Back
[01:06:20] <tjb1> Anyone here good with RAM?
[01:15:07] <s1dev> where can I get a rack and pinion of ~4ft in length?
[01:18:32] <tjb1> moore gear
[01:18:42] <tjb1> http://www.mooregear.com/
[01:18:54] <s1dev> tahnks
[01:30:20] <tjb1> It was $239 for 26' of 20 pitch - 20 pa rack
[01:30:24] <tjb1> shipped
[01:59:27] <RyanS> Dear santa
https://www.youtube.com/watch?v=-s0_lUoA-FE
[02:00:30] <tjb1> Id be happy with this -
http://www.kickstarter.com/projects/andyrawson/ir-blue-thermal-imaging-smartphone-accessory?ref=card
[02:01:23] <s1dev> nice
[02:02:19] <RyanS> the narration is hilarious
[02:04:57] <tjb1> the resolution/speed isnt enough to make me want to buy it though
[02:06:23] <tjb1> Sounds like chinese with uk accent
[02:08:09] <RyanS> How much you reckon something like that is worth?
[02:08:25] <RyanS> 200k + ?
[02:08:35] <RyanS> Probably more
[02:08:41] <tjb1> More than I will ever see in my life
[02:09:03] <DJ9DJ> moin
[02:09:42] <RyanS> is it mainly designed for cylindrical parts?
[02:10:34] <RyanS> I'm not sure how it would clamp square/rectangular parts although you can see it tapping a square block
[02:11:22] <tjb1> RyanS:
http://25.media.tumblr.com/tumblr_mcb7aqOqeA1rhbgojo2_400.jpg
[02:11:25] <tjb1> Thats how.
[02:12:04] <RyanS> right..
[02:13:19] <tjb1> :)
[09:53:12] <mikegg> what's the syntax for loading an ini file at the command line? halrun -i arduino.ini arduino.hal ...?
[09:57:53] <mikegg> i think it's halrun -i arduino.ini -f arduino.hal
[10:04:13] <mikegg> argg. halui pins still don't exist
[10:32:00] <JT-Shop> what are you trying to do?
[10:59:08] <mikegg> i think I got it
[10:59:28] <mikegg> trying to use an arduino to interface with a dual FDM extrusion head
[10:59:39] <mikegg> steppers, heating elements, thermistors etc
[11:00:50] <mikegg> I got the pyvcp to load
[11:01:05] <mikegg> now I need to hookup all the signals and see what I got...
[11:07:11] <mikegg> that brings up a good point - do ya'll know of any real-time analog in methods?
[11:14:58] <pcw_home> what bandwidth?
[11:16:38] <mikegg> uhh, nothing fancy I suppose. would something like twice as fast as the servo thread be reasonable? what would that be, 1/50,000 ns
[11:16:57] <archivist> and number of channels
[11:17:03] <mikegg> two
[11:18:02] <pcw_home> 20 KHz sounds like base thread to me
[11:18:59] <pcw_home> base thread is more like 1 KHz to maybe 8 KHz (on a really fast machine)
[11:19:17] <pcw_home> servo thread I should say
[11:19:30] <mikegg> don't have my machine handy, and I can't remember off the top of my head. I don't think i changed them from the stock hm2 configs...
[11:20:04] <pcw_home> If you have HM2 and servo thread is fast enough , a V-F converter is an easy way
[11:21:00] <mikegg> oh, and then just use an encoder counter or something?
[11:21:46] <pcw_home> Yeah you use the encoder velocity signal
[11:22:27] <pcw_home> http://www.analog.com/en/analog-to-digital-converters/voltage-to-frequency-converters/ad7740/products/product.html
[11:22:43] <pcw_home> for example
[11:24:41] <IchGuckLive> hi all B)
[11:25:10] <pcw_home> Or figure out a real time interface to an Arduino or something
[11:25:23] <mikegg> hah
[11:25:43] <andypugh> You probably don't need real-time for temperature control
[11:26:35] <mikegg> yeah, I know - but I think it might be easier.. cobble it all together with PID and be done
[11:26:36] <pcw_home> Yeah for temperature control a serial/USB interface might be OK
[11:26:52] <IchGuckLive> will the servo prices move down as there are more and more hobbyists going over
[11:28:00] <pcw_home> encoders and wiring will make servos more expensive the step motors
[11:28:14] <pcw_home> s/the/than/
[11:28:52] <pcw_home> although the drive power stages are somewhat simpler for servos
[11:29:02] <IchGuckLive> i sugest around 250euros woudt be atchivable but on 600Eur now its not on me
[11:29:18] <pcw_home> (3 1/2 bridges instead of 4)
[11:30:49] <IchGuckLive> i give up on the servo i got
[11:31:14] <_abc__> Did any of you guys try home carburuzation of broken mill bits to allow diamond resharpening?
