#hazzy Logs

Jul 04 2020

02:11 AM Joco[m]: Lcvette: I hope you are right. I think my steps are on the money. I was doing tests over 75mm travel (metrics version of 123 block). And I was getting the DTI to go from 0 to 0 at the end of travel.
02:11 AM Joco[m]: I believe I had backlash comp at 0 when doing those adjustments.
02:14 AM Joco[m]: pockets are square/rectangle. No circles
02:18 AM Joco[m]: Lcvette: I'll recheck the scale and travel again.
02:18 AM Joco[m]: with the backlash value commented out
05:19 AM JT-Cave: morning
05:22 AM JT-Cave: say you issue Y-2 and you have backlash on X, LinuxCNC does not compensate for X+ or X- as it has no clue what side of the tool is cutting
07:10 AM Lcvette: It should take up lash at ant axis directional change
07:10 AM Lcvette: * It should take up lash at any axis directional change
07:11 AM Lcvette: Doesn't have to know what side the tool is cutting on, it's just removing slack in the mechanical drive system
07:12 AM Lcvette: If you have you cam heart m5 machine take up the lash,
07:12 AM Lcvette: Anyone or changes direction it had a fast little blip
07:12 AM Lcvette: Then returns to normal stepping
07:13 AM Lcvette: But again, this is mach 3 and i only just tried blc on linuxcnc in the large and have boot used it much to notice if it's got an issue
07:16 AM Lcvette: * Any directional changes it has a fast little blip to catch up the axis slack
07:18 AM Lcvette: * But again, this is mach 3 and i only just tried blc on linuxcnc on the lathe and have not used it much to notice if it's got an issue.
07:24 AM JT-Cave: it can't remove slack only offset that axis
07:25 AM JT-Cave: the mechanical play will always be there, the software can't remove that
07:51 AM Lcvette: * The mechanical play remains, but the directional adjustment is compensated
07:52 AM Lcvette: It is of course better to have as little as possible or no backlash, but if you do blc should adjust for it
07:52 AM JT-Cave: which side get compensated for X when Y is moving?
07:54 AM Lcvette: If cutting forces over comes the friction on the dovetails then it will move around in the excursion lash, but on these small machines the dovetails tend to hold position due to stiction
07:54 AM Lcvette: * If cutting forces over comes the friction on the dovetails then it will move around in the existing lash, but on these small machines the dovetails tend to hold position due to stiction
07:55 AM Lcvette: In that scenario the last positioning move in y should put the y acis in position correctly
07:55 AM Lcvette: * In that scenario the last positioning move in y should put the y axis in position correctly
07:57 AM Lcvette: Whether or not y then wiggles while cutting, that is what it is and can only be corrected with better mechanical components
07:58 AM Lcvette: But like i mentioned the poor fitting high stiction dovetails on the g0704/bf20 machines tends to act like a lock
08:09 AM Lcvette: A great example is i never use backlash comp on the z acis because the weight of the head keeps the slide resting on the ballscrew
08:13 AM JT-Cave: but what about the X axis during a Y move?
08:19 AM Lcvette: Why would x need to be compensated for during a y move?
08:23 AM JT-Cave: because the force of cutting could be on the X from one side or the other
08:25 AM Lcvette: I agreed with you on that
08:26 AM Lcvette: But as mentioned, the stiction is so high on these dovetail machines, it rarely is enough force to move the table or y saddle once in position
08:26 AM Lcvette: And when finishing to size the cut force it's generally very low
02:02 PM Lcvette: lcvette has had twin 5 year old girls, mother in law and sister inlaw at his house for 7 days now..... its been loud to say the least
03:13 PM Joco[m]: Lcvette: that sound busy
03:16 PM Joco[m]: Lcvette: JT-Cave re the backlash discussion and movement in X and Y: I have observed that the CAM generated by fusion often has curved lead ins that if you tweak the parameters a bit can help ensure that backlash on the nonmoving axis has been tacken up as part of the entry move into a cut.
03:37 PM Joco[m]: Lcvette: and as you said the finishing cuts tend to be light.
