#linuxcnc | Logs for 2014-12-04

Back
[00:09:17] <Tecan> http://editorial.designtaxi.com/news-pastamodels3108/1.jpg
[00:13:19] <XXCoder> http://www.johndearmond.com/2014/11/01/using-the-induction-heater-aluminum-casting/ interesting
[00:36:51] <Valen> anybody know where I could get fineish module helical cut gears?
[00:36:59] <Valen> like module .8 or something, with about a 40mm OD
[00:37:02] <Valen> metric
[01:32:53] <archivist> Valen, http://hpcgears.com/n/products/13.helical_gears/helical_gears.php
[01:35:16] <Valen> archivist I've never heard of parallel and crossed gears?
[01:36:20] <Valen> also that is exactly what I was looking for
[01:40:16] <archivist> crossed make a right hand 90 deg drive
[01:40:48] <archivist> parallel are parallel axes of the two gears
[01:44:14] <Valen> ahh
[02:07:55] <ssi> https://pbs.twimg.com/media/B3_t57yIIAEcc_x.jpg:large
[02:08:04] <ssi> they're ready to harden
[02:08:10] <ssi> I don't think this is gonna be my final set though
[02:08:17] <Connor> ssi what is that for ?
[02:08:20] <ssi> I dorked one of the dimensions; it's right at 1" if not a little under already
[02:08:27] <ssi> Connor: they're 1/2-1-2 blocks
[02:08:38] <ssi> it's just a heat-treat / precision grinding project
[02:08:44] <ssi> for a learning experience
[02:08:49] <Connor> ah. okay
[02:09:38] <ssi> 123 blocks are too big, the metal costs too much :)
[02:09:42] <ssi> plus I have a couple sets
[02:11:38] <archivist> do the right treatment and they grow
[02:12:03] <ssi> oh yeah? I thought they shrunk
[02:12:07] <ssi> I haven't found and solid data on this yet
[02:12:19] <ssi> so a large part of this first project is recording the dimensions and seeing how they change after treatment
[02:12:35] <archivist> iirc nitriding grows
[02:13:18] <archivist> ok that is a process for cast cranks but they may be other methods
[02:15:32] <Deejay> moin
[02:34:00] <renesis> guys remember when i said fusion360 wont let me select segments?
[02:34:05] <renesis> i think im wrong, testing
[02:41:48] <renesis> oh jezus fuck
[02:42:54] <renesis> you left click on the countour and it pops up open and close contour selection, along with accept delete cancel buttons
[02:43:10] <renesis> does not happen when you right click
[05:27:01] * Loetmichel was just called "antje" by a co-worker (german TV walruss https://www.youtube.com/watch?v=VZju7NZpe7U , american expression would be calling me jamie hyneman) ... it seems i should cut my beard a bit ;-)
[05:36:09] <archivist_herron> I see no reason to cut beards :)
[07:31:18] <seltecc> Hi, yesterday I´ve tried to get 5v out of my relay output from 5axis breakout board. That was a wrong approach so I rewired my external relay to a external powersupply and want to use the onboard relay to open and close the circuit of the external relay. When I invert the pin 9 thenthe external relay switches on indicating that is the right pin etc.
[07:33:57] <seltecc> I can not get the onboard relay to open and close automatically ...The halmeter is indicating spindle-cw -False ?!
[07:38:52] <seltecc> I´ve tried cw,ccw,spindel-on,pwm,dout 00 to 03 non of them works...
[07:40:23] <Tom_itx> if the relay has a diode across it (which it should) maybe you wired it backwards
[07:41:18] <jdh> wiring problem, or hal-wiring problem?
[07:41:18] <seltecc> inverting pin 9 closes the external circuit....
[07:42:02] <seltecc> yes hal problem maybe.....
[07:43:02] <seltecc> aha ok got an idea....
[07:48:09] <seltecc> wow beautiful...now at least the external circuit goes on automatic .....
[07:54:00] <_methods> damn stuck valves
[07:54:08] <_methods> launch that rocket!
[07:55:41] <seltecc> can I tweak the time of the spindel -on ?
[07:59:11] <seltecc> so the external signal switches on with the cw signal on pin 9 and leaves on till the job is finished. is there a chance of switching off when ever z axis moves?
[08:01:33] <seltecc> I want to run a laser...
