Motor drive ratios and speed

[kennected]

Thanks Gerry,

You are right, I AM using VCC not 24v on the positive step and dir connections.

The step is NOT lighting up at all. When I attempt a move. The DIR turns red only in one direction. When I try the other direction, it just goes out.

I will put a switch on the OVR tomorrow and use it. I thought I read that it wasn't necessary just now.

Not sure what you mean "press the e stop on Y104. You mean move the wire to Y104 and try?

Ken

[ger21]

The Dir LED will only be on when moving one direction, and off for the other direction. It's just a 0V, or 5V signal, depending on direction.

So, if the Step LED is not lighting up, then the UB1 is probably not sending any steps.

You said you have an emergency button wired to Y104? If you push the button, does UCCNC go into Reset?

I think I was mistaken about the OVR button. I think you only need it if you are using the OSSD output. The OSSD output should turn off when the safety circuit is triggered, and the OVR button turns it back on.

[1Jumper10]

I do understand that there are a lot of ways to go here. These controllers and Bobs are made to be versatile and that makes them a lot less than "out-of-the-box" ready, but what another fruitless day of floundering. I am and I'm running out of things to try.

The servos are still showing a flickering green light so I'm guessing they are getting at least SOME of what they want. The DRO's advance with arrow key and or flyout buttons. I have the HLFB connections for the Y connected positive to the to the the Y105 and the neg to 0v.
The Enable neg wire is on Y301 and the positive is on 24v.
The Emergency button is normally closed and positive to Y104 and negative on 0v.
And both the positive Step and Dir wires are on a 24v and the Step and Dir neg wires are on the proper Y axis negs on the UB1.

The only real oddity I can think of is that I went ahead and solder bridged the Y104,105,106, and 107 to set up for the limits switches, so after hooking the emergency stop to Y103 (the I/O is set at Y102), and the Y HLFB positive to Y105, I jumpered the others to make the safety circuit continuous through 107 without actually hooking up the switches (which I don't have yet).



I took a couple of screenshots to expose my weak and frustrated underbelly.

Did you mean to say your HLFB was connected to X 105 You said Y but there is no Y105. Same with your E- stop button, you said Y104 but it should be on X 104.


Since your machine isnt in Re-set I dont think you have problem with your Estop circuit yet.

The UB1 like you said is very versatile. That makes it a little harder to configure. BTW that machine is huge. How big is it and what are you going to use for a vacuum source?

[kennected]

Ok, so I will switch the Y leads to another axis and try again. Maybe the UB1 is the problem?

I'll also move the emergency stop to Y104 and see what happens. Funny that you mentioned that, yesterday I found that the E stop only works right if it is configured as active high. Does that make sense? I looked at it like the circuit is hot (high) until the e-stop gets pressed causing the circuit to go low. The switch is definately a normally closed type. And yes, it does trip a reset if I press it. I will try it on 104 as well.

Ken

[kennected]

Thanks Jumper,
You are right, my mind is turning to mush on this. I should have said X not Y for the HLFB. You guys have been great, the least I can do is get it right instead of making this a riddle to solve.

The machine is aprox 6 feet wide and 10 feet long. A lot of hardware for less than $5,000. It seems like it deserves a new life with new electronics and Clearpaths, no? Haven't thought too much about the vacuum yet. I will likely find a turbine type but I'm always puzzled by the specs involved in vacuum pumps. I just finished installing this dust collector and it really does a nice job.

[1Jumper10]

Nice dust collector.

All I can think of Ken is to double check your connections:

Y step negative wire should be on "YS-" on the UB1 and your Y step positive wire on "VCC" Then in UCCNC Y axis configuration your Y axis step pin should be "pin 4, port 2"
Y direction negative wire should be on "YD-" on the UB1 and your Y direction positive wire on "VCC". In UCCNC Y axis configuration your Y axis dir pin should be "pin 5, port 2".
Since you have the flashing green light your enable circuit is probably working correctly. This is how I have mine configured and it is working.
It took me about 3 days of careful work to get mine working.

[kennected]

HEY, HEY!
My negative Step and Dir wires were backwards. I reversed them, no other changes, and just like that, Y moves. I must have mis-interpreted to Clearpath manual as to which one was which...damn.
You guys are awesome! I'll report in again soon, I'm wetting my pants right now and will make some progress today.

Ken

[1Jumper10]

Great!

[kennected]

I'l say!!
Thanks again, amigos.

