[WayneC]

I just spent a day fussing with my CNC, and I can't figure it out. I need some help so I can make a more logical troubleshooting effort tomorrow.

My problem STARTED with an intermittent coil on one X axis motor. The motor would travel one direction, but not the other. I traced this to the connectors at the G202 driver, but I can't tell if its the connectors, me fussing with the connectors, or something intermittent in the G202. When I press on the connector or move the wires just right, the G202 will light up and work.

BUT somehow along the line I've caused another problem- now even when I can get the G202 to the ON state, I can't move the motors connected to it. DRO indicators in Mach2 show movement, but the motors don't turn either direction. Indicator lights on the control box show step pulses, but not direction changes.

In Mach, I checked outputs ports and pins, and these are correct. Mach seems to be working OK. I can't find any settings that are changed- what should I be looking for?

If you can help me to troubleshoot this, it would be great. What I think I need to do is:
1) Check voltage at the parallel port on the X step and dir pins to make sure Mach is sending voltage to the controller
2) Check voltage at the Gecko input to verify input volts (what are these supposed to be- +5V?
3) Check power to the G202s, and compare output of the working motors to the not working one

If the Gecko has power, and input volts, it should produce output, right? If not, I have another Gecko on the way to swap. But my past experience is tht its not likely to be the Gecko. Probably some wiring or Mach 3 SNAFU.

I'm not clear on what to look for in Mach if I don't have output Step/Dir voltage. Other than ports and pins, what controls this?

Thanks for any suggestions. I'm methodical, but I'm running out of what things to check here.

Wayne C
WhiteSalmon, WA

[Richards]

Wayne,
Are you using a break-out-board, or, are you directly driving the Gecko stepper drivers from the parallel port? The G202 requires 15mA per input. Some parallel ports are not capable of that much current.

My G202 stepper drives are connect to Mach 3 via a PMDX-122 BOB. Common is connected to +5V, so an active signal is less than 0.7V, usually very near 0V. Take a meter reading of the step signal going to a G202. It should be either +5VDC or 0V. Change the direction and the signal should be the opposite polarity. Write a line of G-code to move an axis a substantial distance. With an oscilloscope, you should be able to see the step pulses going low for at least 2uS per pulse. After acceleration has finished, the pulses should be evenly spaced.

Some connectors on some Gecko products were faulty. (I believe that Gecko offered replacement connectors.) If you suspect that you have a bad connector, use some small hook clips and attach the jumper's hook directly to a G202 terminal post. Connect the other end of the jumper to your BOB. You'll have to repeat that process for all the terminal connections. Be careful and take your time.

Remember that the step/direction signals are just pulling some current through an opto-isolator. 15mA turns the opto-isolator on. Absence of current keeps the opto-isolator off.

If you have the G202's Common connected to +5V, you should be able to tap a ground lead on the step terminal and get stepper movement. It will take a lot of tapping to see much movement, but you should be able to hear the stepper step each time you touch the ground lead to the step terminal.

[WayneC]

I'm Pacific Coast time, so I'm just getting rolling now (sleepyhead). I'll start working on this shortly.

I use a breakout board that is an Ascension 1000. I believe this was made by Bob Campbell for Custom CNC, but I can't confirm that. At any rate, it seems to be a well made board, with plenty of power supply and good connections, etc.

One starting question I have is a basic one. If the signals are NOT present at the G202, how do I check that Mach3 is outputting the signals? At the step/Dir pins indicated by Mach3 Ports/Pins on the parallel port?

If I'm understanding this correctly, if I get output at the parallel port pins, and not at the Gecko, then the BOB is suspect, correct? (or wiring)

If I get Step/Dir at the G202s and no motion, that would suggest that the Gecko or its connections is defective. I'm getting a replacement G202 tomorrow, so I can just wait till that comes in and swap one out.

Finally, are there settings in Mach3 that can affect the output. I have correct function on Y and Z axis, and it appears that Port/Pin outputs are correct for the X axis. Are there any other settings that could change this?

