Dangerous X5 Plus Uncommanded Spindle Running

Hi guys,

Just trying to figure out if there is a general fault with these machines, or just a faulty board in mine.

My setup is a 2016 Syil X5 Plus, running Mach4 with Ethernet Smoothstepper and Windows 10 Home.

The machine, Mach4 and the ESS run G code perfectly (so far). The problem I have is as follows:-

Using Port#1 on the ESS, the machine runs OK, BUT if when the machine is in CNC mode and not running code i.e. a tool change stop, power to the ESS is turned off (or lost due to the 5V power supply failing), the machine spindle starts and runs up to full speed, only stopping if either the E-Stop button is pressed, or the machine put into manual mode.
This fault does not occur when I use Port #2 on the ESS, as this port reverts to pins 2-9 being set as Inputs by default when power is removed/ lost.
I have tried a second ESS with exactly the same results, so pointing to either an inherant problem between the ESS and the X5 Plus interface boards, or just a faul on my machine.
This is definately not a Mach configuration issue because it only happens if there is NO POWER to the ESS.

If any one with either an X4 or X5 Plus with VFD Spindle drive could try and replicate the above I would be gratefull, as it's taking ages for the UK distributor to try and sort out.

Nick.

I dont even know what an ESS is, but your description seems to say you are using the wrong polarity for the vfd on signal; can you REVERSE (invert) the meaning of the bit that turns on the vfd? seems when something in control or wherever reboots or turns off, it leaves this bit in the state you have described to TURN ON the vfd: you want this power off state to default to vfd OFF not on....

so, yes, t iwill be a mach3 setting in pins and things or whatever screen it is where you set the inputs up - there is an invert button you can pick. then change meaning on the other side too (in vfd, set that ON input to INVERT to match).

Hi Mike,

Thanks for your input, but as I said in my post, this only happens when power is removed from the ESS (a motion controller made by Warp9td, for Mach3 and Mach4), even if the ESS is NOT CONNECTED to the computer.
I've since tried the machine connected directly to a computer with a parallel port and with the computer switched off there is no problem with the spindle starting without warning.
So after trying the above, it's still pointing to a problem between the machines Parallel port and the ESS.

Nick.

Quote:

---

Originally Posted by nick952

Hi Mike,

Thanks for your input, but as I said in my post, this only happens when power is removed from the ESS (a motion controller made by Warp9td, for Mach3 and Mach4), even if the ESS is NOT CONNECTED to the computer.
I've since tried the machine connected directly to a computer with a parallel port and with the computer switched off there is no problem with the spindle starting without warning.
So after trying the above, it's still pointing to a problem between the machines Parallel port and the ESS.

Nick.

---

You have not said how the spindle has been wired for turning it on/off this should be switched through Relays, it sounds like you are not using Relays to control the Spindle Drive, if it is wired correctly this would not happen

Use relays AND watchdog timers-with safety relay, to and from the PC, You MUST NOT rely on software only to control something as dangerous as a spindle.
If the OS fails. Machine stops dead.
If communications to/from the control fails. Machine stops dead.
Momentary power loss? Machine stops dead.
Big red button? Machine stops dead.
Regardless of what fails the safety relay drops out, and machine BRAKES to a halt.
You must not rely on ANY software to just turn off something. Non computer hardware must monitor the complex computer parts.

The warp9td website specifically warns of the differences between the ports.
4) Pin out: It is important to note that Port 2 on the USS and both Ports 2 & 3 on the ESS are functionally different than Port 1. Pins 2-9 of Port 1 are always outputs, whereas Pins 2-9 of the other port(s) may be used as inputs or outputs. The direction of these bits is set in the SmoothStepper’s configuration dialog. These pins have weak (47k) pull down resistors. If a pull-up resistor is desired instead of a pull-down, a pull-up resistor with a maximum value of 4.7k can be added and it will behave similar to the dedicated inputs of each port (pins 10, 11, 12, 13, and 15). When using a breakout board that uses logic ICs to drive those pins (most breakout boards), the resistor becomes irrelevant.

Regardless of this, leave computers out of the safety loop, using careful watchdog circuit design, among other things.