[08:05:51] <Tecan> just remember to cover your last good eye
[08:06:15] <_methods> hehe
[08:06:19] <Loetmichel> archivist: maybe it would work better to find my mouth with the soppn then?
[08:06:25] <Loetmichel> spoon
[08:07:50] <Loetmichel> http://www.cyrom.org/palbum/main.php?g2_itemId=15326 <- it has grown a bit since then ;-9
[08:34:31] <seltecc> can I relate the spindle-cw to the z axis movement ?
[08:35:23] <archivist> that does not seem sensible to me
[08:36:01] <archivist> you can see what other people have done with their laser systems
[08:36:12] <archivist> have a look at the wiki
[08:41:12] <seltecc> ok
[08:51:44] <seltecc> do you have a link?
[08:52:48] <seltecc> i found this: http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Case_Studies
[08:54:45] <JT-Shop> also look on the forum with google site search
[08:56:27] <seltecc> thx
[09:54:58] <skunkworks> JT-Shop, do you have a writup on setting up lathe tooling?
[10:04:14] <ssi> skunkworks: what are you wanting to know?
[10:04:31] <archivist> buy enough tool holders so you can avoid messing with the tool table all the time
[10:04:37] <ssi> agreed
[10:04:51] <ssi> what I like to do is pick one tool as a datum tool
[10:04:58] <ssi> and its cutting nose is 0,0 in the tool table
[10:05:07] <ssi> everything else is referenced to that point in space
[10:05:29] <archivist> that is what I do but using hss and sharpening it screws that up
[10:05:37] <skunkworks> wthat is kind of what I am wondering.. I will probably have to play with it...
[10:05:39] <ssi> yeah I use all indexables
[10:06:01] <ssi> that's how I did it on my grizzly which has a qctp
[10:06:04] <ssi> on the hardinge I do it differently
[10:06:34] <archivist> they dont make cheap indexables to suit my work
[10:16:30] <ssi> I may try to harden those blocks today
[10:24:05] <ssi> holy crap
[10:24:06] <ssi> http://www.ebay.com/itm/S436FXRLZ-STARRETT-MICROMETER-SET-/121107561312?pt=LH_DefaultDomain_0&hash=item1c3292bb60&autorefresh=true
[10:24:16] <ssi> if that was an order of magnitude cheaper, that'd be awesome to own
[10:25:10] <archivist> waaay over priced
[10:25:29] <ssi> well yeah
[10:25:40] <ssi> but I've never seen a 12-24 set for sale before
[10:25:41] <ssi> pretty cool
[10:27:52] <seltecc> pin9 TRUE if Zaxis moves down and FALSE when going up.... how to implement in HAL ?
[10:28:58] <ssi> use an abs component
[10:29:03] <ssi> hook in to the z velocity pin
[10:29:13] <ssi> use the sign output of the abs component as your output
[10:29:35] <seltecc> in HAL-File?
[10:29:41] <ssi> yeah
[10:30:10] <seltecc> I don´t have the commands for doing this...
[10:30:17] <cradek> what should the output be when there is no motion?
[10:30:31] <seltecc> false
[10:31:27] <ssi> abs will give you false at 0 vel, false when moving positive (which is usually up), and true when moving negative
[10:31:28] <cradek> I think that might be what ssi's solution gives you
[10:31:50] <archivist> I think it is wrong to use a motion as an on off, there are better ways, think of z hitting a limit
[10:31:51] <cradek> yeah
[10:32:34] <cradek> I don't know if this is a good solution to a problem, but I think ssi's answer will give you what you asked for
[10:32:37] <seltecc> right now pin9 is true when the job begins and false when job is finished...
[10:33:20] <seltecc> I want to run a ext. relay with a laser attached to it...
[10:33:24] <archivist> look how others enable their laser
[10:34:22] <ssi> I need to start changing out all my chinese drill chucks for albrechts
[10:35:45] <seltecc> it is done with 2 or 3 lines in the hal... and I have a configuration that is differen from others who use the zdir pin from the parallel port... so no use of looking what others do...
[10:36:29] <cradek> I wouldn't want to depend on what a stepgen does with the dir bit while not moving
[10:38:33] <seltecc> I can program the safty hight to almost nothing of the zaxis.... so it is the easyiest way for me to run the damn LASER
[10:40:06] <seltecc> HAL programming is not an issue here....?
