull low for full steps.
  18. I    Y Direction  B
  19. I    Z Step       L    Normally high, steps on low pulse.
  20. I    Y Step       L    Normally high, steps on low pulse.
  21. I    X Step       L    Normally high, steps on low pulse.
  22. I    Enable       H
  23. O    E Stop       L
  24. O    X/Y/Z Home   L
  25. ?    ???          ?

As you can see there are some holes. Notably I'm missing the TTL inputs for the robotic interface. Since I'd just run any robot off the computer directly these other pins were not important to me.

DE-9
Pin    I/O    Desc

   1.   I      + Spindle speed signal
   2.   I      - Spindle speed signal

The spindles works off varying voltage to pins 1 and 2 on the DE-9 connector. Mine came setup for a 0 - 9 volt scale but is adjustable via pot under the back cover of the machine marked sig_adj. It is perfect for a PWM drive signal. Regretfully the pot can’t adjust down low enough for a 5 volt signal to max the spindle. I used an NPN transistor driven by the parallel port’s output hooked to the computer’s 12 volt rail to up the output voltage to 0-12 then adjusted the pot to match. Works like a charm. Only problem is I have no idea how fast the spindle is actually moving so I've just been guessing.

My interface for reference is as follows.

Computer pin    Mill pin    Function
    2    22    Enable
    3    6     X Direction
    4    21    X Step
    5    18    Y Direction
    6    20    Y Step
    7    5     Z Direction
    8    19    Z Step
    9    X     Spindle PWM (goes to a diode->resistor->transistor which switches DE-9's pin 2.)
    10   24    X/Y/Z Home
    11   13    X/Y/Z Limit
    15   23/11 E Stop/Lid Open
    16   1     Acc 1
    17   2     Acc 2

General setup information:
X Travel: 12
Y Travel: 6
Z Travel: 9
Stepper drivers: L297/L298
Steppers: 150 Oz/in 200 SPR
Gearing: 2:1
Thread pitch: 5 tpi
Steps per inch: 4000
IPM: spec'd at 60 but can do 120 if you keep the L298's cool.
Debounce can be left at 0 with proper filter caps.

[Tron144]

Thanks for all your great information Nth. I recently acquired a light machines 1000 and I have a question regarding your interface. Did you splice up a parallel cable to make your connections to the control box?

[Galane]

Some info I've dug up.

Intelitekdownloads.com

Intelitek Downloads

https://docs.google.com/file/d/0Byhq...NhVXBnT00/edit
Intelitek CNC Resource Center.iso - Google Drive

ftp://ftp.robotec.co.il/Techsup/cnc for laptops/

For spectralight ftp://ftp.robotec.co.il/Techsup/cnc software/

ftp://ftp.robotec.co.il/Techsup/To M...ine_resources/ <- content deleted, but still at the docs.google link above

http://www.graco.unb.br/alvares/DOUT...roduction.html
Around the turn of the century, the CADLAB at Purdue University (Calumet, not Lafayette) wrote a cross-platform CAD/CAM package (Windows, Mac and Linux) specifically for the ProLight PLM2000. It didn't directly control the mill (darnit!) but did output G-Code to use with the DOS control software. That G-Code should also work with the PLM 1000 mills. That software WAS available to download for free but it's long gone from any of the school's servers and the professor who ran the CADLAB retired a couple of years ago and his office and class space were cleaned out.

Someone, somewhere, out there, must have downloaded all that and still has it on an old CD-R or backup tape... It would make using the stock ProLight mills so much easier, with a program designed specifically for them.

[Rexford]

Nth,
Thanks for your efforts. I have used your wiring information for the Prolight 1000 and have my mill up and running under Mach3 complete with spindle speed control. I differ in the method a bit as I used a breakout board between the computer and the control box. The breakout board actually takes the place of the missing CDI card that would normally go in the computer. Intelect wanted $1000.00 for the CDI card and I got the breakout board for under $15.00 on Ebay. I use Mach3 and have always been going to try the Prolight software but never have. I have been tole Mach3 is better anyway. The spindle speed as you stated is 0 to 10 volts and the breakout board spindle is 0 to 5. I solved that with a simple op-amp circuit. You get op-amp chips from Radio Shake.

This is a great little milling machine and if one can be picked up for a reasonable price don't let the missing CDI card scare you. I have read where some folks have bought the mills and abandon the control box and bought all new electronics so they could run Mach3, WHY?