03:42 PM Joco[m]: Lcvette: remeasured my X travel and I think I am loosing steps as I am not getting a consistent reading. So back to tweaking the gibs
03:42 PM Lcvette: /o\
03:42 PM Lcvette: losing steps is no fun
03:42 PM Joco[m]: Its not much, must be at the beginning when overcomming the stick. my variation is ~0.1mm
03:42 PM Joco[m]: or even 0.2
03:44 PM Joco[m]: the only thing that makes sense is sticking in the gibs when I measure and adjust scale then it loosens up and I get some readings with no stiction and its going longer than it shoudl.
03:44 PM Lcvette: are you taking up the slack before the move?
03:44 PM Lcvette: steps per should be set wihtout blc on
03:44 PM Lcvette: and with the backlash fully taken up
03:44 PM Lcvette: before making the distance move
03:44 PM Joco[m]: yes. Always move 80-85 back so can come forward to starting position
03:45 PM Lcvette: then something is fishy because it should definitely be consistent
03:45 PM Joco[m]: and doing stuff with backlash commented out
03:45 PM Joco[m]: hence my stiction suspicion. That would account for the variability
03:46 PM Lcvette: stiction should not be an issue over 75mm
03:46 PM Lcvette: it is more an issue on much shorter moves
03:46 PM Joco[m]: it will be it you loose some steps at the start
03:46 PM Lcvette: that allow the screw to bind before breaking free and registering
03:46 PM Joco[m]: wouldn it?
03:47 PM Lcvette: if your axis arelosing steps on an unloaded move, you got issues
03:50 PM Joco[m]: yeah - but what sort.
03:50 PM Joco[m]: I had changed to 10x micro stepping as the digital drivers supported it.
03:51 PM Joco[m]: Maybe I need to stick with the "traditional" valie. And flick back to 8x
03:52 PM Lcvette: what are you at?
03:52 PM Lcvette: microsteps do not offer fantastic precision
03:52 PM Lcvette: the only precise stepper move is a full step
03:53 PM Joco[m]: I have 5mm pitch running 10x so that turns into 2000 steps per turn
03:53 PM Joco[m]: motors are direct connect to ball screws
03:53 PM Lcvette: have you tried just setting it to that and seeing what you get?
03:54 PM Joco[m]: setting to what? 8x?
03:55 PM Lcvette: setting to the screw pitch calculation
03:55 PM Lcvette: no adjustment
03:56 PM Lcvette: maybe its right
03:56 PM Lcvette: set it there and run a test
03:56 PM Joco[m]: so 400 scale. Yes. But probably need to do it and see if I get variation
03:56 PM Joco[m]: I tihink I measured, adjusted then started looking at repeatability
03:58 PM Joco[m]: will set back to 400, make sure its all lubed up and then look at repeatability with stock setting
03:59 PM Lcvette: Try just setting it to the screw pitch
03:59 PM Lcvette: Without adjustment
03:59 PM Joco[m]: so when you say that do you mean 2000 steps / 5 = 400 on the scale?
04:00 PM Lcvette: Yes
04:00 PM Joco[m]: cool - we on same page
04:00 PM Lcvette: Leave backlash comp off
04:00 PM Lcvette: Run test part
04:00 PM Joco[m]: yup - have it commented out in the ini file
04:00 PM Lcvette: Should be under sized
04:00 PM Lcvette: The amount under sized make your blc
04:01 PM Lcvette: Rerun test
04:01 PM Joco[m]: isn't that adjusting the travel amount using blc? When do I adjust the travel scale?
04:01 PM Lcvette: No
04:02 PM Lcvette: Try just screw pitch
04:03 PM Lcvette: By doing it that way you will eliminate any stiction because the cutter is constantly moving it doesn't stop
04:03 PM Joco[m]: ok - so if I do that and the internal pocket test is large what does that tell me
04:04 PM Lcvette: one step at a time
04:06 PM Joco[m]: yeah - but wanting to know what to look for as you might be asleep by the time I get to run all these tests. Have a number of jobs to do today. Else SWMBO will rain full fury on my head
04:07 PM Lcvette: the inner and outer pockets should be off the same amount
04:07 PM Lcvette: remember you are cutting using no backlash comp for the first test
04:07 PM Joco[m]: yup
04:08 PM Lcvette: and use a finish pass that is light
04:08 PM Joco[m]: yup
04:08 PM Lcvette: also check the runout on your tool
04:09 PM Lcvette: i just helped resolve an issue on another forum that the user had 0.007" total runout at the tool
04:09 PM Joco[m]: ok - DTI on the tool flutes?