Mactec54 and neilw20,

Thank you both for your input.

Bear in mind that these machines are sold complete and ready to run (once you add a computer and if needed a motion controller).
The spindle on/off and direction is controlled by relays and the E-Stop system works exactly as it should.
(If you re read my origional post I think I made the E-Stop functionality clear enough)

Turn off the controlling computer or unplug the ethernet cable and the machine stops.
This problem ONLY occurs when using PORT #1 on the ESS and THERE IS NO POWER SUPPLY TO THE ESS,

I am only trying to find out if this problem is an isolated issue with a faulty interface board on my machine or an inherant problem with the design, that people will only find out by accident and hence the reason I have come on here to ask other users of the same machine, to try and replicate the conditions that cause this to happen (or not)!

An update to this issue is that I've tried an old non working (so didn't even plug it into the mains, therefore replicating a power failure to the controlling computer or the ESS) connected directly to the machines parallel port WITHOUT THE ESS and there is NO UNCOMMANDED SPINDLE STARTUP. The machine sits there perfectly dead, untill I put it into Manual Mode and I can then start the spindle, jog around using the MPG. (Please note that the Manual/CNC Mode selection, is via a control pannel on the machine and so is independant of the external computer).

It does appear that there is an electronic difference between the design of Port #1 on the ESS and a standard Parallel Port as fitted to my old computer, that is causing problems with the control boards in the machine.

Nick.

nick952

With no 5v power on the ESS it is not doing anything, the relay must be wired normally closed, when the power is off, or is a faulty Relay,for the spindle to start with power off, this should not be hard to figure out, that something is amiss between the ESS and how the Relay is activated, the Relay should be normally open with the power off, it obviously is not, or is getting a signal from somewhere to turn on, this is what you need to look at, is this Relay

There is a Big difference between Port 1 on the ESS and your computer parallel port just by design,the ESS has it's own CPU, your old computer does not compare, but either port with the power off won't change how the Relay reacts, do some photos of the wiring, someone may be able to see something, that you are missing

Quote:

---

Originally Posted by nick952

this only happens when power is removed from the ESS (a motion controller made by Warp9td, for Mach3 and Mach4), even if the ESS is NOT CONNECTED to the computer.

---

That's because when the ESS is not connected and the input is left floating. This can happen even when the ESS is connected to the CNC but Mach is not running, so the ESS is not initialized. I don't think it is a good idea to have the CNC powered up unless it is connected to Mach4. By that, I mean that the software must be running and computer connected to CNC through the parallel port or the ESS.

I don't have the ESS, but that's my guess, since generally it is NEVER a good idea to leave inputs floating. Another solution is to use pull-up resistors on all inputs, but the best is to follow the rules of never having the inputs in any undefinable state. My CNC behaves the same way if I disconnect the UC300, which is my USB interface, but I make sure it never happens.

If bidirectional watchdog timers are implemented to hold the safety relay on, uninitialized ESS will send no watchdog signal.
Safety relay drops out.
No power to ESS. No watchdog. Safety relay drops out.

Careful watchdog design, will leave the safety relay off.
Your safety relay can be in series with the ESTOP mushroom, requiring minimal interference with existing design/wiring, apart from possibly extra contacts on ESTOP button.

At startup, you need a method to engage the safety relay, until ESS and Mach watchdog signals can latch the relay. ie, you hold a button during initialization.
Also your safety relay needs powering through the ESOP button so the estop also drops out the relay.

Anything floats, or is off. The safety relay drops out, and being in series with ESTOP, is same as pushing ESTOP.

Thanks every one for taking the time to reply and suggest causes and possible solutions.

My reason for posting was only to try and find out if this is an isolated issue, confined to my machine, or more widespread caused by a design fault and in so doing also alerting other users of a potential for serious injury.
As this is a newly purchased machine, any problems are the responsibilty of the importer/distributor to resolve, not mine and so I'm not going to delve too deeply into sorting this myself.

Nick.