[10:44:26] <ssi> I'd love to have one or two of these but damn the money :(
[10:44:26] <ssi> http://www.ebay.com/itm/ALBRECHT-Wrench-Lock-V-Flange-Drill-Chuck-With-Integral-Shank-1-32-1-2-/231381175826?pt=BI_Tool_Work_Holding&hash=item35df648e12
[10:49:28] <skunkworks> we have a few of those - nice chucsk
[10:49:33] <skunkworks> chucks
[10:49:40] <skunkworks> (older versions)
[10:49:41] <ssi> yeah they're great
[10:49:50] <ssi> having a couple cat40 integrated ones for the vmc would be awesome
[10:49:56] <ssi> but I'm not in a position to drop a grand on drill chucks :P
[10:55:05] <seltecc> Pin7 = Zstep, Pin8 =Zdir pin9 = destination.... is it better to use step or dir signal for abs component and how would the exact command look like....
[10:56:09] <cradek> I'm not sure anyone wants to write exact commands for you. have you read the hal tutorial and tried the exercises? It will pay off well to understand hal syntax well enough to be able to experiment with different solutions to your problems.
[10:56:34] <ssi> you don't want to use the step or dir signals, you want to use the axis velocity pin
[10:56:48] <ssi> or motion perhaps
[10:56:58] <archivist> these are motion
[10:57:06] <cradek> the hal syntax is very very simple. there are about 5 commands and you should understand them
[10:57:15] <archivist> hal signals not parport pins
[10:57:20] <ssi> yeah you definitely need to understand hal to do this sort of hackery
[10:58:16] <ssi> archivist: I want to make some tight tolerance tapered bushings, and I'm not sure if it's feasible with the tools I have
[10:58:41] <ssi> I basically need to make bushings which are reamed to .375" id, and then turned to like .705 nominal with maybe a .002" per inch taper
[10:58:49] <ssi> and the whole thing only needs to be .750" long
[10:58:53] <archivist> ssi what!, you have teh southbend :)
[10:58:55] <ssi> then I want to harden them
[10:59:02] <seltecc> where do I find the velocity pin ......here is the board I´m using......: http://softsolder.com/2013/02/17/anonymous-5-axis-parallel-port-breakout-board-pinout/
[10:59:16] <ssi> I think I can figure out how to set it up to OD grind the taper on my surface grinder
[10:59:23] <ssi> but I'm not sure how to finish ID grind the .375 bore
[10:59:26] <ssi> or if it'll even be necessary
[10:59:43] <ssi> but it needs to fit a 3/8" thompson shaft extremely precisely
[10:59:43] <archivist> seltecc, these are hal signals not what is on that board
[11:01:14] <seltecc> sure but redirection of signals can be done in the hal with if true statements etc. ...
[11:02:10] <archivist> seltecc, velocity is an analogue value not a digital signal on a port
[11:03:58] <archivist> ssi I would turn it at one setting to make sure it was concentric
[11:04:32] <ssi> yeah I got the machining part under control; I'm concerned about what happens after heat treatment
[11:04:58] <ssi> I made some of these in aluminum yesterday and just with a reamer, that ID is a hydraulic fit on the shaft
[11:05:04] <ssi> it pushes oil ahead of it, which is perfect
[11:05:07] <archivist> does it need heat treatment it just tightened
[11:05:24] <archivist> ah
[11:05:36] <ssi> these are for an indicator set for precision alignment of rifle actions in a lathe
[11:05:50] <ssi> I've just made them out of aluminum in the past, but a nice precise hardened tool steel set would last me a lifetime
[11:06:00] <ssi> just trying to figure out how to do it :)
[11:06:37] <archivist> harden then finish grind inner and outer in one setup
[11:07:11] <archivist> holding being the hard part then :)
[11:10:34] <ssi> yeah I'm just not sure how to go about ID grind them
[11:10:49] <jdh_> http://oppositelock.jalopnik.com/weldporn-910228664
[11:10:56] <ssi> I guess I'd need to make a toolpost grinder for that
[11:11:25] <archivist> dremel does not make a good toolpost grinder
[11:11:47] <ssi> dremel doesn't make a good much of anything :P
[11:12:27] <roycroft> dermel make good money
[11:12:31] <roycroft> dremel, rather
[11:12:36] <jdh_> I like the little flapper sandpaper wheels in dremels
[11:12:36] <archivist> it does a few jobs well
[11:12:44] <seltecc> I am going back to dremel...
