Let the journey begin

[mvdveer]

3 minutes ago, smirk said:

Alexa-connected-colour-changing LED bulbs

@smirk Who's Alexa, Oh now I remember, she's Siri's sister 

[smirk]

Sssshhh.....don't say her name, she'll go and buy something!

[mvdveer]

Hey Siri......  Alexa.....   Move my Voron

 

Its alive

 

 

[Buurman]

Its alive!!!!! 

[smirk]

Wonder how long before it starts learning at a geometric rate and becomes self-aware.......I always did say Vorons would make excellent robotic overlords.

 

[mvdveer]

On 6/28/2022 at 11:56 PM, claudermilk said:

Be sure to check out the Steve Builds stream where he installed one: https://www.youtube.com/watch?v=hpS2G_02OlQ It worked out in the end, but the macros were difficult.

@claudermilk you are sooooo right. The macros are a pain in the proverbial. Gonna rewatch Steve's build, do some more reading and figure it out. Can pickup the probe from the dock, but that's about all. All the other macro's throws errors. Watch this space

[mvdveer]

Now, I am either stupid or tired or both. Cannot get the probe to function. Connected via Stealthburner PCB to the BTT Octopus stop 7 (PG15) . When queried probe shows triggered as it should. Picking up the probe, it still shows triggered, not open. If I invert the pin (!PG15) the probe shows open and when picked up shows triggered, but that won't work as the M401 expects the probe to be triggered before pick up.  Have replaced the probe and PCB - same results. What am I missing? Everything else working smashingly

 

Have tried a different end stop (PG11) No luck either

Edited July 16, 2022 by mvdveer

[smirk]

Ok, so it's not a wiring fault as the behaviour changes, it's not a faulty port as the issue follows to a different port. Part of me wants to say pullup resistor but I'll need to think about the logic of that

[mvdveer]

On 7/16/2022 at 8:22 AM, smirk said:

Ok, so it's not a wiring fault as the behaviour changes, it's not a faulty port as the issue follows to a different port. Part of me wants to say pullup resistor but I'll need to think about the logic of that

Think you may be on to something. According to the documentation on the Euclid site: 

Quote

If Euclid Probe is to be wired to the Endstop Port (pin PG11), you will need to use the BAT85 diode and set the probe pin HIGH in order for the firmware to recognize the operation (^PG11). See Section 2.7 in the Octopus User Guide under Probe Port Wiring

Set the Pin  in "high" - no difference - pull up resistor seems the logical answer - but why.

Might print and setup a click and see what happens

[smirk]

I'm presuming the BAT85 diode is inside the detachable probe/head part (so only enters the circuit when the probe is picked up).

I'll need to go and look at the wiring for the euclid, been a while (like my soldering) since I've mucked around with pull-ups and switching. It's simply enough but only when you have the brain cells. Guess it all depends whether the input is triggering on a high or low signal (inaccurately +v or 0v) and how the circuit you've got wired to the input is hooked up.

For a system you've total control over you pick/decide which way your inputs trigger but not sure if klipper/octopus has some set notions of whether an input will only trigger on a high signal or not (need to read up).

[mvdveer]

The bat diode is on the PCB. Been mucking around with it all day. Going to call it a day. 11pm here on Oz. and I’m grumpy. Thanks for the help
 

 

[smirk]

Scratching my head about this. (sorry some of this is me "thinking out loud")

The diode makes sense to me - I was forgetting that the tool head/PCB would be 24v and the Octopus inputs would only be 3.3V logic. However, one thing that is troubling me now that I've thought about it more, was your:

2 hours ago, mvdveer said:

If I invert the pin (!PG15) the probe shows open and when picked up shows triggered,

Which was different to the operation/status when it was just PG15 (i.e. always triggered). That shouldn't have happened you should have only ever got "open/open" the inversion is just that, a logic NOT operation. That's weird.

1 hour ago, mvdveer said:

(^PG11).