Thanks so much for your help with this. I can troubleshoot OK, but I don't fully understand the operation and interaction of these pieces, so your help is invaluable.

Wayne

[Richards]

I'm assuming that you don't have an oscilloscope or a logic probe to check signals. If that's the case, you can build a simple LED tester to use in place of a Gecko (to verify that the BOB/parallel cable/Mach 3 are all working properly).

1. Disconnect the X-axis Gecko G202 from the Break-Out-Board
2. Get a standard LED and a 220 ohm resistor (1/4 watt is fine).
3. Connect the Anode of the LED to +5V.
4. Connect the Cathode (usually flat portion of LED is Cathode) to one end of the resistor.
5. Connect the other end of the resistor to the X-axis step terminal on your break-out-board.

If everything is properly configured and if a signal is received by the BOB, the LED should turn on when steps are being sent to the X-axis. The brightness of the LED will depend on the pulse rate, but you should see some activity on the LED.

If the LED does not turn on, then you'll need to check your equipment and/or your Mach 3 settings.

With my PMDX-122 BOB, I have parallel port pin #3 supplying the step pulse for the X-axis. In Configuration -> Ports and Pins -> Motor Outputs, the X-axis is enabled (green check mark), the step pin is set to #3, step low active is checked (green check mark), and the step port is set to 1 (which matches my parallel port setup info).

Except for the step and direction pin numbers, you should have identical settings for the X, Y, and Z axes. As long as the motors are identical, you should be able to switch the G202 terminal headers between motors to easily check whether a motor has a problem or whether the G202 has a problem. Of course, verify that the power is off and that the power supply has drained before exchanging G202 headers.

To make life easier when wiring things together, you might consider using ferrules (www.automationdirect.com). Sometimes a single stray strand of wire can bridge two terminals. If your eyes are as old as mine, that could cause a serious problem. Ferrules help keep those stray strands from straying.

[WayneC]

My control box has LED step/Direction indicators. Do these work to detect the presence of signal?

Also, what sort of logic probe would you recommend? I can see that I may need to take at least this step in test equipment. I have a nice old Simpson meter, but it can't follow logic pulses at .7 V.

I'll assume that the LEDs on the box are indications of the presence of signal, and see what happens when I swap plugs. Thanks for the support, Mike.

I'm wondering if I have a bad parallel port in the PC. I think that Mach is correctly configured, but the X motors don't work. If Mach is putting out signal, but its not triggering the LEDs then the parallel port is suspect, right?

Wayne

[Richards]

The Ascension box LEDs should indicate axis activity. (I have never seen an Ascension box, but your description makes me think that the LEDs are there for a purpose.)

My logic probe is something that I bought at Radio Shack around 1975 and it has worked perfectly for checking the occasional signal. Almost any logic probe should work as long as it is compatible with TTL and CMOS logic. Jameco and Digikey both offer one for about $40.

The parallel port OR the parallel port cable OR the BIOS setting for the parallel port could each cause problems. Right now, with the G540 in place, I have the BIOS configured for EPP mode; however, if I remember correctly, before installing the G540, I just had the BIOS configuration set to standard parallel port. With a helper, you can check continuity on the parallel port cable. Some cables only have a minimum number of pins connected internally. The 'enhanced' cables (that HP required for their earlier LaserJet printers) are the type that I use.

[WayneC]

Mike,
I measuree the voltages at the Geckos step pin. On the X axis 202's I read -5V from common to step. I don't get a change when I change directions using the jog key. Also, when I measure the Z axis (works correctly), I don't read the -5V, but I don't actually read any voltage on my Simpson, even when I change direction. As I say, this is a working axis. So I'm confused by this result.

I'm in the process of installing Mach 3 on another spare computer, to try and eliminate the parallel port as a potential problem. I did swap motor connectors, and the problem travels with the swap, meaning the motors and their cables appear fine.

Why am I getting no volts from the step pins on the working axis, and -5V on the step pin of the NON working axis?

Thanks for your patience here.