[11:31:18] <IchGuckLive> no move within 5month 4weeks intesiv trying around 150 houres of wor k
[11:32:17] <IchGuckLive> _abc__: the reasharping of a good bit is on manafactur cheep but a new russion bit is cheeper
[11:32:49] <_abc__> *carburization. Ie stick in oven with carbon packing and let sit at 950C for a few hours.
[11:34:09] <_abc__> I shape my own bits using a diamond wheel sharpener with 3 axes. I use broken mills as raw material. I would like to make the bits I shape harder
[11:34:41] <_abc__> The bits i make are used to mill delicate things
[11:34:43] <IchGuckLive> http://www.wolf-beschichtung.de/
[11:35:11] <_abc__> Thanks
[11:36:13] <IchGuckLive> i use DLC3000 for punches this is 30times more effectiv
[11:36:37] <_abc__> Anyway I make bits with 0.1 to 0.5mm tips. I think I can make my metallurgy fit on top of an unmodified BIC lighter....
[11:37:08] <IchGuckLive> V -shape
[11:37:22] <_abc__> So large solutions will not help so much. I need very small.
[11:38:02] <IchGuckLive> 0.5mm mill bit is 3.90 so no sharping
[11:38:27] <_abc__> 3.90 ?
[11:38:32] <IchGuckLive> Euros
[11:38:51] <_abc__> Lol the ones I use are 45€
[11:39:34] <_abc__> Also they do not have the right shape for what I do.
[11:41:15] <andypugh> It probably makes more sense to spend €3.90 on the wrong cutter than €45.
[11:42:09] <archivist> if the old endmill was through hardened why bother carburizing, just dont over heat when grinding
[11:43:14] <_abc__> Heh the 45€ ones are used as is for one thing and I take the broken ones and do my thing with them after that.
[11:43:36] <IchGuckLive> _abc__: shape photo please
[11:43:43] <_abc__> Archivist the tools are surface hardened only.
[11:44:31] <TekniQue> most hss drills you get nowadays only seem to be hardened at the tip
[11:44:47] <TekniQue> impossible to resharpen
[11:44:52] <_abc__> Shape i need is mostly V with angle adapted to the job, sometimes with cut off tip. Trapeze groove kind of.
[11:45:06] <TekniQue> once the tip is gone, there's no way to grind them sharp
[11:45:32] <andypugh> TekniQue: Buy better drills?
[11:45:43] <IchGuckLive> _abc__: on you bit cost it is worth to get a dimond V-bit grinder
[11:45:45] <_abc__> Tek exactly. They have the gas nitrurizing / carburizing tuned for just surface hardening.
[11:46:45] <_abc__> Tek and you can resharpen them if you carburize them. This time, not surface only...
[11:47:02] <_abc__> Thus my questioning here.
[11:47:27] <IchGuckLive> http://stores.ebay.de/Werbematerialien-und-Gravierbedarf/Gravierstichel-/_i.html?_fsub=1129709018&_sid=129145738&_trksid=p4634.c0.m322
[11:47:49] <IchGuckLive> this are the ones you got
[11:49:00] <_abc__> So ill take a hollow bolt, pack it with carbon and stick a mill bit into it, pack a little high heat plaster around the open and and heat it on a gas flame, with a K thermocouple also wedged in there. Has anyone tried this?
[11:49:41] <JT-Shop> you may be the first
[11:49:55] <_abc__> Meh I have a doubt.
[11:51:10] <_abc__> Any idea how hot i can get something not insulated on an open flame? Red heat for sure, but I need white?
[11:51:55] <IchGuckLive> if you need angels of tool grinding here is a pdf that holds all tools
http://www.lehrer.uni-karlsruhe.de/~za685/cnc/cnctools.htm
[11:52:01] <IchGuckLive> tool 30 is your bit
[11:52:09] <andypugh> Oxy-acetylene welding rather relies on the fact that a flame can melt steel.
