I am setting up a new controller box using a smoothstepper(using plug V0150gx2). The box contains:
-(4) Geckodrive G201
-(1) Geckodrive G203V
-(1) 53VDC power supply(homemade: transformer, bridge, filter cap)
-(1) 12VDC power supply(wall wart)
-(1) 24VDC power supply(wall wart)
-(1) 5VDC power suppply(wall wart)
-(1) CNC4PC C4 Rev6.2Charge Pump Sense board
-(1) Homann Designs ModIO V1.1
-(1) Homann Designs MA-02 pendant BOB V1.0
-(2) Homann Designs MB-02 BOB V5
-(1) Homann Designs Digispeed V2.0
-(1) Minarik DC Speed Controller
-(2) custom 5-relay boards
-(6) high-current relays(AC power, and spindle DC power)

The box is wired up and hooked to a PC with a fresh install of Mach3(Version R3.042.037)
The PC is  a Dell Dimension 2400 with:
-2.4GHZ Celeron
-512MB RAM
-using PCI graphics adapter
-nothing installed except Mach andModBus reader

I have not yet hooked the controller up to my machine, it is on the workbench, with only the PC and the Homann MP-01 pendant hooked up.

Test go well until I get to the point where I want to simulate a limit switch. When I do so, the limit switch is seen by Mach, as indicated by the status display, and then after anywhere from 1 to 5 seconds, the SS locks up(red LED stops blinking, and Mach reports lost communication, or locks up also.

I have the SS filtering options set at 1,000, except for limits which is set at 10,000.

I also tried with the modbus disabled, and also with the HV stepper supply and the DC spindle motor controller off.

Same results.

Anybody else having this type of behavior with a smooth stepper? I am not sure how high I can go on the input filtering.

Andy Wander

How are your limits wired up? Do they do anything except send a signal to Mach? In other words are you turning off power to anything when you have a limit tripped. What is the status of the LEDs on the SmoothStepper when you see this problem?

Hi Jeff:

My limit switch inputs themselves go directly to the BOB, and thence to Mach3. When a limit is triggered, though, the Estop condition in Mach turns off 2 enables, which control the AC power relays to the Spindle DC controller, and to to my HVDC supply. Mach controls the BOB, the BOB controls a small 5VDC 1A relay, and the small relay controls a (relatively) large 24VDC relay. All relay coils have 1N4003 suppression diodes installed across them.


When I trigger the limit, and then remove the triggering voltage, Mach resets. This drops the enables and turns off the AC power to the HVDC and the Spindle DC controller.This drops out my charge pump, which causes the C4 board to open the "enable" ckt. The enable ckt drops out another small 5VDC relay, which removes the enable signals from the Geckos, and from the BOBs, as well as supplying voltage to the BOB ESTOP pin, and thence to Mach. This disables all of the other BOB outputs.

At this point, Mach will be in RESET mode, and the LEDs on the SS are still the same as when it is running: Green steady, Red blinking, Blue steady.

When I press my hardware reset button, an enable signal is applied to the BOBs, and also to Pin 10(estop) which resets Mach. This turns on the charge pump, which means I can take my finger off the button(my hardware reset button is in parallel with the C4 Charge Pump Sensor contacts). If I have gotten into RESET mode by pressing an estop button, or by pressing the RESET button on the screen, then at this point, Mach resets smoothly, but if i have gotten here by triggering the limit input, then as soon as I press the hardware reset button, the SS Red LED drops out, and Mach locks up.

...AND while I was testing and writing this up, I think I have found a way around it!! It seems that if I click on the screen RESET button before I press the hardware reset button, the SS will NOT lock up. I am not sure why this would be??

Wow, my head hurt just trying to follow all of that logic. Is there a reason you must remove servo power when a limit is hit? Depending on a signal to make it from the SS through you BOB, through Mach, back through you BOB, etc is adding a huge amount of complication and no additional security (as you have way to many pieces of electronics in the loop.) You would be as well off just letting Mach handle the limit and cease driving the axis (bring them to a controlled stop). If you must remove power from them then you should do so directly by having the limits drop out the mains supply to the drives power supply. In a similar vein IF all your your BOB outputs should be 'off' when a limit or E-Stop occurs you can't depend on a BOB to take care of this, you should be removing drive power to those things.

Turning power supplies on/off can cause a lot of electrical noise which can throw many things for a loop. If your relays are in close proximity to your BOB and SS that could be causing a problem. I'll be releasing a device to reduce this arc noise in a few weeks.

Wow, my head hurt just trying to follow all of that logic.

Think how mine felt when i was writing it...

Is there a reason you must remove servo power when a limit is hit?

No, not directly, but it is a function of the actual estop circuitry. i have a loop which passes through my 2 estop switches, and through the Charge Pump detector, and that loop also cuts off the power to the steppers and the spindle.

I don;t think that the turning off of the power supplies is doing this. If i press the Estop (physical) button, nearly the same sequence of events happens. Certainly all the same things are shut down. But in that case, i can just reset the physical Estop button, and press my physical reset button, and get Mach running fine again.

If the sequence is initiated by a limit switch, the power supplies and all are shut down, and the SS keeps communicating UNTIL I press the physical reset button. At that time, no switching of any relays occurs(because Mach never sends out the CP), but the SS then loses communications.

Thanks,

An E-stop even and a limit even are two seperate things. Your limit swithces should not be part of your E-Stop chain. My suggestion is to just let Mach handle the limits by ceasing sending out stepping commands, etc. Then you can use Mach's built in functionality to allow you to only jog in the correct direction to get off of the limit.

I'm not being clear.

My limit switches are NOT  part of the Estop chain.

However, my limit switches. which feed directly into the PC thru the BOB, DO affect the Estop chain in the following manner:

When a limit switch is opened, Mach shuts down the charge pump, and the enable outputs. The absence of these signals from Mach causes the Estop loop to open and keep the system in Estop, as well as shutting off the AC supplies, etc.

I'm not being clear.

My limit switches are NOT  part of the Estop chain.

However, my limit switches. which feed directly into the PC thru the BOB, DO affect the Estop chain in the following manner:

When a limit switch is opened, Mach shuts down the charge pump, and the enable outputs. The absence of these signals from Mach causes the Estop loop to open and keep the system in Estop, as well as shutting off the AC supplies, etc.

Your limits are part of the Estop as you just said this in the second part of your quote that i emphasised

And until you take them out of the equation you may not achieve what you want

Phil

Hmm-is there a way to have Mach3 maintain the Charge Pump and the enable signals when a limit switch is triggered? It seems that in my installation, the limit being hit sends Mach into an ESTOP condition. And that is no matter what I have hooked up to what else.

I have just tested my set up.

Touching a limit will trigger an Estop this is a software Estop

You do not need the charge pump with a smooth stepper

A charge pump is a software safety device, lose communication you get Estop

If SS loses communication you get an Estop, you can configure how much data is held in the SS buffer before you get an Estop

I would read as much as you can about the SS, threads etc to confirm

Phil