Rexford

[Fastest1]

Nth,
Thanks for your efforts. I have used your wiring information for the Prolight 1000 and have my mill up and running under Mach3 complete with spindle speed control. I differ in the method a bit as I used a breakout board between the computer and the control box. The breakout board actually takes the place of the missing CDI card that would normally go in the computer. Intelect wanted $1000.00 for the CDI card and I got the breakout board for under $15.00 on Ebay. I use Mach3 and have always been going to try the Prolight software but never have. I have been tole Mach3 is better anyway. The spindle speed as you stated is 0 to 10 volts and the breakout board spindle is 0 to 5. I solved that with a simple op-amp circuit. You get op-amp chips from Radio Shake.

This is a great little milling machine and if one can be picked up for a reasonable price don't let the missing CDI card scare you. I have read where some folks have bought the mills and abandon the control box and bought all new electronics so they could run Mach3, WHY?

Rexford

Because we bought the machines before this information was available or we were too new to understand what we were doing. New electronics have support. 1000.00 card for antique electronics makes people move on.

Now with the sharing of info and hard work by those who understand it. The community will hopefully benefit.

[Alan]

To: "Nth"

I'm the proud owner of a ProLight 1000 as of yesterday, and would like to get it running under Mach3. I have two other machines CNC machines using Mach3 and a background in electronics, so I'm decently familiar with the software and what's required.

Now, I COULD just build a whole new controller and in fact might even do so one day, but for now, I'd like to use the existing controller. The information in your post gives just about everything I need to get started, and I thank you for that!

I do have one question though; is it possible to control the spindle On/Off by a signal on the controller's 25 pin connector?

I see from your pinout that you're using the PC's pin 9 and some additional electronics to generate a 0-10v signal from Mach3's PWM output. Supplying 0 volts should therefore cause the spindle to stop.

But I'd like a more positive option, one that removes power entirely.

Is there a way to do this?

Thanks for any thoughts.


Alan

[Alan]

I’ve been working on backwards engineering a Prolight 1000’s control box

To try to fill in a few of your question marks, I've had a look at my new-to-me controller and have just a couple of things to add about the DB25 connector labelled "Computer".

1. All the inputs to the box are (potentially) optically isolated. There is a jumper that connects (almost) all of the LED anodes to either an internal source of +5 volts or pin 22 of the DB25 connector.

There are two implications to this jumper.

a. With the jumper set to the internal +5, the optoisolators are powered internally and so there is no actual isolation. Any noise travelling along these lines will show up inside the box, maybe resulting in missed steps, possibly damaging the electronics at the other end and so forth. However, this is the required setting for use with just a parallel port. (More on this in a second.)

b. With the jumper set the other way, such that the anodes of the LEDs in the optoisolators are connected to pin 22 of the DB25, the input signals are isolated.

This is a good thing! Genuine optoisolation is good.

However, a strong source of +5 volts must be delivered or the LEDs in the optoisolators will not transfer the bits properly, each and every time. I can certainly see missed steps or erratic behaviour if pin 22 is driven from a logic signal, i.e., a pin on the parallel port. It would be pretty rare .. extremely rare .. for a parallel port to source enough current to power the LEDs for a dozen signals. (Modern computers .. the last of those that actually came with parallel ports .. likely couldn't drive a single LED.)

As stated above (twice), optoisolation is a very good thing and if a person wants to run their machine off a parallel port using Mach3 or LinuxCNC and have isolation, there are two "fixes" to the problem.

The first is to buy a breakout board that includes optoisolation. There's a good number of breakout boards to choose from, but sadly, only the expensive boards deliver optoisolation.

The other option (which I've used in the past for other devices) is to supply +5 from the computer itself. The easiest way to do this is to use a USB cable attached to the PC, cutting the other end off and running the 5 volt wire (usually the red one) to the +5 input, which in this case is pin 22.

(By the way, this means that pin 22 is not actually a logic input, it's more of a "power" input. It's still OK to consider it an enable input but only with the jumper set to external power. With the jumper set to internal optoisolator power, pin 22 does nothing that I can see.)

Sadly, there's still one problem with this scheme. For true isolation, the "ground" connections must be separated as well and with the ProLight controller, they're not. Pin 7 of the DB25 connector is connected to the "ground" (DC zero reference) of the ProLight controller and must also be connected to pins 18-25 of the parallel port of the PC. This could result in grounding problems (of a sort too technical to describe here). Thankfully, those problems are minimized if the computer and the controller are plugged into the SAME duplex outlet.