So far I have hooked up all 4 motors and they are moving nicely. The slave X-A setup was easy. I simply set the DIR's as 1 active High and the other active Low. They all GGLLIIIIDDDDDDEEEEEE.
I'm calibrating right now and it seems to be simple enough.
I'm curious to how these compare to your motor settings, Jumper. At this minute I'm at: 1280 steps per unit, Velocity 750, and 50 for acceleration.

[1Jumper10]

Steps per unit depends on your final gear ratio and mine is 10:1 with a 40mm pinion gear. That said, my steps per unit is in the 3200 count range. Do you have the enhanced servo's or the regular models? I think I have my servo firmware set to 1600 counts per rotation. That makes a difference too. I was at 900 IPM and 80 accel, but I've since reduced it to 650.

750 IPM at 50 Accel, is pretty respectable. If you've got a good tune on them you'd probably be able to push to 1000 IPM comfortably. If the green light keeps blinking fast then you're not taxing the motor. When you push it and get a slow 3 green blinks, that's about the limit.

Glad you got them running. When you get ready to set up your OSSD/E stop circuit let me know. You can use K1 relay as your OSSD override switch.



Sent from my Pixel XL using Tapatalk

[kennected]

I bought the CPM-SDSK-3421S-RLN standard motors. Man, they sure sound nice compared to my nema 34 steppers on the other machine. Not that it means much but it lets you hear the racks and rails glide. It made me feel good that the slaved X setup was not a big deal. Yesterday I couldn't make a motor move and today I have 3 calibrated axis' running and I get to play with stuff like accelaration.

I have something to think about in the tuning. I was able to tune my Y axis no problem. I took my Y configuration file from the MSP usb connection and copied it over to both of the X motors. I have to say the paired motors sound identical to the tuned Y with only the exception that there is a little more unrelated rumble coming from the gantry's extra mass moving. I know what this pig gantry feels like when just heaving it by hand but it accelerates like it just doesn't care. Smooth and not a hint of racking. You could close your eyes and think there is only one motor involved. Now, performance might be a different thing, but I guess I'll cross that bridge soon enough. I did screw up once today. For one of the first jog moves of the gantry I had the jog feed at 100% and velocity at its stock 2000 units. It jumped hard for a split second, I was only tapping the button. I reduced the jog feed to 10% and it just took off again. No harm, no fowl.

So Jumper, is the relay safety circuit using K1 like using a relay shut down a group of pins? Is the idea that its quicker?

Where's a good place to buy new limits? Mine look like these:

but I would like them to be closer to 1/4" and if I understand correctly, they need to be NPN. These are 12mm. Too fat.

[1Jumper10]

I bought the Pepperrel and Fuch's set from CNC router parts. They're about a 1/2" in diameter so bigger than what you are looking for. There are lots of places that sell them. And the UB1 has slots for PNP and NPN sensors so you can get either. Factorymation sells good stuff and they have a variety Inductive Proximity
CNC drive is over in Hungary I believe, but they have the right sensors for CNC and they wrote the UCCNC software. Sensors - CNCdrive - webshop
This is the sensor I bought from CNC router parts: Pepperl + Fuchs Proximity Sensor | CNCRouterParts Go easy on yourself and be sure whatever sensors you get are normally open.

Ref using the onboard relay on the UB1; I'm away from my build book at the moment so I'm going by memory. The relays are there because I think most people use them as the start signal for their VFD's. I just used a Y output for my VFD start signal on mine so my relay wasn't being used. There is a solder bridge for that relay that links it to an output that can be controlled by UCCNC. Since it wasn't being used, and I need a switch for the OSSD override, I just wired my OSSD override into the normally open side of that relay and mapped an output trigger linked to the "override limits" software button found on the main screen of UCCNC. To recover from an E-stop, all I have to do is just click the "Override limits" software button, this closes the relay and thus the OSSD interlock is overridden. Well, that was the intent anyway. Mapping it to the "override Limits" SW button didn't work. It just toggled the relay and it needed to keep it latched. So I've been looking for a different output or HotKey I can map it to, to make it stay latched until the Reset button clears the E-stop condition. Maybe Gerry could suggest something

[kennected]

Now that surprises me, "normally open". In my way of thinking about limit switches, is the idea that there is a circuit of 1 or more switches providing current in if the high current is broken (switch see metal, wire gets cut, connection fails) the fault is registered as a fault. Kind of a fail safe way of thinking. All is good as long as high current flow is maintained. With normally open switches, the circuit current goes high when the switch closes and UCCNC goes into fault. On my home machine, my limit switches are all in a loop so if any one of them opens, (or other fault) there is a reset condition. I have seen some pretty heated debates here about switches and I understand that some are attempting to engineer some real fine homing accuracy, but in my simple mind, the purpose of a limit ****ch is to first stop everything. If one wants to multi-task their limit switches to do homing, they may have a point with really expensive switches but for the purpose of limits, it seems like anything that reliably interupts this safety circuit would suffice. What am I missing with my logic?