[11:12:45] <ssi> I wonder if one of those proxxon spindles would work well as a toolpost grinder
[11:12:58] <roycroft> i use a foredom flexshaft tool for things that i used to use dremels for
[11:13:44] <ssi> probably a decent air pencil grinder would work well
[11:14:28] <ssi> archivist: I could cut the thing and leave it unparted, maybe just v-notched but still attached to the shaft
[11:14:53] <ssi> then heat treat it, then chuck it back up by its shaft and dial it in, ID and OD grind in one setup, then cut it off with an abrasive disk and clean up the cut end
[11:15:20] <archivist> I was going to say leave a holding extension on the parts for rework later
[11:15:52] <ssi> I could do all that work in a collet, and that'll be plenty accurate enough because the final collet setup will be the finish ID/OD grind
[11:16:06] <archivist> a lot of my stuff is made with a chucking stub
[11:16:28] <ssi> ugh I have to build tools so that I can build tools so that I can build stuff
[11:16:38] <seltecc> with my spindle the dust is spread all over the place since I don´t have a dust shoe or any other solution ...
[11:16:54] <archivist> to understand recursion you need to understand recursion
[11:16:57] <ssi> :)
[11:17:41] <ssi> ok so now I need to use my fancy southbend lathe to make a nice little cartridge spindle
[11:21:07] <archivist> but then you need to grind the collets for the new spindle /me ducks
[11:22:13] <ssi> :)
[11:23:01] <archivist> or cheat http://www.ebay.co.uk/itm/C10-ER11A-100L-Straight-Shank-Collet-Chuck-Holder-CNC-Milling-Deep-Hole-Working-/271560849987
[11:23:58] <ssi> yea I thought about starting with an ER straight shank
[11:24:03] <ssi> I built a spindle out of one like that before
[11:24:10] <ssi> unfortunately I lost that spindle in the fire :(
[11:24:28] <ssi> only thing about that is I'd prefer to have threads on the back end so I can preload the bearings
[11:24:38] <ssi> otherwise I'd have to preload using a stop collar or something, not ideal
[11:24:44] <archivist> I use one for my homebrew cutter grinder
[11:24:46] <ssi> or maybe a spacer and screw a bolt in the end
[11:24:50] <ssi> if it's threaded in the end
[11:25:18] <ssi> projects spawn projects in the most horrible way!
[11:26:28] <archivist> yup I had to make the cutter grinder in a rush to make a bevel gear
[11:27:05] <pcw_home> At some point you hope for convergence...
[11:27:26] <ssi> I don't think it's ever going to happen :)
[11:27:42] <ssi> this whole machine shop nonsense started because I didn't want to pay someone $50 to waterject the instrument panel for my airplane
[11:29:35] <fenugrec> I hear the only way out is to take up knitting or something
[11:31:27] <ssi> golf
[11:31:28] <ssi> :P
[11:37:19] <Nick_name_> is it a good idea, use a usb hid device as cnc pendant with linuxcnc?
[11:44:45] <archivist> some do that happily I believe
[11:45:42] <archivist> I think there is some info on the wiki
[11:49:52] <jdh_> I have a ShuttleExpress(tm)
[11:50:09] <jdh_> awkward to use, but no issues.
[11:50:26] <jdh_> might be less awkward if I used it more, or labeled the axes
[11:55:54] <ssi> archivist: thought:
[11:55:56] <ssi> http://www.astrotools.com/index.php/air-tools/die-grinders/1-8-pencil-type-die-grinder-56-000rpm.html
[11:56:01] <ssi> I can get one of those locally for $65
[11:56:11] <ssi> looks like it might be a reasonably decent quality tool
[11:56:21] <ssi> might be worth starting there for a toolpost grinder, as it'd be quick to set up
[11:57:04] <archivist> nice and fast
[11:57:42] <ssi> I may grab one today and make a toolholder for it
[11:59:15] <archivist> I held a spindle of similar max rpm at a show, had to touch the collet to feel the rpm
[12:00:40] <seltecc> Yes I did it !!! This simple line brings my desired function : net zdir => parport.0.pin-09-out
[12:00:58] <cradek> well done
[12:02:55] <seltecc> Thx
[12:04:42] <seltecc> now to the next point .... when the Laser is on and I push the F2 key the Laser should go off too, but is still on.....How to fix this?