[11:52:37] <_abc__> I mean with gas in air :)
[11:52:47] <_abc__> Thanks ichgucklive
[11:53:30] <_abc__> And with somewhat reasonable gas consumption
[11:54:28] <_abc__> Radiant heat loss at 1000C is high, it makes sense to shield the part or insulate it as much as possible
[11:54:33] <andypugh> I have hardened things with a MAPP gas blowtorch. How big is the whole assembly?
[11:54:38] <archivist> start from known metals
[11:55:14] <archivist> random old end mills could have any composition
[11:55:14] <_abc__> Andy probably 10mm hollow bolt, 20mm long or a little longer.
[11:55:27] <andypugh> A firebrick tent is probably called for.
[11:56:02] <_abc__> Archivist I agree, but I have faith in what the original mill bit maker choose.
[11:56:34] <archivist> was it hss,carbon steel or coated or wtf
[11:56:38] <andypugh> Don't spend too much time and effort, I anticipate failure.
[11:56:41] <_abc__> Andy yes, I was thinking of embedding the item halfway in the brick
[11:57:16] <_abc__> It was Tungsten Carbide I think but nobody knows for sure.
[11:58:01] <_abc__> I plan to coat the mill bit end with high heat clay to minimize air contact
[11:58:38] <archivist> I expect total failure with tungsten carbide
[11:58:46] <_abc__> Why?
[11:59:12] <_abc__> Its called that but it is not an insert
[11:59:21] <archivist> they are made from solid you should have no need to harden in any way
[11:59:24] <_abc__> It is all metal. The bit.
[11:59:54] <_abc__> So its just the usual marketing buzzword thing
[12:00:31] <_abc__> I expect good old hss high alloy with carburized outside
[12:00:35] <andypugh> Not solid carbide?
[12:00:46] <archivist> what colour are they, bright or grey
[12:01:10] <_abc__> Does not look like it. My diamond sharpener eats into it too easily
[12:01:30] <_abc__> Grey outside brighter inside
[12:03:38] <archivist> I would say hardening and tempering is a black art where you really need to know your starting point and metallurgy
[12:04:03] <_abc__> I had some older ones which were deep gray throughout. Those were sharpenable and kept for quite a while after that. I suspect solid carbide.
[12:04:34] <andypugh> http://www.cromwell.co.uk/publication_page_pdfs/1187/143.pdf
[12:04:41] <_abc__> Archivist i agree. But for the price of a bolt, i might find out.
[12:04:53] <andypugh> Are some coated HSS cutters, which might be what you have.
[12:05:07] <_abc__> I cant see pdf right now, am on a phone
[12:05:23] <andypugh> I don't think that you harden HSS by carburising. All the required elements are already in there.
[12:05:51] <_abc__> So why the different surface color?
[12:06:07] <andypugh> It's a coating.
[12:06:12] <andypugh> Mainly anti-welding
[12:06:21] <_abc__> Hmm
[12:06:43] <archivist> some coatings are for wear resistance
[12:06:48] <_abc__> And why is the sharpened bit not as tough?
[12:07:27] <archivist> no coating choose which resistance you no longer have
[12:08:04] <andypugh> Toughness is normally inversely proprtional to hardness.
[12:08:14] <IchGuckLive> _abc__:
http://www.kirba.de/multilang/faq/faq-sch.htm
[12:08:21] <_abc__> Hmm so simply attempting to temper them as is might do the trick?
[12:08:40] <archivist> remember if you try to retreat a part the depth will vary and it will likely become brittle
[12:09:02] <_abc__> I will look at the links later when near a normal computer, thanks.
[12:09:07] <andypugh> Is it toughness you want, though?
[12:09:58] <_abc__> Tough, hard. I use the cutter in the same way as when new, it keeps much less.
[12:10:16] <_abc__> Loses edge way too fast
[12:10:42] <L84Supper> is cutting softer materials an option? :)
[12:11:02] <_abc__> Visibly too. I can see the bluntness after a few meters on a microscope. Initially it cuts great.