If they cannot both be plugged into the same outlet (or if weirdness results even when they are), a breakout board with genuine optoisolation is the only option.

2. It appears as if Pin 12 and Pin 25 are outputs. They are connected to the collector of an optoisolator with a pullup to +5, the diode inputs of those devices going somewhere unknown.

3. It appears as if Pin3 is an input, although for what purpose, I can't say.

4. It appears as if Pin 10 is not connected to anything.

Hope this helps someone else who, like me, got a deal on a neat machine and wants to use modern software but doesn't necessarily want to build a whole new controller.

Alan

[Rexford]

Alan,I have a Prolight 1000 and am running it with the original black control box and using Mach3. I do not have the pro light control card. I am using a $15.00 break out board from Ebay in its place. I use a little OpAmp circuit to change the break out board PWM 0 - 5 V to 0 - 10 V for the spindle speed. Connections: The Computer to breakout board via parallel cable. From breakout board to Black control box via a parallel cable that has had one end cut off and wired to the breakout board with the connections found else where in this thread. Then that parallel cable is plugged into the Black box. This works only on 32 bit computers. I have just last week updated to a 64 bit computer by placing a Smooth Stepper in be tween the computer and break out board. The Smooth Stepper operates with a USB cable to the computer. If you choose this route I would recommend getting the mill running with a 32 bit and break out board first then Change to the 64 bit with the Smooth Stepper.If there is interest in what I did I may try to put some drawings , pictures and detail on a web sight some where.While I am here I have a Prolight 2000 which has servo motors I would like to switch to mach3 if anyone knows how to do that. It run now on the old DOS system. Rexford

[Alan]

Alan,I have a Prolight 1000 and am running it with the original black control box and using Mach3.

Thanks for the reply!

I'm certainly making progress getting my setup going. I've discovered that in the case of my box at least, there's a bit more required that what was posted by Nth above. (BTW, thank you Mr. Nth, your work has cut hours and hours off my task.)

Late last night I discovered that the RESET line on the L297 drivers is connected to a pin on the DB25 connector (3 or 14, I just forget which). I've also discovered that pin 22 does double duty, depending upon a jumper setting. In one setting, it requires genuine 5 volt power (from an external source), in the other a logic "1" will do. But in both cases, it feeds the ENABLE input of the L297s.

What is slowing me down a wee bit is the fact that there are numerous jumpers on the controller board, some labelled "Old Controller", some labelled "New Controller", some not labelled at all. Sadly, I think my controller falls half ways in between as some (but not all) of the OLD jumpers are set and some (but not all) of the NEW jumpers are set.

Today's task is to try to figure out, if I possibly can, what these various jumpers do.

I have to say, there is an AWFUL lot of logic on this board, for reasons not clear to me...

The Smooth Stepper operates with a USB cable to the computer.

I suspect a Smooth Stepper, or something like it, is in my future, as some of my difficulties with this project are self inflicted.

I have four other machines run by Mach3, all using controllers I designed and built, so I do have some familiarity with all this. However, all these machines are desktops, as the use of notebooks is strongly discouraged by the Mach3 programmers. But I wanted to use a notebook, at least to try this all out, simply for space reasons; a notebook makes the most sense given where the machine is temporarily located. Sadly, the notebook I chose, an older one that I did not think would have problems does have a problem, albeit a very very weird one.

The Prolight controller effectively pulls all inputs to logic 1 internally. What the computer has to do to make a pin "not" 1 is to ground it, generally a simple task. Annoyingly, in the case of this controller, grounding the pin is not the same thing as supplying a logic 0.

The difference is a subtle one of voltage and current. In the modern age .. everything since 1980 or so .. logic signals are voltage levels. In the case of a 5 volt system, a signal above 2.5 volts is seen as a logic 1 and a signal below 1 volt is seen as a logic 0, with the area in between being an undefined state.

In the olden days though, current flow is what determined a logic level; so much current flowing out of a pin or not flowing out of a pin is what was important.

Most modern electronics interface nicely to either; it's a simple matter to "fake out" modern voltage sensitive logic to work with older current flow logic.