04:09 PM Lcvette: the shank is fine
04:10 PM Lcvette: it should give you an idea what the tool runout is
04:10 PM Lcvette: on mine i have about 0.0003" runout at my tools
04:11 PM Lcvette: you can adjust the tool diameter to account for the runout amount
04:11 PM Joco[m]: k
04:11 PM Lcvette: that should eliminate that from your final results
04:12 PM Joco[m]: ok - off to get my test part setup in vice
04:12 PM Lcvette: what is a light pass to you?
04:12 PM Lcvette: for me i finish at 0.005"
04:12 PM Joco[m]: I'm running 0.04mm chip load with a 0.2 to 0.3mm step over
04:12 PM Lcvette: if you are running er collets you can take some of the tool runout out with some adjusting
04:13 PM Lcvette: that shjould be ok
04:13 PM Lcvette: 0.2
04:14 PM Joco[m]: ok - I had been using 0.1 but was wondering if starting to rub.
04:14 PM Lcvette: i never go under 0.005"
04:14 PM Lcvette: 0.127mm
04:14 PM Joco[m]: I switched to .2 to .3 and started getting a nice fine chip
04:15 PM Joco[m]: ok - i'm doing roughly sensible stuff then
04:15 PM Lcvette: doing good!
04:15 PM Lcvette: ok go bang outn those tests!
04:15 PM Lcvette: standing by
04:18 PM Joco[m]: In the shop. Tally ho.
04:51 PM Joco[m]: quick run out check is 0.02
04:52 PM Lcvette: Mm?
04:53 PM Joco[m]: on the tool
04:54 PM Joco[m]: so tweaked its diameter, got part in vice and zero set. About to have a crack with 400 scale and no backlash
04:56 PM Lcvette: Great
04:57 PM Lcvette: Stranding by
04:58 PM Joco[m]: starting
05:05 PM Joco[m]: measurement tim. Target on X is 76mm for outside and 41mm for inside.
05:05 PM Lcvette: Ok
05:06 PM Lcvette: Done yet?
05:06 PM Lcvette: That's a big test
05:08 PM Joco[m]: 76.12mm for outside
05:10 PM Joco[m]: 40.90 or 40.92mm on inside
05:11 PM Joco[m]: oops - inside wrong.
05:11 PM Joco[m]: 41.12
05:11 PM Lcvette: So tippy are oversized same amount
05:11 PM Lcvette: That's more normal
05:12 PM Lcvette: * So both are oversized same amount
05:12 PM Joco[m]: yup
05:12 PM Joco[m]: between 0.1 and 0.12mm oversize
05:13 PM Lcvette: In testing usually they would be undersized
05:14 PM Joco[m]: double checked and 400 steps per unit and backlash commented out.
05:14 PM Lcvette: Ok try backlash comp at 0.12
05:15 PM Joco[m]: ok
05:15 PM Lcvette: I'm guessing this will make our worse though
05:16 PM Lcvette: * I'm guessing this will make it worse though.. usually backlash causes undersized parts
05:18 PM Joco[m]: would it be half this? Since that error is caused by both sides? I am cutting the left and right sides here. Not just one side
05:18 PM Joco[m]: the left and right sides of a square.
05:19 PM Joco[m]: * the left and right sides of a rectangle.
05:20 PM Lcvette: Can try
05:20 PM Joco[m]: ok - will give it a blast
05:20 PM Lcvette: Like i said this method usual leaves an undersized part
05:21 PM Lcvette: And then blc brings the tolerance out
05:21 PM Lcvette: But I'm curious to see back to back results here with blc
05:22 PM Lcvette: Just take a skim off outside and inside existing pockets
05:23 PM Joco[m]: Skim? Was reducing dimension by 1mm. Is that ok?
