PMDX-126 Rev B K1/K2 Relay issue

[thosj]

I've had my PMDX-126, Rev. B (I think!) running on my BP clone for years, Mach3, ESS. I've used K1 and K2 for coolant/mist all along. Suddenly, I get an eStop when issuing M8 or M7 in MDI/Gcode, or via my Pokeys mapped physical switches. I'm using  port/pin 1-1 and 2-14 and as I said, been working all along.

Obviously noise somewhere, but to my knowledge nothing has changed in my setup. I don't know where to start to try and fix this. The coolant relays are 110v, so the power is coming from the Mains power on the PMDX-126 to K1 and K2, simply jumped up to the relays. I've checked all the wiring connections and everything seems OK. I've tried Smoothstepper Noise Filtering of Inputs settings, up to 250, but I'm unsure what field might effect this so I've put 250 in all to NO effect. I think Mach3's Debounce is ineffective with the SS, correct?

So, what to try? Capacitance across the K1/K2 and if so what capacitor? More SS Noise Filtering value?

I'm baffled and open to suggestions!

*****Edit********

The ESS is powered remotely, it's own stand alone 5v PS.

Thanks in advance!

Tom

[Steve Stallings]

The large K1 relay on the PMDX-126 has a MOV arc suppressor permanently wired across its contacts.

The K2 relay has a MOV terminal that can be jumpered to the N/O contact if desired.

These MOVs should normally suppress the worst of the electrical noise.

I think you are correct that any debounce or software noise filtering must be done in the SmoothStepper configuration.

[Bob at PMDX]

Suddenly, I get an eStop when issuing M8 or M7 in MDI/Gcode, or via my Pokeys mapped physical switches.

(1) By "via my Pokeys mapped...." do you mean you have mechanical switches connected to your pokeys board and those switches are configured in Mach to manually turn the flood/mist relays on and off?  If not, what do the switches on the pokeys board do?

(2) Can you look at the PMDX-126 while you issue an M7 or M8 command?  Does the red EStop LED on the PMDX-126 turn on when the relay(s) engage?

(3) Disconnect the 110VAC that is being switched by the relays, so the relays can close and open but won't actually switch power to the flood/mist pumps.  Then try M7 or M8.  Do you still get the EStop?

(4) Disconnect any EStop circuit you have connected to the PMDX-126's EStop and Fault inputs.  Then take a (very!) short piece of wire and connect the EStop input to the adjacent GND terminal.  Leave the Fault input unconnected.  Then try M7 or M8.  Do you still get the EStop?

(5) Are you powering anything from the "+12U" terminals on J11 or J12 of the PMDX-126?

(6) Are you powering anything from the "+V Raw" terminal on J7 (right next to the AC voltage selector switch)?

(7) A drastic test is to disconnect everything from the PMDX-126 except the 110VAC input power, the SS, and the EStop jumper to GND.  Then try M7 and M8 and see if you get an EStop.

There *may* be something loading down one of the PMDX-126 power supplies, or there is a faulty component in the power supply section.  Then when one of the relays tries to energize it loads down the supply  even further and causes the PMDX-126 to malfunction.  We'll see what your answers are to the above questions.

Bob

[thosj]

Suddenly, I get an eStop when issuing M8 or M7 in MDI/Gcode, or via my Pokeys mapped physical switches.

(1) By "via my Pokeys mapped...." do you mean you have mechanical switches connected to your pokeys board and those switches are configured in Mach to manually turn the flood/mist relays on and off?  If not, what do the switches on the pokeys board do?

(2) Can you look at the PMDX-126 while you issue an M7 or M8 command?  Does the red EStop LED on the PMDX-126 turn on when the relay(s) engage?

(3) Disconnect the 110VAC that is being switched by the relays, so the relays can close and open but won't actually switch power to the flood/mist pumps.  Then try M7 or M8.  Do you still get the EStop?

(4) Disconnect any EStop circuit you have connected to the PMDX-126's EStop and Fault inputs.  Then take a (very!) short piece of wire and connect the EStop input to the adjacent GND terminal.  Leave the Fault input unconnected.  Then try M7 or M8.  Do you still get the EStop?

(5) Are you powering anything from the "+12U" terminals on J11 or J12 of the PMDX-126?

(6) Are you powering anything from the "+V Raw" terminal on J7 (right next to the AC voltage selector switch)?

(7) A drastic test is to disconnect everything from the PMDX-126 except the 110VAC input power, the SS, and the EStop jumper to GND.  Then try M7 and M8 and see if you get an EStop.