But the Prolight is not one of those devices! Since the inputs drive the LEDs inside optoisolators, the computer must genuinely sink current, enough current to turn on an LED and my little notebook won't! It only took me half a day to figure that out because I've never heard of such a thing. Sure, there's devices that don't want to source current but I've never ever run across a thing that didn't want to sink current .. until now.

Well, I've got a decent stock of parts, and I found a decent driver chip and wired it up in between the computer and controller. Now the notebook would indeed "ground" the pins as required. But what it would not do is release the "ground". Turns out that the notebook, when outputting a logic 1, produces a 3.5 volt signal. This is well above the 2.5 volt requirement, and so this is a valid digital logic signal. My driver chip though, is a completely analog device, and needs the signal to go all the way to 5 volts! At 3.5 volts, it's still sinking enough current to turn the LEDs on.

Figuring that out was the other half day.

I'm going to guess that if I hooked this up to a desktop computer, most of my problems would go away. The notebook printer port is likely just fine for driving an actual printer, it's just not OK for this slightly left of center task. I suppose it's time to stop cutting bait and actually fish, which is to say, time to build a proper breakout board with logic input on one side and current sink and source on the other.


Thanks again for the reply.


Alan

[carman2002]

Rexford, I just recently picked up a Prolight 1000 and am looking to do this retrofit using a breakout board. I am interested in more of the detail on how you hooked up the connections from the computer to breakout board to black box. During your retrofit how did you wire the spindle motor. Did you run a 9 pin connector from the computer to the breakout board or was the spindle IO's somehow run out of the 25 pin connection from the parallel port on the computer? I was curious if you could provide any information as to which break out board you purchased on ebay as well. I am new the CNC retrofitting and am excited to get my machine running again, and appreciate all the very useful information that everyone has already posted on this topic.

[Galane]

While I am here I have a Prolight 2000 which has servo motors I would like to switch to mach3 if anyone knows how to do that. It run now on the old DOS system. Rexford

Send me an email to g_alan_e at yahoo and I'll send you a bunch of technical documentation and software for the Animatics servo controller used in most of the Prolight 2000 mills. A person at Moog Animatics who stayed with the company after the merger found it all on an old backup drive. There's a lot that controller can do which wasn't exploited by the DOS software. One such feature is storing g code in its RAM then running it as a canned cycle without an external computer.

With the info it should be possible to add support to Linux CNC and other software. The thing with the 2000 is it's a "smart" machine. The control computer just feeds it g code and monitors the status data the mill sends back. That's why it can run from pretty much anything that can run MS-DOS and had a serial port.

So it *should* be relatively simple to write Windows control software for it since the software doesn't have to do any fancy math, that's all done by the Animatics box in the mill.

[carman2002]

Alan,I have a Prolight 1000 and am running it with the original black control box and using Mach3. I do not have the pro light control card. I am using a $15.00 break out board from Ebay in its place. I use a little OpAmp circuit to change the break out board PWM 0 - 5 V to 0 - 10 V for the spindle speed. Connections: The Computer to breakout board via parallel cable. From breakout board to Black control box via a parallel cable that has had one end cut off and wired to the breakout board with the connections found else where in this thread. Then that parallel cable is plugged into the Black box. This works only on 32 bit computers. I have just last week updated to a 64 bit computer by placing a Smooth Stepper in be tween the computer and break out board. The Smooth Stepper operates with a USB cable to the computer. If you choose this route I would recommend getting the mill running with a 32 bit and break out board first then Change to the 64 bit with the Smooth Stepper.If there is interest in what I did I may try to put some drawings , pictures and detail on a web sight some where.While I am here I have a Prolight 2000 which has servo motors I would like to switch to mach3 if anyone knows how to do that. It run now on the old DOS system. Rexford

Rexford,

Would you be able to provide any of your drawings or pictures of your breakout board retrofit using the original controller box? DId you have much trouble setting up the Mach 3 parameters in the software?

[Mike13815]

Just wanted to say thank you for this writeup, it made everything way easier getting mine running as everything was fried before I bought it due to a water pipe breaking.

I gutted the electronics and set up GRBL with an ESP32. After everything, replacing that control box was about $50. Replacing the spindle controller was about $40.

I'm already milling out the hole in my remote control for the hand wheel. This is a wonderful machine and I'd suggest to anyone else that is looking to get one seriously consider it, as it's going to outperform any entry level machine, at harbor freight prices. Just be prepared to do some wiring to give it modern, universal control.