05:23 PM Lcvette: Good news is they are both of same side of target
05:23 PM Lcvette: Yes but isee same finish pass
05:23 PM Lcvette: * Yes but use same finish pass
05:23 PM Joco[m]: Yup two finish passes of 0.02mm each.
05:24 PM Joco[m]: Was in last test as well.
05:24 PM Joco[m]: Wish I could change blc without having to stop start lcnc
05:25 PM TurBoss: hello
05:25 PM Lcvette: o/
05:25 PM Joco[m]: Hi TurBoss
05:26 PM Joco[m]: Off we go. Test 2
05:26 PM Lcvette: Great
05:26 PM TurBoss: Joco, Lcvette !
05:26 PM Joco[m]: 75 and 42mm targets.
05:28 PM Lcvette: What's crackin turboss? Get your pot working?
05:30 PM TurBoss: yes got the pot working
05:30 PM TurBoss: I'm playing with the serial DRO
05:30 PM Lcvette: Hurray! \o/
05:30 PM TurBoss: is USB moving to RS232
05:33 PM * TurBoss uploaded an image: VectorImage_2020-07-05_122348.jpg (424KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/ibzAFNKpiIKEECqhLCQhBuDQ >
05:36 PM Joco[m]: Lcvette: targets of 75 and 42. Got 75.06 and 42.05
05:36 PM Joco[m]: that was with blc = 0.06
05:36 PM * TurBoss uploaded an image: VectorImage_2020-07-05_122700.jpg (286KB) < https://matrix.org/_matrix/media/r0/download/jauriarts.org/PvNlYCuStFWnjJqjjfQQJEmc >
05:36 PM Joco[m]: going to try blc=0.12
05:36 PM Joco[m]: TurBoss: that looks dang cool
05:37 PM TurBoss: was driving the mgp and exis selector
05:37 PM TurBoss: now only dro
05:38 PM Joco[m]: a little arduino nano/mini the heart of the beast?
05:38 PM Lcvette: So go the whole 0.12
05:38 PM TurBoss: yes
05:38 PM TurBoss: xD
05:41 PM Lcvette: Awesome turboss
05:42 PM Joco[m]: Lcvette: yup. Just setup for it. Hitting go
05:43 PM Lcvette: Great!
05:52 PM Lcvette: Talk to me goose
05:52 PM Joco[m]: ok ....
05:53 PM Joco[m]: taregts 74 and 43mm. Got 74mm and 43.1
05:54 PM Joco[m]: in fact I think its 43.12
05:57 PM Lcvette: so one is good and one id off
05:57 PM Joco[m]: yup
05:58 PM Joco[m]: wondering if should run test again and see if consistent
06:01 PM Lcvette: Yea
06:01 PM Joco[m]: rgr - setting up
06:02 PM Lcvette: Is it square in both axis?
06:03 PM Joco[m]: not been doing the Y tests
06:03 PM Joco[m]: * not been doing the X tests. No test for cuts based on Y postionign
06:03 PM Joco[m]: * been doing the X tests. No test for cuts based on Y postionign
06:04 PM Lcvette: Oh I thought you were making like pockets and I'm outside square
06:04 PM Joco[m]: have been. But have just been changing the X dimension at the moment.
06:05 PM Joco[m]: working on the princple of "one thing at a time"
06:06 PM Joco[m]: lets see if get the samne result oafter this test
06:09 PM Lcvette: Ok
06:10 PM Lcvette: Standing by
06:13 PM Joco[m]: 73.98 and 44.14
06:13 PM Joco[m]: 0.02mm variation
06:13 PM Joco[m]: * 0.02mm variation which could be measuring to be honets
06:14 PM Joco[m]: * 0.02mm variation which could be measuring to be honest
06:17 PM Joco[m]: so pretty much the same.
06:19 PM Joco[m]: not sure where that gets things. :-/
06:20 PM Joco[m]: ¯\_(ツ)_/¯
06:28 PM Lcvette: Go back to 0.06
06:29 PM Joco[m]: seems to be a happier medium
06:31 PM Lcvette: Maybe try adjusting steps per from there
06:32 PM Joco[m]: will do the same process on Y now. Then see where lands. Just have a quick late breakfast
06:32 PM Joco[m]: got a big hungry
06:33 PM Lcvette: Lol
06:33 PM Lcvette: Second breakfast?