Bit like you I'm struggling with this (or why it make no difference for you). I don't think it determines how the pin triggers (high/low) it simply governs what the pin sees as it's default state. My simple thinking is this "^" (caret) activates a pull-up resistor on input PG11 which in turn makes sure it is permanently connected to +3.3V (logical HIGH signal). This permanent connection stops the connection floating and randomly deciding on input signal. The fact there's a [pull-up] resistor that prevents excessive current flow (short-circuit) when you temporarily connect to Ground (logical LOW signal) when you activate the switch (which is NC). You need a pull-up resistor, as much of the time the probe is not connected so the input pins are ultimately hanging in the wind so would be otherwise floating. The logic of pull-ups makes sense to me (now I've thought about it).

I'm maybe making a dangerous assumption here: that is there is essentially no difference (circuit-wise) between the euclid probe not being connected (picked up) and the probe being connected with the microswitch activated (clicked/pressed). In other words the euclid probe doesn't have some other mechanism for knowing if it's simply picked up the probe independent of the actual activation state of the switch?

They've got a nice circuit simulation on the Euclid website and that clearly shows (for the 24v setup) that with the switch open (akin to the probe not being connected) that the input pin sees a logic high (+3.3v) signal. Close the switch, it's default state (since it's NC) and the input pin sees a logic low (0v).

That's nice,a (moving) picture is worth a thousand words. Now, I might've been tempted to ask if the probe's switch was actually set to NO rather than NC or if it was the 5V part rather than the 24V part but......

I keep coming back to that non sequitur with the inversion of PG15 and the difference in states between the application of the inversion (!) Unless that's just a coincidence and the difference is caused by a loose cable/magnet?

You've probably already done it, appreciate you spent most of the day on this, but I'd have the multimeter out and checking what electrons were falling out of the wires at the Octopus end and just manually connecting/disconnecting the probe.

 

Sorry that's really rambly and probably hee-haw help, other than to say "you're not going mad"

[mvdveer]

7 hours ago, smirk said:

 

 

7 hours ago, smirk said:

Sorry that's really rambly and probably hee-haw help, other than to say "you're not going mad"

No that is very helpful.

Also it is a 24V Euclid board and my thinking is the PCB only supplies 5V to the probe - should not make a difference to the function. According to the documentation:

"Octopus can use either the standard 5V or 24V Euclid models whether Euclid Probe is going to be connected to an open endstop connector or 24V capable probe port."

 

Depends how you read this it may mean 24V probe needs a 24V capable port and 5V board the end stop port. 

Going to rewire the probe and bypass the PCB. Currently t is wired as follows:

 

Thanks again for the efforts

7 hours ago, smirk said:

That's nice,a (moving) picture is worth a thousand words. Now, I might've been tempted to ask if the probe's switch was actually set to NO rather than NC or if it was the 5V part rather than the 24V part but......

Thank you - I think you are on to it

[smirk]

34 minutes ago, mvdveer said:

Depends how you read this it may mean 24V probe needs a 24V capable port and 5V board the end stop port. 

I would read it as the 24v (euclid) probe can "survive" 24v rather than needing it as such. It's just a switch (so beyond rating) it wouldn't care, well I guess there's LEDS but they're <2V. Suppose, I was thinking if it was a 5v board on a 24v set up then something has burned out (bit like the straws I'm clutching). As far as I know, the higher rated ports on some MCUs are for those power hungry inductive type probes that generally operate (18v-24v).

I guess I was thinking they provided a 24v (safe) version if you only had the "inductive" probe connector on your MCU to use (guess not every board is as blessed as the spider or Octopus inputs) or (as in your case) you're using a toolhead PCB, but looking at that diagram (of the PCB) it is just an extension of the higher voltage inductive probe port. It's late here, so my brain is shutting down.

Its probably a good step to by-pass the PCB and connect the Euclid direct to the "regular"end-stop port (the 24v board will work at 5v (Well I'm sure it's really 3.3v), perhaps LEDS on it might be dimmer I guess.

 

I just realised, did you wake up early or have I been awake long enough to "see" a new day down-under? Clearly I need to go to bed,

Good luck with the rewiring.

[mvdveer]

10 hours ago, smirk said:

Its probably a good step to by-pass the PCB and connect the Euclid direct to the "regular"end-stop port (the 24v board will work at 5v (Well I'm sure it's really 3.3v), perhaps LEDS on it might be dimmer I guess.