There *may* be something loading down one of the PMDX-126 power supplies, or there is a faulty component in the power supply section.  Then when one of the relays tries to energize it loads down the supply  even further and causes the PMDX-126 to malfunction.  We'll see what your answers are to the above questions.

Bob

Answers by the numbers!

1. Yes, mechanical switches hooked to PoKeys and mapped with Pokeys software. All buttons working as always, only M7/M8 buttons tripping, just as MDI M7/M8 do.

2. No, the PMDX Estop red LED does NOT light when this happens. I have LED's connected to J7 and on my panel and that Estop doesn't light either, of course!

3. NO, it does NOT Estop with the coolant switches disconnected, PMDX K1 and K2 relays click and LEDs light up. M9 clicks and turns off LEDs, all as it should be. So the act of tripping the coolant switches is doing this, apparently.

4. Did NOT do this, yet. Thinking we're on to something else, but perhaps not.

5. Nothing powered by +12U on J11 or J12

6. Nothing powered by the +V Raw terminal on J7

7. Did not go here yet either.

Again, this all worked as it is for years. Only recently started with no wiring changes anywhere and no other symptoms, everything else runs and works as always!

Thanks

Tom

[Bob at PMDX]

Are you sure that Mach3 is seeing the "EStop" signal?  Does the message in the Mach3 status line say "External EStop Requested"?  A limit switch going active will also make the Mach3 "Reset" button flash but will display a different message in the status line.

If the PMDX-126's EStop LED does not turn on, then the problem is NOT the EStop signal coming into the PMDX-126 board.  The PMDX-126 samples and filters** the EStop input, and outputs the EStop signal to the PC.  The EStop output to the PC is asserted for at least 1 second whenever the PMDX-126 detects the EStop input.  So even narrow pulses on the EStop input should result in visible EStop LED activity.

** I am pretty sure the rev B boards filter the EStop input but I haven't gone back through the code to verify this.

It is  possible that noise is getting coupled onto the EStop output (port 1 pin 10) between the PMDX-126 and the ESS.

How far away from the PMDX-126 are the external relays from the PMDX-126?

One other test would be to re-route the 110V that goes through the K1 and K2 relays and keep the 110V wiring as far away from the 126 and ESS as possible.  Except, of course, where it has to come into the AC input and the K1/K2 connectors.

Bob

[thosj]

I DO get the External Estop Requested in the Mach 3 status line, but no red EStop LED on the PMDX. I don't know why I'd get External Estop Requested in Mach and no EStop on the PMDX nor why it started up just now.

The K1/K2 relays switch solenoids for air, both are air, one is a Kool Mist unit and one is a cold air gun, same solenoid model on each relay.

I'm not sure I even have a Rev. B, my s/n is 27514. It doesn't say Rev. B or Rev. anything on it anywhere I can find.

I actually have two PMDX bobs, the other is a PMDX-125. s/n 27346. I think I bought the second one to get the probe inputs that can be linked or whatever it's called. I am not using the first one at this time. I guess I thought they were both 126's and that the second one would be Rev. B, but..........

Looking on your site, I really have a Rev. A as it seems Rev. B starts at s/n 28601. Does this mean the Estop is NOT filtered? Would a new Rev. C PMDX-126 fix this, do  you think?

I guess I could rig up a temporary setup with an external 110 feeding the K1/K2 relays, not taking the 110 from the same place the 126 is getting it's mains power, and see if that works. I could then route the 110 to the solenoids way away from anything.

Still mystifies me that this setup has worked for 5 years and now it doesn't!! I guess that's the way it is with "noise." And, sorry I don't "get" this to well, it's been a long time since I set it up and it's just been humming along all this time, so I have to refresh my memory every time I look in the box.

Thanks again,

Tom

[Bob at PMDX]

As for which version of the PMDX-126 you have, I can't access our serial number records from home.  But we can figure this out.  All location descriptions below are based on the 126 oriented so that the 25-pin "D" connector for parallel port #1 is at the bottom edge, and the transformer is at the top of the board.

Rev A boards: The serial number block is near the lower right hand corner of the board, just below the ESTop LED.  There should be a silkscreen block just above the serial number for the revision, and should be marked "A1", "A2" or "A3".  If you can look at the underside of the board, in the lower right corner (lower left now that you've flipped the board over), you should see in copper "PCB-485A".

Rev B boards: The serial number block is along the lower left edge of the board, just to the left of the 25-pin "D" connector for parallel port #1.  The silkscreen also says "(Revision B") in the center of the board just below the transformer and the "PMDX-126" text.  Again, there should be a silkscreen block for the revision just above the serial number block.  It should say "B1", "B2" or "B3".