[12:11:09] <archivist> better cooling and lubrication
[12:11:36] <_abc__> I cut fiberglass copper bronze and aluminum with this
[12:11:39] <archivist> or in some cases move to no cooling
[12:12:11] <_abc__> The tests are usually in fr4 with copper
[12:12:41] <archivist> carbides can/will crack and fail if the coolant is intermittent
[12:12:48] <andypugh> Toughness is the resistance to fracture. Hardness is the resistance to deformation. Unless you are breaking the bits then toughness isn't what you want.
[12:13:21] <_abc__> Dry cutting, 30krpm, 1 to 3 mm/sec speed at say 0.2mm depth of groove
[12:13:44] <_abc__> This is fr4
[12:14:11] <archivist> there is a limited life cutting glass
[12:14:27] <archivist> 30k may be a bit too slow
[12:14:29] <_abc__> Andy i am not sure if the edge chips off first and is then rounded or if it is rounded
[12:15:26] <_abc__> The bad things happen after 2-3 linear meters. The same bit in the same machine same conditions lasts weeks
[12:16:04] <_abc__> 30krpm is the upper limit of this 2kW water cooled spindle's vfd
[12:16:17] <_abc__> Brb
[12:17:41] <_abc__> Re
[12:20:38] <_abc__> At 1mm/sec 30krpm and 0.1mm tip I get about 160:1 milling speed vs feed. This should be fine enough for what i do
[12:21:39] <_abc__> Also chip shape is the same as with original mill, so no shape problems. Just weird material issues.
[12:47:43] <awallin> this just advertized on the finnish forum..
http://www.atomia.net/shop/index.php?route=common/home
[13:07:56] <IchGuckLive> awallin:
http://shop.embedded-projects.net/index.php?module=artikel&action=gruppe&id=11
[13:10:00] <IchGuckLive> im off have a nice sunday
[13:29:27] <DJ9DJ> namd
[13:37:40] <Aero-Tec> had my network go down for a bit, so not sure if anyone tried to contact me
[13:38:27] <Aero-Tec> new update, the new setup is getting close to 75% faster then it was before I started to tweak it
[13:39:06] <andypugh> You could, of course, have been finished by now with the old setup :-)
[13:39:09] <Aero-Tec> I had said 50%, but 75% is closer
[13:39:17] <Aero-Tec> lol
[13:39:26] <Aero-Tec> what fun would that have been
[13:39:41] <Aero-Tec> this is a reoccurring order
[13:39:43] <andypugh> And next time you can charge less.
[13:39:56] <Aero-Tec> every 2 to 3 months I have to do it
[13:40:51] <Aero-Tec> I have learned so much by going through the effort of getting this going like I did
[13:41:13] <Aero-Tec> no chance of charging less
[13:41:28] <Aero-Tec> not charging enough from the start
[13:41:33] <Aero-Tec> filler work
[13:41:49] <archivist> I had a choice a few weeks ago, stop and sharpen cutter and re setup or struggle for just a few more gears, I struggled
[13:42:19] <Aero-Tec> there are some times when it gets to be a hard choice
[13:42:45] <Aero-Tec> trudge on or stop and recode or re tool or re what erver
[13:43:14] <Aero-Tec> if it was not a ongoing job I would not have put in the effort like I did
[13:43:37] <archivist> I think your real change would be to move to a servo system
[13:43:43] <Aero-Tec> there need to be a pay off some where in the job
[13:44:25] <Aero-Tec> I got a return in 2 ways
[13:45:00] <Aero-Tec> faster out put, and a real education on the inner working of the program and setting it up
[13:45:31] <andypugh> I think we all learned something. I am not sure cradek wanted to learn what he learned.
[13:45:48] <Aero-Tec> also got more familiar with the Gcode of EMC
[13:46:04] <Aero-Tec> what was that?
[13:46:30] <Aero-Tec> that X and B could not do a G64 move?
[13:47:13] <andypugh> I am unclear of the details, but I think he became disatisfied with the way blending works when axes have very different accelerations.
[13:48:04] <Aero-Tec> could it not just be tweaked?
[13:48:47] <archivist> tweaking maths is a bit hairy
[13:48:48] <Aero-Tec> so it was the difference in settings, as in steps and accel that made it not work?
[13:49:11] <Aero-Tec> I thought it was the X and B axis were not set up yet for blending
[13:50:12] <Aero-Tec> I have X and Z set up exactly how X and B were set up
[13:50:51] <Aero-Tec> so it has nothing to do with the difference in acceleration or steps or anything like that
[13:50:55] <andypugh> I am wary of putting words into someone else's mouth, or thoughts into their heads, but I think he found an issue when looking at the deeper details.