Did and all working as expected. Must have been something on the PCB. Rewiring was NOT fun. Thanks for all your help @smirk. On with the build.....

[smirk]

Glad you got it working despite the extra wiring. I presuming it must be a design "flaw" with the PCB rather than a manufacturing flaw as you swapped the PCB and got the same results (and it's meant to be a valid config). 

 

[Buurman]

Remember when I said.. the more electronics you add...........  

Always shit to run into these things when a certain combination doesnt work as expected and it hard/impossible to find kindred spirits running into the same thing...

But... thats the price you pay for being a entrepreneur   Good fix though...compliments, also for smirk!! 

[mvdveer]

15 minutes ago, Buurman said:

Always shit to run into these things when a certain combination doesnt work as expected and it hard/impossible to find kindred spirits running into the same thing..

@Buurman That's beauty of this forum. Filled with people a lot more knowledgeable than I am, and more importantly - always happy to help. 

16 minutes ago, Buurman said:

But... thats the price you pay for being a entrepreneur

Entrepreneur, don't think so, just like a challenge. And I have another one which I am currently reading up on. Printer homing well, picking up probe, QGL good but the Bed Mesh.... ARGGGH - got an error: " horizontal_move_z can't be less than probe's z_offset" which I am currently reading up on in hope to find a solution. Only way we learn...

[mvdveer]

Build coming along nicely but not without its frustrations. First it was the toolhead PCB that did not activate the probe correctly.

Love how the Voron Parts logo turned out - at least something turned out as it should have

Edited July 18, 2022 by mvdveer
Formatting

[Buurman]

post you bed_mesh part from the config.php please.

Thats where this part is I think, I will add my part from bed_mesh.

NOTE, the part under this, the quad_gantry_level is grayed out, since thats what is the advise using Klicky, but I dont know how that is for your MOD.

I use a 350 bed, I think you did too.

[bed_mesh]
#special note the PROBE is offset approx 22 to 25Y to the nozzle take this into account for bed size so you dont probe over the bed edge
speed: 300
horizontal_move_z: 10
mesh_min: 40, 40
#45,25 matches the QGL consider the probe offset of 22mm add it to Y 275+22=275 same as QGL
mesh_max: 310, 310
#matches QGL
probe_count: 5, 5
algorithm: bicubic
#relative_reference_index = ()(x points * y points) - 1) / 2
#3x3 mesh = 4
#5x5 mesh = 12
#7x7 mesh = 24, etc
relative_reference_index: 12

move_check_distance: 3

algorithm: bicubic
fade_start: 1
fade_end: 10
fade_target: 0
split_delta_z: 0.0125
mesh_pps: 2,2
bicubic_tension: 0.2

 

Edited July 18, 2022 by Buurman

[mvdveer]

2 minutes ago, Buurman said:

post you bed_mesh part from the config.php please.

[bed_mesh]
speed: 50 
mesh_min: 40, 40
mesh_max: 260,260
fade_start: 0.6
fade_end: 10.0
probe_count: 11,11
algorithm: bicubic
relative_reference_index: 60
 

Aha, now I see it:

Have no :"horizontal_move_z: "

[mvdveer]

Thanks @Buurman - got it all working now. See - someone with more knowledge - right on the mark

[Buurman]

It a tricky part this, had to read up on it also when I wanted to adjust the amount of probes, and how...

You now do 11,11. thats a huge amount if you ask me, but its good for the fisrts time and you want to check your bed level.

Keep in mind that the textured plate isnt "flat" you can turn it around to get a better view of bed level.

I myself use 5,5 and ref point 11 (now to think of it, maybe it should be 13 I still dont fully understand that part) 

I think having over 100 ref points is just overkill, but thats me

 

[mvdveer]

Just now, Buurman said:

You now do 11,11. thats a huge amount if you ask me, but its good for the fisrts time and you want to check your bed level.

That's just for the initial run - then I'll switch to 5,5. Thanks for your help - saved my hours of reading

[Buurman]

Happy to help! like you did with me also  and why I have squash balls now  hahaha