Rev C boards: The serial number block is in the same location as the rev B boards.  The silkscreen says "(Revision C)" in the center of the board just under the "PMDX-126" text.  The revision block would have "C1", "C2" or "C3".

A couple more things to look at, after you try #4 from my previous message:
(1) If you have a volt meter or DMM - With the 110VAC disconnected from the external relays, turn on the PMDX-126 K1 and K2 relays.  Measure the voltage between any of the "PC+5V" and "PC gnd" terminals on J1 through J6.

(2) Reconnect 110VAC to the external relays.  Disconnect the wire for your panel EStop LED from connector J7 on the PMDX-126.  See if you still get the EStop when the relay(s) energize.

(3) Where is the ESS mounted relative to the PMDX-126?  Is it mounted on top of the 126?  If not, how far away from the 126 is it?  And how long are the ribbon cable or cables from the 126 to the ESS (if the ESS is mounted on the 126 and you bought the cables from us then they are 6" (approx.).

Bob

[thosj]

Bob,

I'm going to get to the rest of your recommendations/tests tomorrow, but if you see this before then, my board is a Rev. A board, A1 sn 27514 on the silkscreened blocks on the lower right as you mention. Does this mean I definitely do NOT have these MOV's of which Steve speaks across the relay's, and/or does NOT filter the EStop inputs as you mention? If that is indeed the case I might be inclined to buy a new one!! If that's not the case I'll keep plugging away at this.

Tom

[Steve Stallings]

The PMDX-126 boards have always had the MOVs on the relay contacts.

The enhancements to the E-Stop were introduced with Rev C. These
changes were most beneficial to users who were using an actual parallel
port and matter less for users of a SmoothStepper because the cable
from the PMDX-126 to a SmoothStepper is usually quite short.

I think the more important thing to do is to figure out what changed. We
all like to think that "nothing has changed" but the results tell us otherwise.
While it is possible that the reaction of the PMDX-126 to noise could have
changed, it is more likely that the source or strength of the noise has
changed.

The first order of business should be to confirm the source of the noise.
Removing the power that flows through the relays to see if the problem
goes away on a trial run should confirm if it is coming from arcing of the
relay contacts. Usually this is worse when turning off the current but can
occur when turning it on.

If it is confirmed that the E-Stop only happens when the circuits are drawing
current through the relay contacts, then look to see if the wiring harness has
been moved or other wiring changes have been made in the wiring that is
connected to your devices that are controlled by the M commands.

If you are convinced that no changes have been made, then it would be most
cost effective to add an additional MOV device across the coil of each load
device rather than to replace the PMDX-126 board. The MOV should be selected
with an operating voltage that is similar to the load device.

[thosj]

OK, Bob,

Step 4 on your original email, I jumpered (with a very short piece of wire!) EStop to gnd, still get the EStop on M7/M8.

Step 1 of the newer email, 5.00vdc on all J1 thru J6.

Step 2, disconnect all panel LEDs, still get EStop.

Step 3, ESS is about 4" from where it would be if connected directly to the 126, Port1 to ESS 6" ribbon cable, Port 2 to PMDX-108 6" ribbon cable, PMDX-108 to ESS 8" ribbon cable. That one could be 6" but I must not have had one and do not have one now. I did swap for a new 8" cable to no effect.

Now for the interesting part. I wired up a separate 110 into the K1/K2 relays. If plugged in to a different house circuit, it did NOT EStop. If plugged in to the same house circuit, it DID EStop!! All machine 110 is thru a conditioning UPS. This seems to point at a fault in my house wiring. I can easily run a new circuit to the machine and give it it's own dedicated 110 circuit, something I should probably do anyway. I can also easily run a separate 110 for just the relay's if necessary.

Interested to hear your thoughts.

Thanks to both you and Steve for trying to help me out. I'm pretty good at machinist stuff but when electronics go berserk, I'm less hopeful!

Tom

[Steve Stallings]

It sounds like the noise is getting onto the power line and either the computer
or the SmoothStepper does not like it. The use of a separate circuit may be
acting like a filter to some degree.

Is the power for the for the loads controlled by the relays coming from the
output of the UPS and shared with the computer and the SmoothStepper power
supply? If so, move it to the other side of the UPS.

[thosj]

The computer was behind the UPS but the control box was NOT. I moved the control box behind the UPS and all is well. Well, "well" with the power to the K1/K2 relays now coming from a separate source but still plugged into the same house circuit as the UPS and all it's stuff. So, even though I don't get it, I'm fixed.

Thanks once again, PMDX IS the best!

Tom