[13:51:40] <Aero-Tec> it is purely that X,B are not set up to do it and X,Z is
[13:51:41] <andypugh> I wonder if you could wind springs with a G33?
[13:52:26] <Aero-Tec> could if the spindle was turning and had a encoder on it and a good speed control
[13:52:45] <Aero-Tec> right now the spindle in just a tool holder
[13:52:58] <Aero-Tec> the wire is feed through it
[13:53:19] <skunkworks> andypugh: that is how I understood it also
[13:53:21] <andypugh> I did find myself pondering today if you could allow IJK words in a G1 block, to tell the system what the working radius was. (as you were discussing)
[13:54:01] <Aero-Tec> I set up Z and angular and wrapped, every thing was exactly the same from B and Z
[13:54:17] <skunkworks> I think mach as a place to give a 'radius' of the rotory for calculating feeds..
[13:54:56] <skunkworks> if I understand it correctly
[13:55:11] <andypugh> Aero-Tec: The fact that _your_ system now works doesn't mean that there isn't an underlying problem that can be seen when you look carefully enough.
[13:55:26] <Aero-Tec> true
[13:56:18] <Aero-Tec> one thing, the program complains about gcode Z axis being out side the limits of Z
[13:56:38] <Aero-Tec> but if you say continue anyways it works
[13:56:51] <Aero-Tec> the B axis did not do that
[13:57:22] <Aero-Tec> but then again the B axis did not work for blended moves
[13:58:25] <Aero-Tec> also, even tho Z is set up to be angular and wrapped, it appears if is not
[13:58:33] <Aero-Tec> it is not wrapped for sure
[14:04:56] <andypugh> I just looked at the code, and it only even looks for the WRAPPED_ROTARY parameter in the [AXIS_3], [AXIS_4] and [AXIS_5] block. Which seems a bit short-sighted.
[14:06:54] <andypugh> You can only have an indexer on A, B and C too.
[14:07:36] <JT-Shop> hmmm I'm getting a gouge error but the code runs fine :(
[14:07:39] <Aero-Tec> 10.7.2 Linear Motion at Feed Rate - G1
[14:07:41] <Aero-Tec> (a) For linear motion at feed rate (for cutting or not), program G1 X~ Y~ Z~ A~ B~
[14:07:42] <Aero-Tec> C~, where all the axis words are optional, except that at least one must be used. The G1 is
[14:07:44] <Aero-Tec> optional if the current motion mode is G1. This will produce co-ordinated linear motion to
[14:07:46] <Aero-Tec> the destination point at the current feed rate (or slower if the machine will not go that fast).
[14:07:47] <Aero-Tec> (b) If G16 has been executed to set a polar origin then linear motion at feed rate to a point
[14:07:49] <Aero-Tec> described by a radius and angle G0 X~ Y~ can be used. X~ is the radius of the line from
[14:07:51] <Aero-Tec> the G16 polar origin and Y~ is the angle in degrees measured with increasing values
[14:07:52] <Aero-Tec> counterclockwise from the 3 o'clock direction (i.e. the conventional four quadrant
[14:07:54] <Aero-Tec> conventions).
[14:07:56] <Aero-Tec> Coordinates of the current point at the time of executing the G16 are the polar origin.
[14:07:57] <Aero-Tec> It is an error if:
[14:07:59] <Aero-Tec> ¨ all axis words are omitted.
[14:08:01] <Aero-Tec> If cutter radius compensation is active, the motion will differ from the above; see Cutter
[14:08:02] <Aero-Tec> Compensation. If G53 is programmed on the same line, the motion will also differ; see
[14:08:04] <Aero-Tec> Absolute Coordinates.
[14:08:06] <Aero-Tec> from the mach manual
[14:08:23] <Aero-Tec> it is the same for G0
[14:08:42] <Aero-Tec> not sure if that helps
[14:09:01] <Aero-Tec> never used G16 in mach
[14:09:12] <Aero-Tec> I never
[14:09:47] <andypugh> That's not the same thing
[14:10:11] <Aero-Tec> JT-Shop: what code are you testing?