Wayne

[Richards]

I'm looking at the instruction manual for Bob Campbell's version 1.8 BOB. The step and direction pin-outs are different than the PMDX-122 that I use.

X-axis, pin 2=step, pin 6=dir
Y-axis, pin 3=step, pin 7=dir
Z-axis, pin 4=step, pin 8=dir
A-axis, pin 5=step, pin 9=dir

Assuming that your meter's positive lead is Red and its negative lead is Black, put the Red lead on the +5V screw terminal and the Black lead on the direction terminal. When you jog, one direction should show nothing, i.e. direction is at the same voltage (or nearly the same voltage) as +5V. The other direction should show +5VDC (or nearly +5VDC). If you attach the Black lead to Common, you will get -5V when you place the Red lead on the step or direction pin and the step or direction pin goes active.

[WayneC]

Mike

Thanks for the follow up. I replaced the computer, installed Mach 3 on a new machine and verfied the ports/pins etc. On the new machine I have exactly the same problem, so its not likely the old computer's parallel port.

I have a small difficulty with your instructions. My G202s don't have a +5V terminal. There is a common and a step and direction terminal, but no term labeled +5v. Where should I be reading voltages on the G202?

Also, the Campbell BOB pinouts are the same as mine.

Wayne

[Richards]

On a G202, the Common terminal is connected to +5V, so it is always +5VDC. (In the beginning, when computer logic was young, most TTL chips could SINK current much better than they could SOURCE current. Sinking current means that you pull an Input Line LOW to turn it on. ) So, on the G202, the Common terminal is connected to +5V and the break-out-board pulls the line low to turn the line on. The way that I measure voltage for that type of input is to put the meter's Red lead (positive) on the Common terminal and the meter's Black lead (negative) on the input pin. If you see about 5V, then you know that the input line is On or Active (Active Low). If no voltage appears, then the input line is Off or Inactive (Inactive High voltage).

I know that referring to Common as +5V is confusing. Normally Common is Ground or 0V, but on the G202, Common is the Common voltage source for the pins, or +5VDC.

It might help to refer to the G202 manual:

(TERM. 8) DIRECTION Connect the DIRECTION line to this terminal.
(TERM. 9) STEP PULSE Connect the STEP line to this terminal.
(TERM. 10) COMMON Connect +5VDC from the STEP / DIRECTION generator to this terminal.
These inputs are optically isolated from the rest of the drive. Term. 10 is the common anode connection for the opto-isolators and must be connected to the +5 VDC supply of your indexer or pulse generator.

These inputs are meant to be driven by standard TTL logic or other driver capable of sinking 16 mA of current. The minimum logic “0” time is .5 uS while the minimum logic “1” time is 4.5 uS. Microstepping occurs on the falling edge of the step input.

[WayneC]

Mike
Got it. Common is 5V. I need to step back for a few hours or I'll loose it and pull the rest of my hair out. I'm getting inconsistent results, or there are some other variables that are at play here.

I measured the volts from common to the direction term, and I get 5V, and then 0 V when I change direction. But when I try the same test on the the working Y axis, I get 5V when I jog, but it stays at 5V when I change direction. I will have to try again later, maybe I was not touching the right terminals, but I don't think so.

Also, somehow I've changed settings in Mach 3, so I have little confidence that its not something in Mach 3's setup. It somehow changed limit switch settings on me, which I found and fixed. But perhaps some other settings have changed, or Mach 3 loaded the wrong set file.

Finally, when I restarted the machine a short while ago, one of the two X axis motors was turning even though there was no command to move- no program, no GCode, no jog. (On my machine I run two X axis motors from one Mach 3 output, the X output, and feed that to 2 G202s.) It was just spinning, but only one motor when 2 G202s are given the exact same signal. I made a change in the General Config file to a setting that somehow switched itself on, and the axis quit moving. The setting was the "Z is 2.5 on output 6". Since my X axis motor outputs Direction signal on pin 6, I can see how this could cause problems. Somehoe I must have loaded the wrong screen set- unlikely, but possible.

