Hi all,

I have replied to this in the Warp9TD forum.  It is a problem that I will address tomorrow.  If the SS loses communications with the PC it should stop jogging, but it doesn't.  Even if it doesn't lose communications, the current implementation could be a problem if a jog-off command were swallowed.  Motion from gcode and step jogs will stop as soon as the SS runs out of data, but continuous jog movement is generated on-board.

Thanks,

Greg

Greg,

Can I suggest that as soon as the SS detets a problem with the USB comms that it disables all outputs and enters  a safe state.

Cheers,


Peter.

Each machine will have a different requirement to be safe in case of an Estop
When using inverters and AC motors on the spindle, the last thing you want to do is "switching off" the inverted as it will have no means to brake
Sending it a stop signal while keeping it energized is far better.

A lot of commercial kit Estops in sequence, where after a short while they also switch of the power to the drives
On heavy gantries with free running ball screws the same counts, (think punching machines), taking power away from the drive will allow nature to take over (inertia).

It is striking to see the changes in the EU machinery directive over the last few years where the safety schemes have changed from prescriptive to functional as the editors understood that each machine design has its particular challenges, and what works for one may actually be unsafe for another.

I have to agree that only relying on the charge pump is a bad thing, the Estop when pressed by the user should override all, and bring all moving parts to a stop in the fastest way possible without causing a danger in doing so (think of machine becoming unstable due to braking)

The above does not take away that SS bug should be repaired as they said they wood

machinery safety... I find it one of the most fascinating things to think of as there are so many conflicting issues, a real puzzle

Each machine will have a different requirement to be safe in case of an Estop
When using inverters and AC motors on the spindle, the last thing you want to do is "switching off" the inverted as it will have no means to brake
Sending it a stop signal while keeping it energized is far better.

A lot of commercial kit Estops in sequence, where after a short while they also switch of the power to the drives
On heavy gantries with free running ball screws the same counts, (think punching machines), taking power away from the drive will allow nature to take over (inertia).

It is striking to see the changes in the EU machinery directive over the last few years where the safety schemes have changed from prescriptive to functional as the editors understood that each machine design has its particular challenges, and what works for one may actually be unsafe for another.

I have to agree that only relying on the charge pump is a bad thing, the Estop when pressed by the user should override all, and bring all moving parts to a stop in the fastest way possible without causing a danger in doing so (think of machine becoming unstable due to braking)

The above does not take away that SS bug should be repaired as they said they wood

machinery safety... I find it one of the most fascinating things to think of as there are so many conflicting issues, a real puzzle

I have to disagree here. To rely on power to apply a brake during an EStop is not going to pass a safety inspection. Well not mine anyway. In my mind, if a system needs a brake to stop a gantry, or a ball screw Z axis, then the system needs to be designed so that the brake requires power to release it, not apply it.  All machine safety brakes I know of work this way.

In your system when a truck wipes out the power pole to a factory, all the ball screw Z-axes will drop, and the gantry will move under its momentum. In my system, all the machine brakes will be applied.

I may be wrong, but there is probably no system that cannot be made safe when power is removed. Keep in mind that some Large machines will blow servo drives and break mechanical components when the EStop is pressed. This is accepted as OK as after all it is an "Emergency" stop.

Cheers,

Peter.

If I had not set up the limit switches directly in to the servo amp

Which is stupid, but I have seen a few commercial machine set up this way.

The servo fault lines and spindle fault lines all go in to an external e-stop input on the PLC.

While controls should participate in enabling a drive they should should not be relied upon for disabling it in an EStop. The estop chain should disable the drives in the proper manner and as a side effect tell your controls there has been an external EStop. In addition an EStop may be able to initiate an EStop but triggering an external safety relay, but the EStop chain should not be routed through the electronics. With a properly functioning EStop chain anytime you hit an EStop button the machine will stop period, no matter how brain dead your controls are at the moment. The controls should be smart enough to sense the EStop and set their own EStop outputs so the controls have to be reset as well as the EStop switch reset to restore power to the machine. This allows you to, for instance, more a robot arm out of the way before allowing a rotating fixture table to reset and rotate into position. Ensuring the proper order of resetting things is also very important.

It seems as though you have made a good attempt to wire things up properly but are still missing the mark a bit.

so no smartie pants telling me that USB power is the solution to my problem.

Well, if you act like an A$$ all you'll get is sh^t. Most of these folks responding to you have loads more experience actually building machines than you do, it would be wise to swallow your pride and listen to their advise. Greg also said (soon after you post on the Warp forum BTW) that he was looking into the problem on the SS end immediately. Please, check you attitude at the door...

I would like to reinforce what Jeff has just said - I couldn't have put it better myself.

Tweakie.

it has been sorted in the latest plugin as far as I know although I have not tested it.
Hood

I am verry happy to stumble on this post.  I am a machine retailer with mach3 soft controller and was thinking of using the smoothstepper board untill I have seen this post.  I think we will still keep on using the trusted pdmx-122 board with the charge pump circuit untill this bug is out of the smoothstepper board.  When machining safty is the most important thing to keep in mind.  I would not be happy if a customer stops by my door telling me he lost one or more body parts becouse of a wire someone tripped over.  We often install machines when the machines are on a different powerbreaker then the pc is on.  It would be easy possible the fuse of the pc is down and the machine is still hot.

You may be fooling your self if you assume all is well if the charge pump is present. If you look back through the Mach yahoo archive or mabe the Master5 (The original CNC software Art wrote) group, you will see that the charge pump ois not there for as an emergency stop feature.

The purpose of the charge pump signal is to cater for the twidling of the the parallel port outputs during the PC power up, and to detect the the Mach software is running.

When the PC powers up, there can be some twiddling of the parallel port outputs, presumable as the bios checks whats there and sets up the port mode. This can cause the steppers to move a couple of steps, changing the axis position. This is annoiying if you shut the machine down at a known position and expect it to be in that position at a restart.

Thesecondary purpose is that the chargepump informs the BOB that Mach is running. It DOES NOT mean that Mach is running correctly. Yes you can configure Mach to remove the chargepump when you press the ESTOP button on the Mach3 screen. But to rely on this is fooling yourself. I've had an axis runaway, and had to hit the "Real" ESTOP button, removing power from the controller.

Cheers,

Peter.

I would like to ask the moderators to remove this thread. Even its title is blatantly false and inflammatory. It accomplishes nothing but besmirching a good product and company. The OP's main issue was that he did not know how to wire up a proper Emergency stop circuit and refused to listen to anyone who tried to help him. He did however have a good point about the SS continuing to run even with a loss of communication, a problem which I believe was swiftly rectified by Greg.

As others have stated the chargepump and com drop shutdown are great features but they should not be confused with proper safety measures.

I am in full agreement with Jeff here. I have had the SS for some time and it is an excellent product. OK I use the E-Stop function but my real Emergency Stop disconnects power to all functions of the machine instantly. SS has not killed a man as the thread title states (in the event that I am wrong then prove it !).

Tweakie.

Jeff and Tweakie,
I hear ya. It took a shocking statement for someone to get attention about a concern of which i will not pass judgement on. Now that there seems to be some closure on this thread, I would like Gum??? to consider changing the thread title. Maybe call it "E Stop Concern". As a thread there is some valid discussion points in  here worth keeping ("I" need to re-read). Personally don't like what the title insinuates. I need to do a Google and see if this thread pops up outside of the forum ( i think it does ).

Any other opinions out there?

RICH