Nick,

The problem couldn't be replicated on my X4+ (Also using ESS). A floating pin (Pin 9 of port = Safety Charge Pump) & a parasitic frequency of 12.5Khz could result in this happening to you. As Neil points out, having pull-up/down resistors is going to reduce the likelihood of spurious signals generated inside/outside of your X5 cabinet. That being said, Syil swap circuitry and other components on a frequent basis. Mine's a 2015, possibly 2014. So a direct comparison of this behaviour isn't always possible.
Are you not able to stick with Port 1?

Pete

Hi Pete,

Thanks for trying to replicate the problem. This does seem to be an issue with the electronics on my machine.
I will stick with Port #2 on the ESS as this works perfectly, but obviously if there's a fault with the machine or it's electronics, I want it to be sorted before the guarantee runs out (and it's hard work getting Syik UK to actually sort problems out).

Nick.

Quote:

---

Originally Posted by nick952

Hi Pete,

Thanks for trying to replicate the problem. This does seem to be an issue with the electronics on my machine.
I will stick with Port #2 on the ESS as this works perfectly, but obviously if there's a fault with the machine or it's electronics, I want it to be sorted before the guarantee runs out (and it's hard work getting Syik UK to actually sort problems out).

Nick.

---

No, it's not necessarily an issue at all, it is normal condition. You should NEVER have inputs floating and outputs uninitialized. Perfectly normal condition if those basic electronic rules are not followed. A floating input and uninitialized output can have ANY state and can even start to oscillate uncontrolled, causing any possible unwanted behavior. Installing pull up resistors is not always a good solution because if your ESS has built in pull-ups than adding another one will reduce the resistance and may cause overload on the ESS output.

Quote:

---

Originally Posted by nick952

Hi Pete,

Thanks for trying to replicate the problem. This does seem to be an issue with the electronics on my machine.
I will stick with Port #2 on the ESS as this works perfectly, but obviously if there's a fault with the machine or it's electronics, I want it to be sorted before the guarantee runs out (and it's hard work getting Syik UK to actually sort problems out).

Nick.

---

If port 2 works how everything should, then it's as simple as re-configuring port 1, it most likely just one pin state to change or reassign, Both ports have different pin assignments, so just a simple change of a pin would fix it

Quote:

---

Originally Posted by mactec54

If port 2 works how everything should, then it's as simple as re-configuring port 1, it most likely just one pin state to change or reassign, Both ports have different pin assignments, so just a simple change of a pin would fix it

---

Not quite, the ESS's Port #1 pins 2-9 are fixed as Outputs, where as Ports #2 and #3 pins 2-9 can be configured as either Input or Output and I think Ports #2 and #3 revert to Input by default when the power is OFF and therfore possibly (probably) why I don't have the problem when using Port #2.

Nick.

Quote:

---

Originally Posted by nick952

Not quite, the ESS's Port #1 pins 2-9 are fixed as Outputs, where as Ports #2 and #3 pins 2-9 can be configured as either Input or Output and I think Ports #2 and #3 revert to Input by default when the power is OFF and therfore possibly (probably) why I don't have the problem when using Port #2.

Nick.

---

It would be a good idea to study how the ESS works, you have something setup incorrectly for port 1 not to be working, as it should

Dedicated Output Pins

Pins 1, 14, 16, and 17 of all 3 ports. Plus pins 2 through 9 of Port 1

These pins are rated for outputting 0 to 5V, but may go as high as 5.5V if your power supply is sourcing that voltage. Their output current is designed to drive (sink) +(-) 24mA.

Please note that each of these pins has a dedicated 47.0kOhm pull down resistor.

When the ESS is powered, but not yet controlled by MachX, these pins will be pulled low, down to GND. After Mach takes control of the ESS, the unconfigured pins will stay the same, and the configured pins will be determined by your settings. Any external circuitry that is sourcing or sinking a current to/from the pin will change the state of the pin, but since this is a dedicated output, there should not be any circuity trying to control this pin. The external circuitry should only be reading this pin's value.

This pretty much sums it up as to what is going wrong, so your system is either missing a Relay that is connected to the ESS for your spindle control, this would isolate it from the power source that is turning your spindle on, your spindle is turning on because of the manual control, not because of the ESS

Post some photos of how it has been wired, it will be easy to see what is feeding the relay that is turning the spindle on