[14:10:19] <andypugh> You can use polar coordinates in LinuxCNC, but I am not sure anyone does.
[14:10:39] <JT-Shop> it's just a radius that errors out if the Z is not a positive value
[14:11:05] <Aero-Tec> I never heard of a radius option in mach with G1 or G0
[14:11:44] <andypugh> http://linuxcnc.org/docs/devel/html/gcode/overview_pl.html#_polar_coordinates
[14:11:58] <JT-Shop> http://pastebin.com/mtYr7Vie
[14:12:01] <Aero-Tec> in Lathe you can set radius or diameter mode
[14:12:35] <Aero-Tec> not sure if one can change it by Gcode in mach
[16:15:53] <DJ9DJ> gute nacht
[16:43:28] <petitbrochet> il ya quelqu'un
[17:01:13] <skunkworks> http://www.youtube.com/watch?v=GK4rbD6C17o
[17:06:59] <andypugh> That actually looks rather handy/useful
[17:08:52] <andypugh> It doesn't necessarily do anything that you can't do with ngcgui, but perhaps a little more turnkey
[17:10:07] <skunkworks> right
[18:08:13] <JT-Shop> takes to long to watch
[18:10:09] <Tom_itx> it is kinda slow ehh?
[18:13:51] <JT-Shop> yea
[18:14:24] <tjb1-mobile> JT-Shop: Did you see that thc I sent you?
[18:14:36] <JT-Shop> no
[18:17:08] <skunkworks> http://linuxcnc.org/index.php/english/forum/21-axis/25947-axis-tool-path-colouration
[18:17:52] <skunkworks> I think he is talking about how the tool path history disapears
[18:18:50] <skunkworks> only a finite amount of history is displayed because it slows the display/computer
[18:21:12] <cradek> yeah, long ago, it used to display forever until you cleared it, and that caused trouble.
[19:51:15] <cradek> someone had to go and mention the metric system on emc-users
[20:01:26] <Nick001> <cradek>I put that -showkeys into terminal and got command unknown - am I missing a package?
[20:03:02] <cradek> sorry, I don't know what you're talking about
[20:04:03] <Nick001> you gave me this yesterday to check my keyboard
[20:04:29] <cradek> maybe it was some other cradek...?
[20:07:16] <Nick001> sorry - it was <cpresser>
[20:15:57] <pcw_home> do you mean the showkey command?
[20:20:12] <Nick001> yes
[20:20:52] <pcw_home> thats just showkey
[20:21:09] <Nick001> try to find out if my keyboard is bad of I have a prob in linuxcnc
[20:21:13] <pcw_home> maybe showkey -a if you want ascii
[20:21:44] <Nick001> its a terminal command?
[20:21:49] <pcw_home> yes
[20:22:07] <pcw_home> > showkey -a
[20:22:22] <Nick001> will it also show F5 key?
[20:22:57] <pcw_home> shows it as a ANSII sequence
[20:23:17] <pcw_home> if you want keycodes you need -s or -k
[20:23:41] <pcw_home> (but I think that needs sudo showkey)
[20:24:51] <Nick001> ok - I'll try it in the morning when I get back to the shop
[20:26:08] <pcw_home> pcw@pcw-G41M-Combo:~$ showkey -a
[20:26:11] <pcw_home> Press any keys - Ctrl-D will terminate this program
[20:26:12] <pcw_home> 5 53 0065 0x35
[20:26:14] <pcw_home> 5 53 0065 0x35
[20:26:15] <pcw_home> 6 54 0066 0x36
[20:26:17] <pcw_home> 7 55 0067 0x37
[20:27:39] <Nick001> what should Control K show?
[20:28:12] <pcw_home> ^K 11 0013 0x0b
[20:29:42] <pcw_home> or scancodes:
[20:29:44] <pcw_home> keycode 29 press
[20:29:46] <pcw_home> keycode 37 press
[20:29:47] <pcw_home> keycode 37 release
[20:29:49] <pcw_home> keycode 29 release
[20:29:59] <Nick001> thanks - I guess I should dual boot a laptop so I can do some of this at home -)
[20:30:07] <pcw_home> (29 is left control)
[20:30:21] <Nick001> is this documented anywhere?
[20:31:26] <pcw_home> man showkey
[20:32:28] <Nick001> thanks - will persue this in the morning
