Blasting through Z stop on Cobra X3

[richf]

I had a Z axis skip the other day that caused the side router head to plunge through a piece and into my vacuum jig. I had attributed this to static electricity even though I had grounded my dust collection hose and haven't had an issue in months. That was last weekend.

Today I went to cut a replacement jig. I reinitialized and everything worked which is typical for a static electricity issue. After the cut, I manually moved the Y axis to get the head away from the piece and I heard a clunk which I knew was a Y axis skip. I confirmed by moving back to x0y0 which was way off. I went to reinitialize and my Z axis kept on running. I found that it pushed the Z stop switch and it bent the bracket and Z stepper kept running. I did some troubleshooting and the Z switch works properly by closing the circuit when it is pushed. What it doesn't do is light up the indicator on the control board, hence appears it can't communicate to the control software.

The voltage applied to the other axis stops, X,Y, is 24 volts and when the button is pushed the voltage drops to zero. I noticed the Z axis voltage is somewhat inconsistent but hangs out around 14 volts. Note that the Z stop and tool measuring plate share the same voltage. So Z high and Z low which are p0b2 and p0b3 respectively on the control board both show this lower voltage. Also, none of the axis will move now nor will spindle power up even though the control board is getting the signal for the spindle (I know by the click and the indicator light coming on). Also, laser control port works.

I tore apart the tool measurement box (it looks like a headphone jack) where you plug in the 1/2" plate and wiring looked a bit messy where it attaches. I moved it around a bit and looks like it made the Z stop voltage vary. What I don't understand is why all of the axes appear to be offline and not just Z. Now as I'm typing this, I think I need to remove this Z low wire to clear up the voltage but I wouldn't think this would affect the other axis.

I appreciate any input on this. I'll contact support on Monday if it doesn't get fixed.

[Gary Campbell]

Varying voltage is a sign of a load that is too heavy, or a short. Possibly a bad power supply.

Voltage at the 24vdc power supply?
Voltage when nothing is connected to it?

[richf]

Looks like I found issue. The two board inputs to Z hi and Z low show 16 volts even if I remove all the Z axis connections. It might be a solder issue on the back of the board. I put my finger back there and tried to swipe it and used air to clean as best I could in case a piece of conductive debris was laying back there - that didn't change anything. I should be able to relocate this control block assuming the wires reach to the new location. If not, I should be able to relocate the adjacent block somewhere else and use it.

P0B2 and P0B3 measure at 16 volts as shown. I measured between there and where the COM power connects into the power supply. The other adjacent pins measure right about 24V. Otherwise, can't imagine having to pull that board and try to resolder those - it would be a lot of labeling and meticulous rewiring. Luckily, Camaster redid this whole control board when it visited the plant back in 2012 so it is somewhat up to date. I bought it from Mick Martin.

Edit update: I went back looking at the adjacent location. I assumed that P0B5,B6, and B7 would have power but they don't. Is that by design or is there something wrong? The COM input power is coming in to that block and P0B4 is powered. I assumed that COM input would power that block of 4 pins since P0B0-B3 are powered.

I also realized that it would be pretty easy to remove board as long as you keep track of the green block locations. For some reason I was thinking you would have to remove all the wires.

[richf]

I added additional info in my last post. Does anyone know approximate costs to replace the whole control board board for this X3. Board date is 2010. The board has lots of inputs. It is made to add a lot of peripherals.

[Gary Campbell]

Rich...
A couple notes...
None of the input pins should show any voltage when empty. They are activated by applying the INVERSE of the COM voltage. i.e, there is -24V to the com, it takes +24v to trigger the input. +24 to com, -24 to trigger, etc. There should be voltage (24vdc) across com and a triggered input


Did you measure voltage at the 24vdc power supply?

Inputs can be blown as you describe with a static jolt


Replacement CN1/CN2 IO boards are ~$500

[richf]

Thanks for taking time out on the weekend to look at this.

I did go directly to the power supply and it pretty much showed exactly 24V at the points that looked obvious that were + and -.

So it appears the two empty pins showing 16V (not sure I have polarity correct) is an issue. If that is the case I should be able to update my wincnc.ini file to utilize the P0B5 and P0B6 pins I assume. Note I did these measurements with the CNC1/2 cables pulled off the board.

[richf]

Well, I used another powered port, P0B4, that is utilized for the Z probe which I have never used and utilized this for the Z high. When I press the Z stop, at least the indicator light comes on just like for the X and Y stops. I updated the wincnc.ini for this location and commented out the probe but behavior is same. No axis will move. I hear a whining in the Z stepper I believe when I do an iniitialize. When I click T2 tool, there is a long delay and eventually the T2 router drops down. When I click T1, it retracts like normal. The long delay is not normal.

I did some resistance measurements between the COM port and the adjacent pins (power off obviously). The ones that appear to work show a resistance of about 2 million ohms but the ones that don't work appear to be completely open (infinite resistance).

I'm guessing my board is fried. Will open up a support ticket with Camaster and reference this thread. Unfortunately, I won't be home during the day on Mon and Tues in case I need to do a procedure.

[richf]

Tech support gave me instructions for testing the WinCNC card in the computer and it was fine. Looks like it is the main board so ordered a new one today. Apparently my grounding was done wrong, so I likely killed it with static electricity. I know multiple ports were non-functional. I had the wire going through the dust hose which is grounded on the dust collector frame and then attached it to the spindle/router structure. It isn't supposed to touch the CNC frame metal.

[Gary Campbell]

Rich...
Look this thread over: https://www.camheads.org/showthread.php?t=8650

Video and a PDF to D/L

[richf]

Thanks, will be reviewing that.

[richf]

I installed a brand new control board and I get exactly the same symptoms, none of the axis will move. The only difference is that the port used for Z hi is working properly since light comes on now when Z stop is pressed so I know there were some bad ports on the old board.

Maybe it is one of the internal cables or some pin that would prevent movement of any axis. I don't know what would prevent all of the axis from moving unless it was a bad ground. I can still turn on the spindle and laser ports but the spindle won't power up.

[richf]

Found a wire hanging loose that is pin 10 common for all the Gecko drives. That would explain why they aren't responding. I'm trying to determine where that one red wire goes. Each gecko has pin 10 connected to each other and the X gecko has a long disconnected red wire. I'm trying to determine where that one goes. I'm guessing I knocked it out (must have been loose) when I pulled the CNC1 and 2 connectors out so I could access the connections to the Z hi board connector.

[richf]

Fixed. The red wire needed to be connected to a +5V pin on the controller board, of which there are many of those ports available. At least all of my wires going to the control board have labels on now.

[The real JP]

Fixed?
Is the machine operating normally now?

[richf]

Everything was operating normally and I cut 12 cutting boards on Sunday and when I got to the 4th board this evening, the Z axis started skipping. Even if I operate it up and down, there is something failing. Don't know if it is a gecko driver issue or a hardware binding issue at the moment.