06:33 PM Joco[m]: nah - didn't have one
06:33 PM Joco[m]: I get focused on something and food gets forgotten abou
06:33 PM Joco[m]: * I get focused on something and food gets forgotten about
06:33 PM Lcvette: Lol
06:35 PM Joco[m]: reminds me - need to remember to go up to the lockup and get some alisheet. Have a repair I'm doing to someones fishtank. Its a BIG tank. about 1.5m long and 600mm wide, 600mm deep I think.
06:35 PM Joco[m]: Anyway - need to get my aluminium welding mogo back. Not practiced in months.
06:35 PM Lcvette: Nice
06:36 PM Lcvette: So done for the day?
06:36 PM Joco[m]: Nah - going to sort the Y axis next
06:36 PM Joco[m]: once hit lunch will need to pack it in and sort out other jobs. 11:30am here
06:38 PM TurBoss: DRO installed and working!
06:38 PM TurBoss: no more USB stuff
06:38 PM Joco[m]: your are charging along TurBoss
06:39 PM Joco[m]: I might have found a surface grinder. :D
06:39 PM TurBoss: yes?
06:39 PM TurBoss: great
06:39 PM * Joco[m] uploaded an image: image.png (12887KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/JvLfBbQrtvkDOqpbVoibljlu >
06:39 PM Joco[m]: going on auction along with a bunch of other stuff on 18-July
06:40 PM TurBoss: sek I need to config some stuff in my matrix server
06:40 PM TurBoss: I can't download stuff bigger than 10Mb
06:40 PM Joco[m]: ok
06:41 PM Joco[m]: and If I could figure out where to put it:
06:41 PM * Joco[m] uploaded an image: image.png (12766KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/SqHLOlGMuoSXhaFbLCuZYrbt >
06:45 PM TurBoss: nice!
06:47 PM Lcvette: Joco: nice
06:47 PM Lcvette: @Turboss nice!
06:47 PM Lcvette: * TurBoss nice!
06:48 PM Joco[m]: Got to win one of them yet. But I have a chunk of cash and a lot of people are feeling nervous in current economic climate so I’m hopeful.
06:49 PM Joco[m]: There is a bunch of mills and lathes and tool and cutter grinder that make my fingers itch. But I need to realistic one space.
06:50 PM Joco[m]: The surface grinder is what I really want. It opens up a bunch of options for precision and surface finish I can’t do at the moment. A grinder do the lathe would open up round grinding.
06:51 PM Joco[m]: Something like this would be good: http://www.scottmachinery.co.nz/used.php?s=300&ss=700&scode=A-LATH-1-0101
07:03 PM Lcvette: Yeah grinders are cool, but everything i wanna grind is big..lol
07:04 PM Joco[m]: heh - my target size is more in the 300x150mm range
07:06 PM Lcvette: 1000x400 would be what i wanted
07:06 PM Lcvette: :D
07:19 PM Joco[m]: that big
07:19 PM Joco[m]: right - starting Y tests now
07:32 PM Joco[m]: Lcvette: tested Y. Pretty good. Targets were 55mm outer and 33mm inner. Got 54.95 and 33.03
07:32 PM Joco[m]: Going to run another test before adjusting anything.
07:32 PM Lcvette: With what blc setting?
07:38 PM Joco[m]: no blc
07:40 PM Lcvette: Did you set steps top 400?
07:41 PM Joco[m]: yup
07:48 PM Joco[m]: ok - feel this is more representative. targets: 54 and 34. Got: 53.98mm and 33.95mm
07:50 PM Joco[m]: so I think maybe 0.02 on blc will be plenty to modify for small veriance
07:55 PM Lcvette: Yup
07:55 PM Lcvette: Think this is closer than you were
07:56 PM Lcvette: Needed to dial x in
07:57 PM Lcvette: I think the steps per can get your base tolerance closer
08:08 PM * Joco[m] uploaded an image: ima_a696901.jpeg (75KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/PImrdxrqWJznVPGuVSpsjHrW >
08:09 PM Joco[m]: Lcvette: target on left, actual on right. X is my problem child.