Frustrating, in case you can't tell that. Thanks so much for your help. I'll get a logic probe tonite and try again tomorrow. Maybe I need to get someone to check my XML file for stupidness as well.

Wayne

[Richards]

Wayne,

It might be a good idea to have another Ascension user send you his Mach 3 xml file so that you can compare it to your xml file. If the settings are messed up, the results will be unpredictable.

I'm concerned about motor movement with no pulses being sent to the BOB. Stepper motors require one step pulse per motor step. Somehow, something is set to produce pulses. I'm wondering whether you somehow have a spindle setting mixed up with an axis setting.

[WayneC]

Mike, you are right. Actually, I have a disk with the unchanged original XML file, and I'll be able to copy that to the computer and see what that yields. I'll also spend a few minutes rechecking my troubleshooting results to see what I'm getting.

Mike, thanks so much for the help. I'll let you know what happens.

Wayne

[WayneC]

I'm getting closer to resolution. I removed the CNC4PC speed control apparatus, and turned off the spindle settings in the Ports/Pins config panel. When I do this, I can get the X axis to operate correctly. I did also replace a Gecko 202, but I'm not convinced there was anything wrong with it.

I did also find a loose wire, which was probably causing some of the intermittent stuff. The new Gecko plugs fit much tighter, so I think that may resolve that problem as well.

Now I just need to see if I have a problem with the CNC4PC stuff, or with the Port/Pin config for the spindle. I may have caused a spike when I was poking around on the Gecko plug that might have damaged the speed control board, or I might have screwed up a setting somehow in Mach3.

THanks Mike for your help.

[ehalpern1]

I'm new to CNC and to the CNCZONE community.

I recent bought a hotrod ShopBot PR. The only remaining ShopBot hardware is the uni-strut frame. The system runs Mach3, has two x motors and uses an Ascention 1000 controller.

I haven't had any problems with it yet, but I have no documentation on the Ascension 1000.

Is there a download somewhere for the Ascension 1000? OR, can someone scan and PDF their Ascension manuals?

What do you do if an Ascension 1000 blows up?

As an engineer, I look inside my controller and see four Gecko 202s, two large transformers, and a circuit board that has power supply hardware next to parallel port decoding circuits.

The part that weirds me out is the CMOS logic board on the far side of the case from the parallel port decoding circuits.

I would feel better knowing exactly what I have. Schematic documentation on the large printed circuit board and/or the entire controller package would be awesome.

Thanks in advance.
Eric (Wayne, PA)

[Richards]

Take a look at Bob Campbell's, campbelldesigns.net web site.

The parts are basically generic. You can use the PMDX-122 bread-out-board as a replacement for the one that you have. You can build your own power supply, or buy one from www.antekinc.com (or other supplier). The Gecko G202s come from Geckodrive. Software is Mach 3.

If you really want to make that machine do all that it can, replace the motors with Oriental Motor PK296A2A-SG7.2 motors. Wire those motors half-coil. (Half-coil = Black wire to A-coil, Yellow wire to /A-coil, Red wire to B-coil and White wire to /B-coil. Tape off individually and insulate the Green wire and the Blue wire. The Oriental Motor SG gear box cannot handle the full torque of the PK296 motor, so wiring the motor half-coil will give all the torque that the gearbox can handle and all the speed that the motor can produce.) Power those motors with a power supply giving 35VDC to 40VDC 12A that has 20,000uF to 30,000uF capacitors and you'll have a real "Hot Rod" Shopbot. (Visit MechMate.com to see how many others have either modified their Shopbots or build their own machines from the ground up. Gerald Dorrington has provided free plans for anyone who wants to build their own machine.)

[Gerald_D]

As an engineer, I look inside my controller and see four Gecko 202s, two large transformers, and a circuit board that has power supply hardware next to parallel port decoding circuits.

The part that weirds me out is the CMOS logic board on the far side of the case from the parallel port decoding circuits.

I would feel better knowing exactly what I have.

Eric, I very much suspect that these are your gray areas:
http://www.campbelldesigns.net/Combo-board.php

That "CMOS logic board" is probably behind the external LED's?