Also, if my existing switches are PNP (delivering high current upon activation, you're saying I could use X310-314, though they would no longer be part of the "safety circuit".

[1Jumper10]

Thats one way to do it. Personally, I didnt want my home sensors as part of my OSSD circuit just because if any of the inputs of the OSSD fail, the OSSD cuts out and shuts everything down. This would include a limit sensor if its tied to the OSSD. I never want my limit sensors disabled. I have (eventually, ) hard mechanlcal, travel stops at the limits of each axis that function as the last barrier to injury and extreme anguish . And software limits would be in place and take control before the hard stops.

I think you have a better understanding of NPN vs PNP than I do. I can follow a connection diagram and get them to work but beyond that I'm lost.

I wanted an input for each sensor so I can see its status on the diagnostic screen. Thats a nice thing about the UB1/UC300ETH, lots of inputs. If all your sensors are daisy chained in series to one input you lose that capability. I can test each sensor with a piece of metal and see that it is functioning as expected by watching the diagnostic screen. And each sensor has a local LED.

Its personal preference. Just my 2 cents.

[ger21]

What am I missing with my logic?

Nothing?

My belief is that your limit switches are normally closed, mechanical switches, in series. Your Home switches should be separate switches, inside the limits. This is how all of the industrial machines that I've used work. If you have a limit switch failure, the machine stops.

[ger21]

There is a solder bridge for that relay that links it to an output that can be controlled by UCCNC.

There's a solder bridge that optionally has the charge pump activate the relay, is that what you are talking about?
By default, all 3 relays are controlled by outputs.

Since it wasn't being used, and I need a switch for the OSSD override, I just wired my OSSD override into the normally open side of that relay and mapped an output trigger linked to the "override limits" software button found on the main screen of UCCNC.

So, you have a wire from SafetyOVR to NO3, and from CM3 to 0V?

Do you have the Output Trigger mapped to LED 27, and set to Port 3, Pin 17?

Here's the issue I think.
1) If you actually hit a limit, until you jog off the limit, the OSSD output will keep triggering the Reset, so you kinda have a loop where it Overrides it, but then it instantly gets tripped again.
2) The Limit override only stays on while a limit is triggered. If you use an Estop, you can't turn on the limit override.

What I would try, is to use a macroloop, and a custom LED. I don't know the specifics of your setup to go into more detail. But you'd have the macroloop turn on when the limit or estop is hit, Have the output triggered by the state of the custom LED.

[1Jumper10]

Gerry,

I'm using relay 1 so I have a wire from SafetyOVR to NO1, and from CM to 0V.
And I have the Output Trigger mapped to LED 25, and set to Port 1, Pin 1.

I Initially tried mapping it to LED 27 but it behaved exactly like you described. Using LED 25 works pretty well except that it clears the E-stop condition very quickly. Watching the diagnostics page when a limit is hit or the Reset software button is pressed, the E-stop LED illuminates for ~ 1/2 second then clears. I tried looking for a more suitable LED to use but found none. I was thinking about a HotKey that would latch Relay1 but couldnt quite figure it out. I still get the interlocking feature of the OSSD with this method and it's easy to recover from Reset. But it sounds like a macroloop - custom LED is probably the better solution. However, I dont know how to write a MAcroLoop. Concerning HotKeys, if I made a custom LED and assigned a HotKey to trigger it, and then assigned an output trigger to that custom LED to close Relay 1, would that work?

[ger21]

I think there's an easier way. I'll look into it tonight when I get home.

[kennected]

On your commercial level, do you also use soft limits and are they setup inside your mechanical limit switches? I have rarely turned mine on but at least in Mach, they sure have a nice method of slowing down the potentially runaway axis before hitting the abrupt result of a limit switch.

Also, I'm trying to understand why UCCNC can't be set up to home to a switch and then backup till it clears? Or is this not what is being discussed here? Is it simply the inability to clear a limit switch fault long enough to jog off of it?

[1Jumper10]

UCCNC backs off the homing switch automatically. There's nothing really to set. It homes to the sensor then, backs off about a half inch.

I was discussing using one of the UB1's on board relay to function as its own OSSD override switch. If the OSSD safety circuit is going to be used as designed, it needs an override switch to recover from a shutdown. The on board relay is a convenient switch to use, it just needs a good way to actuate it. Thats what Gerry and I were discussing.