08:09 PM Joco[m]: well the outer measure on X. The inner one was pretty good.
08:11 PM Lcvette: Try adjusting steps
08:12 PM Joco[m]: Lcvette: as you say, perhaps some tweaking to the steps per unit will help. Any specific rules on that? e.g. making sure the scale is not some horrible odd fraction.
08:23 PM Lcvette: Just play with it until you get closer
08:23 PM Lcvette: Remember single nuts have about 0.002" backlash
08:23 PM Lcvette: So if you go beyond what that would fix your moving away from target
08:24 PM Lcvette: I m still surprised taking proper measurements for scale and backlash gave such bad results
08:24 PM Lcvette: It shouldn't have
08:24 PM Lcvette: But because it did.. this method might get you closer
08:25 PM Lcvette: Even though it's kind of wonky
08:25 PM Lcvette: Wad hoping the screw pitch scale and blc would get you there
08:25 PM Lcvette: Seems it did on y
08:25 PM Joco[m]: Cheers. I’ve been pondering if I shouldn’t fit my 5u scales and use them with dro to help.
08:25 PM Lcvette: Wonder why x is such a butthead
08:26 PM Joco[m]: not sure. Might need to take it off and look at all the parts.
08:26 PM Joco[m]: Maybe it’s not clamping well into the saddle.
08:26 PM Joco[m]: y carriage thingy
08:26 PM Lcvette: could be
08:27 PM Lcvette: You could also try taking a cut then move take another cut and measure
08:27 PM Lcvette: Start with that's axis kash taken up
08:27 PM Joco[m]: Will see how I go. Thinking a bearing repack in the future might help things a bit. Not keen on springing for whole new screw/dual nut setup.
08:28 PM Lcvette: Similar to the block measuring but with an actual cut
08:28 PM Lcvette: Dual nut is really a big plus
08:28 PM Lcvette: C5 dual nut is even better
08:29 PM Lcvette: But wil run about $500usd for a set
08:29 PM Lcvette: You can get by with uber powered spindle but if your axis are out of whack you'll never make great parts
08:30 PM Lcvette: Put it on the wish list
08:30 PM Lcvette: C5 are ground screws
08:30 PM Lcvette: They have about 0.0001 to 0.0003" backlash
08:31 PM Lcvette: Not even enough to worry with blc
08:31 PM Lcvette: And you set the actual pitch scale is bang on
08:32 PM Joco[m]: ahh simple life. Like it. lol. Wish list
09:07 PM TurBoss: Lcvette: whats up with lathe stuff?
09:08 PM TurBoss: also got no response about the G92 error :/
09:09 PM Not-944e: [02qtpyvcp] 07TurBoss commented on issue 03#42: Wrong preview of rounded rectangle path in probe-basic. - 13https://git.io/JJkAp
09:17 PM Lcvette: TurBoss: lathe needs vtk stuff, but you looked busy on other victories
09:17 PM Lcvette: :D
09:17 PM TurBoss: the 90 90 tools?
09:17 PM Lcvette: The non rotation
09:18 PM TurBoss: hmmm what?
09:18 PM Lcvette: Don't allow rotation
09:18 PM Lcvette: In preview
09:18 PM TurBoss: ah
09:18 PM Lcvette: Backpkot window
09:18 PM TurBoss: ok
09:18 PM TurBoss: got it
09:19 PM Lcvette: * Backplot window
09:19 PM Lcvette: But tonight is last night of company being here
09:20 PM Lcvette: They leave in the morning so we're are in the living room talking
09:20 PM Lcvette: No dev until later maybe
09:20 PM TurBoss: okay no problem I will take a look
09:20 PM TurBoss: have fun
09:31 PM Not-944e: [02qtpyvcp] 07TurBoss pushed 031 commit to 03VTK_LATHE [+0/-0/±1] 13https://git.io/JJkxc
09:31 PM Not-944e: [02qtpyvcp] 07TurBoss 03411ddc7 - disable rotation for lathe mode in vtk backplot
09:31 PM Not-944e: [02qtpyvcp] 07TurBoss created branch 03VTK_LATHE - 13https://git.io/fhCou
09:33 PM travis-ci: kcjengr/qtpyvcp#1644 (VTK_LATHE - 411ddc7 : TurBoss): The build passed.