[12:05:27] <ssi> tie machine enable to a laser enable
[12:05:27] <cradek> the "machine on" state is available on the axis.0.amp-enable-out pin
[12:14:18] <seltecc> I love it. ready to test out my LASER..!!!
[12:15:14] <seltecc> Klick, Klack etc. on and off , beautiful
[12:20:23] <PetefromTn_> Afternoon folks
[12:24:20] <seltecc> bye everyone and thanks
[12:25:11] <jdh_> seltecc: got insurance?
[13:12:25] <skunkworks> so if you guys are touching off z how do you touch off tools that are right facing (like tool 1,4?)
[13:13:18] <PetefromTn_> I can
[13:13:29] <PetefromTn_> can't wait to play with CNC lathe stuff LOL
[13:13:48] <skunkworks> ' is too close to Enter...
[13:15:14] <_abc_> Hi guys. the Debian based live iso, does it have a hdd install option please?
[13:15:33] <skunkworks> _abc_, yes - it is at the boot menu stage
[13:15:41] <_abc_> hm?
[13:15:53] <_abc_> Oh it asks whether to install at boot time? Nice.
[13:15:59] <_abc_> That is a live iso, right
[13:16:00] <skunkworks> yes
[13:16:04] <_abc_> Thanks
[13:16:14] <_methods> skunkworks: you can put a straight edge across the face after you face it then touch off to that
[13:16:18] <_methods> not the most accurate way
[13:16:23] <PetefromTn_> I would think that you might be able to face off a point and then using a straight edge touch off the right facing tool with it.
[13:16:28] <PetefromTn_> LOL
[13:16:36] <skunkworks> heh - great minds??
[13:16:48] <_methods> heheh
[13:16:54] <PetefromTn_> ;)
[13:17:05] <_abc_> I am not following this exactly. You need to touch off on X or Y after planing the top?
[13:17:16] <PetefromTn_> this is on a CNC lathe
[13:17:26] <PetefromTn_> he is I believe asking about a righ facing turning tool
[13:17:32] <PetefromTn_> right
[13:17:39] <_abc_> Hmm.
[13:18:08] <_abc_> Well what do you touch it off "on"? If the previous tool was left cutting.
[13:18:30] <PetefromTn_> Well I got another job opportunity from the large local company I did some work for a little bit ago. Bidding it right now. Glad to see they thought enough of my previous work to consider giving me more!
[13:18:41] <_abc_> Personally I optimize my designs for one tool cutting even if I have to make a special tool for it.
[13:19:30] <PetefromTn_> as we suggested you can face off the end of a piece of stock and use the flat face to locate the opposite facing tool using a straight edge not the best way perhaps but it would work
[13:20:09] <PetefromTn_> _abc_ Do you have a CNC lathe?
[13:20:30] <_abc_> I use a 3d cnc mill and I usually mill very small things or engrave smd prototype boards.
[13:20:41] <PetefromTn_> ok
[13:21:03] <_abc_> Let me put it this way: the largest cnc mill I used on that mill, ever, was 6mm shaft and point tip vee cutter.
[13:22:09] <PetefromTn_> nice so you must be well versed in 3d milling then.
[13:22:34] <_abc_> Well I'm more of an electronics and embedded guy. So I wrote some cnc optimizers for what my tools output ;)
[13:22:52] <_abc_> But usually it's prototypes and special jobs and very small series.
[13:24:25] <PetefromTn_> interesting
[13:24:34] <_abc_> I heard grinding Tungsten Carbide bits a lot causes people to have fascinating lung x-rays for a while after it. W-C is high contrast stuff for xtay.
[13:24:37] <_abc_> *ray
[13:25:06] <_abc_> Will be interesting when I get to it. My diamond wheel universal grinder is right next to the mill. I make my own mills, as I said.
[13:25:13] <PetefromTn_> I grind my own tungsten for the Tig welder and I buy the non- radioactive stuff as much as possible heh
[13:25:50] <_abc_> It's not radioactive but it is a heavy metal and it will show up in high contrast on any chest xray. Your doc will probably ask if you smoke or what. I don't...