09:33 PM travis-ci: Diff view : https://github.com/kcjengr/qtpyvcp/commit/411ddc74a366
09:33 PM travis-ci: Build details : https://travis-ci.org/kcjengr/qtpyvcp/builds/705040966
10:01 PM Lcvette: TurBoss: you did it?
10:02 PM Lcvette: :o
10:04 PM TurBoss: yu
10:04 PM TurBoss: <TurBoss "yu"> yup
10:09 PM Lcvette: \o/
10:09 PM Lcvette: Hurray!
10:09 PM Lcvette: I will test in a little bit
10:09 PM Lcvette: Almost time to make a new release i think
10:13 PM TurBoss: great
10:13 PM TurBoss: but requires QtPyVCP release
10:13 PM TurBoss: and duno some feature status
10:19 PM Joco[m]: sounds like some good lathe progress.
10:21 PM Joco[m]: Lcvette: tweaked the X axis scale for 0.02 excess travel and new cut is: taregets: 71, 52. Measured 71.01 and 51.99
10:22 PM Lcvette: Sounds good to me!
10:22 PM Joco[m]: hard to argue with that on this wee machine. Given the class of gear on it
10:22 PM Lcvette: Now test an xy combined cut
10:22 PM Lcvette: Top
10:22 PM Joco[m]: that was from an x/y combo cut
10:22 PM Lcvette: * Yup
10:23 PM Joco[m]: doing a complete circuit around the rectangle
10:23 PM Joco[m]: cuts on Y are within 0.03 with no scale adjustment
10:23 PM Lcvette: Picture of the past
10:23 PM Lcvette: * Picture of the part
10:24 PM * Joco[m] uploaded an image: ima_eea3ed1.jpeg (180KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/NMkfBiatOYGDTNMpaAvSteCi >
10:24 PM Lcvette: Do inside yet?
10:24 PM Joco[m]: yup in same op
10:25 PM Lcvette: But only 2 listed dimensions?
10:25 PM Lcvette: Need x y external and internal
10:27 PM * Joco[m] uploaded an image: ima_4d7c014.jpeg (69KB) < https://matrix.org/_matrix/media/r0/download/matrix.org/GcECkQReyIXhyRFurnfVHSsH >
10:27 PM Joco[m]: here are measurements. Target on left, actual on right. all done in the same program.
10:27 PM Joco[m]: I confused things in last messages. Ignore that.
10:28 PM Joco[m]: the 36 is the Y dimension in the pocket
10:28 PM Lcvette: That's great results
10:28 PM Lcvette: Was with blc on?
10:28 PM Joco[m]: yup from our last lot of work.
10:28 PM Lcvette: You should be pleased
10:29 PM Joco[m]: only new thing is I tweaked the X scale value to account for 0.02mm excess travel
10:29 PM Lcvette: Excellent so it would seem this is your proper settings for tolerance parts
10:29 PM Joco[m]: X scale set as: 399.611488830303871
10:29 PM Joco[m]: lol
10:29 PM Lcvette: Don't think anyone could complain about the precision there!
10:30 PM Joco[m]: yup - for this grade cnc its absolutely fine.
10:30 PM TurBoss: you did it Joco ?
10:30 PM Joco[m]: TurBoss: yup - with a lot of patient help/advice from Lcvette
10:30 PM TurBoss: Great!!!
10:31 PM Lcvette: It looks like he did it! \o/ hurray!
10:31 PM * Joco[m] crosses fingers and toes
10:31 PM Lcvette: People scoff when i tell them it's possible to get beautiful parts to a good tolerance from those machine
10:32 PM Lcvette: I usually give up because they settle for less than it's capable of
10:32 PM Joco[m]: if I had better grade ballscrews or double nuts it would be easier. Next will be improved bearings and getting ride of the vibration inducing gears
10:32 PM Lcvette: You did good!