[13:27:11] <PetefromTn_> http://www.twi-global.com/technical-knowledge/faqs/health-and-safety-faqs/faq-the-use-of-thoriated-tungsten-electrodes/
[13:27:22] <PetefromTn_> actually they are slightly radioactive
[13:27:41] <PetefromTn_> they contain thorium
[13:30:07] <_methods> hehe i haven't ground an end mill since school
[13:30:20] <_methods> we just send them out for regrind
[13:31:28] <PetefromTn_> The only grinding I ever do on an endmill is to cut relief for the flutes on the shank so it does not rub at depths beyond the flute length.
[15:06:00] <jdh> and your teeth, at night.
[15:07:32] <PetefromTn_> what do you think I am a Ferengi? hehe
[16:33:53] <unfy> erf. look up a largish nema34. do the math on a cheap lead screw that has a 1/2" diameter and when you couple the MA of the lead screw with the force of the motor you discover the motor could snap the lead screw O_o
[16:43:45] <MarkusBec> did anyone noticed the mesa 7i92?
[16:43:59] <MarkusBec> is it a new mesa card?
[16:44:21] <MarkusBec> I cant find andy documentation
[16:51:42] <Deejay> gn8
[18:18:28] <Tom_itx> MarkusBec, http://www.mesanet.com/fpgacardinfo.html
[18:18:43] <Tom_itx> Ethernet Anything I/I card
[18:20:57] <Tom_itx> ethernet is fairly new yes
[18:21:40] <SpeedEvil> 1978 at a guess?
[18:23:34] <_methods> token ring
[18:39:18] * Jymmmm smacks _methods with the token
[18:42:58] * PetefromTn_ steps in front of methods and deflects Jymmmmm's pitiful girlyman slap and bitch slaps Jymmmm back. Nobody slaps my pal Methods hehehe
[18:48:28] <PCW> 3 mb/s Ethernet maybe only at Xerox in 1978
[18:50:10] <_methods> hehe
[18:50:32] <_methods> he slapped me with the token and stole ring
[18:50:35] <PetefromTn_> ;)
[18:50:38] <_methods> shady dog
[18:54:00] <Tom_itx> PCW, what would it take to talk you outta a couple chips for my boards?
[18:54:26] <PCW> let me get a bag...
[18:54:42] <Tom_itx> hopefully the rest is still good
[18:54:54] <Tom_itx> i checked the 5v and it's good
[18:55:07] <Jymm> PetefromTn_: So, what you are saying is _methods is your bithc and you are her pimp?
[18:55:08] <Tom_itx> dunno what happened really...
[18:55:42] * Tom_itx thinks Jymm should act like an op in a respectable channel
[18:55:47] <_methods> hey i have feelings
[18:56:03] <PetefromTn_> hehehe
[18:56:08] <Jymm> _methods: Your are pretty
[18:56:09] <PetefromTn_> jealous?
[18:56:14] <_methods> hahah
[18:56:36] <Jymm> Tom_itx: Find me a respectable channel, this IS irc afterall
[18:56:52] <_methods> #reprap lol
[18:57:05] <_methods> #ubuntu
[18:57:07] <Jymm> _methods: I said respectable, not pathtic
[18:57:08] <pfred1> #recrap
[18:58:44] <pfred1> think these brackets will be heavy enough for my CNC machine's corners? http://i.imgur.com/CDEloWf.jpg
[18:59:12] <pfred1> I could always weld gussets to them if they're not
[18:59:30] <Tom_itx> how big is your cnc gonna be?
[18:59:48] <pfred1> the whole machine?
[18:59:58] <pfred1> I think it is about 3x4?
[19:00:10] <Tom_itx> plasma?
[19:00:13] <pfred1> like 3 feet wide and 4 feet long
[19:00:24] <Tom_itx> what are you making?
[19:00:29] <pfred1> wood router
[19:00:34] <Tom_itx> probably
[19:00:53] <pfred1> I had 1/8" anglesi n the corners and the thing bent up like a doamond on me
[19:00:58] <pfred1> diamond
[19:01:17] <pfred1> I wasn't too happy about seeing that
[19:01:28] <Tom_itx> use square tubing instead
[19:01:37] <Tom_itx> or weld a triangle
[19:01:46] <pfred1> so now I put angles on the outside threaded rod on the inside and these brackets in the corners
[19:02:01] <zeeshan> has anyone run emt conduit before?
[19:02:06] <pfred1> well, I am only done with one end
[19:02:11] <pfred1> I have
[19:02:28] <pfred1> but I'm no wizard bending it
[19:02:44] <pfred1> I know people who are so I know I'm not
[19:02:58] <zeeshan> do you run a ground
[19:03:00] <zeeshan> even with emt?