10:32 PM Lcvette: Yes
10:33 PM Lcvette: If you get that grinding setup, you can also regrind your spindle
10:33 PM Lcvette: Get it dead nuts zero runout
10:34 PM * Joco[m] rubs hands together
10:34 PM Lcvette: Then with belt drive, quality tooling, flood coolant, and some good cam it will make mirrors
10:34 PM TurBoss: 🤞
10:35 PM Joco[m]: next new project (not in flight on) is full enclosure.
10:35 PM Joco[m]: Going to start modeling that out in CAD
10:35 PM Lcvette: You have nachi tapered rollers?
10:35 PM TurBoss: freecad is good for me
10:35 PM TurBoss: :P
10:35 PM Joco[m]: I'm a fusion bunny
10:35 PM Joco[m]: had used freecad. But find fusion easier
10:36 PM TurBoss: 🐰
10:36 PM Joco[m]: lol
10:36 PM TurBoss: ya.. true frecad sux n some ways but is what i do
10:36 PM TurBoss: :P
10:36 PM Joco[m]: right. I think we call this tuning done. And I get on to making two more motor mount standoffs for my mate
10:37 PM Lcvette: Enjoy!
10:38 PM Joco[m]: cheers. Then this evening try and get my brain engaged enough to get back to testing the gcode editior. Lcvette your mate still looking for a notifcation system based on some event trigger?
10:38 PM Joco[m]: Was pondering starting a basic plugin for that which could be extended and made smarter in time.
10:38 PM Lcvette: Yeah that would be sweet!
10:39 PM Joco[m]: idea being to set in yaml the trigger/action pairs.
10:39 PM TurBoss: on blendercam once asked for a email notification sistem
10:39 PM Lcvette: Awesome!
10:39 PM TurBoss: instead I ended speeding up the process 10x
10:39 PM Joco[m]: I want to be able to make it so a trigger can have more than one action. e.g. play a sound AND set a HAL pin high
10:40 PM Lcvette: Awesome
10:40 PM TurBoss: a plugin
10:40 PM Joco[m]: TurBoss: wow - impressive and a better option :)
10:40 PM Lcvette: I wasn't too grey a sexy voice model to make some notifications
10:40 PM Lcvette: * I want to get a sexy voice model to make some notifications
10:41 PM Lcvette: Hey big boy, put your tool in..lol
10:41 PM TurBoss: https://github.com/kcjengr/qtpyvcp/blob/master/qtpyvcp/plugins/clock.py
10:42 PM TurBoss: https://github.com/kcjengr/qtpyvcp/blob/master/qtpyvcp/yaml_lib/default_config.yml#L14
10:42 PM TurBoss: Joco: ^^ plugins
10:43 PM Joco[m]: Cheers. Will look at. I have done one plugin already. The one for exposing certain HAL pins.
10:44 PM Joco[m]: But need to learn the yaml parsing side.
10:45 PM TurBoss: https://github.com/TurBoss/JauriaCNC/blob/master/tnc/config.yml#L17
10:45 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/CgdGYoQdIaEgRKDVYxtknesV >
10:46 PM TurBoss: provider qtpyvcp.plugins.filename:Class
10:46 PM TurBoss: then you can add a list of args
10:46 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/mwJOzRkSTyEaTgCRfMICFALr >
10:46 PM TurBoss: kwargs:
10:46 PM TurBoss: menu: ( default_menubar )
10:46 PM TurBoss: ui_file: {{ file.dir }}/ui/window.ui
10:47 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/lRaZXDqJpNwXWolPjwSpIxnS >
10:48 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/mcRojMdRNVxdPlOnmpDJziXO >
10:48 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/ZdUYqjYewGjpzuujFPYCCUTH >
10:48 PM TurBoss: Joco: something like this^^
10:49 PM Joco[m]: Sweet
10:49 PM TurBoss: ui_file can be snd_file
10:49 PM TurBoss: just forgot to rename
10:52 PM * TurBoss sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/zYuQPCGKMbWiYCMqPhLNwklb >