[19:03:11] <zeeshan> er
[19:03:13] <Jymm> This clips are not half bad, 6/$1 at dollarstore http://www.dhresource.com/albu_276218779_00-1.0x0/multi-purpose-colorful-clip-clothes-folder.jpg
[19:03:13] <zeeshan> thats not what i meant
[19:03:20] <zeeshan> i mean do consider 4 conductors
[19:03:22] <zeeshan> when picking the size
[19:03:29] <pfred1> one good idea is to leave some blow line in
[19:03:38] <pfred1> in case you ever want to run more wire
[19:03:40] <Jymm> Thean can hold a sheet of paper and be screwed down
[19:04:03] <Jymm> without marring the matieral
[19:04:36] <Jymm> stronger than a clothespin
[19:05:02] <_methods> emt is cheap i always get enough to double the conductors i'm gonna pull
[19:05:06] <_methods> makes it easier to pull too
[19:05:32] <_methods> double the conduit size for conductors i mean
[19:06:22] <pfred1> zeeshan a ground is not required but considered good practice so it is up to you
[19:06:26] <Jymm> good for clamping laminate, veneer, pcb, etc
[19:07:31] <pfred1> yeah with EMT there is a limit to how many wires you're supposed ot stuff into it
[19:07:53] <pfred1> I mean you can't just pack them with as many wires as you can fit
[19:07:59] <PCW> Tom_itx OK 3 422 driver chips 74HC132-tssop and a couple FAN2001s (probably nearly impossible to solder) in a envelope
[19:08:01] <PCW> PM me your address and will mail tomorrow
[19:08:40] <Tom_itx> ok thanks
[19:09:40] <Tom_itx> U2 3 & 6 on the 7i47 and U9 on the 7i43
[19:10:40] <andypugh> So, I am making a clock with an Arduino. It is driving a 1 pulse every 30 seconds pulse-clock.
[19:10:57] <zeeshan> http://cdn.mrsupply.com/images/P/Siemens-HNF361_m.jpg
[19:11:01] <zeeshan> is there a non fused version of this?
[19:11:13] <zeeshan> 100A disconnect
[19:11:26] <Jymm> zeeshan: Why non fused?
[19:11:34] <Jymm> or do you want breaker?
[19:11:37] <zeeshan> i already have upstream breaker.
[19:11:43] <Valen> Tom_itx: did you break something?
[19:11:46] <Jymm> zeeshan: how far away?
[19:11:50] <zeeshan> 32 feet
[19:11:57] <andypugh> So that it has some idea of the last position it left the hands at after re-start or re-set I think I need to write the time of the last pulse to eeprom.
[19:12:00] <Tom_itx> Valen, yep
[19:12:04] <Jymm> zeeshan: get a breaker at least, it never hurts
[19:12:07] <Tom_itx> no idea what happened really
[19:12:09] <Valen> did you do something silly?
[19:12:15] <zeeshan> Jymm: there already is a breaker
[19:12:21] <zeeshan> upstream to the future disconnect switch
[19:12:24] <zeeshan> thats why it doesnt need to be fused
[19:12:24] <Jymm> zeeshan: more never hurts
[19:12:30] <andypugh> Now, the eeprom only has 100,000 writes and that gets used up pretty quickly with 30-second writes.
[19:12:31] <Valen> heh I defeated the current limit on mine with phat stacks of copper wire soldered onto the fet legs
[19:12:48] <Jymm> andypugh: why is it WRITING at all?
[19:12:55] <_methods> yeah why use the eeprom vs the tcr0a?
[19:13:17] <andypugh> Because I want to be able to recover from power failure
[19:13:22] <_methods> ahhh
[19:13:30] <Jymm> andypugh: thats what gps are for =)
[19:13:37] <Jymm> andypugh: and WWV =)
[19:13:41] <PCW> need some FRAM
[19:13:58] <andypugh> The whole basis of the clock is an MSF radio receiver.
[19:14:15] <andypugh> But it needs to know where the hands were last parked..
[19:14:24] <pfred1> I have one of those atomic radio clocks its is great
[19:14:26] <andypugh> The hand-control is open loop.
[19:14:27] <PCW> is there 12 noon index?
[19:14:35] <andypugh> No, no index.
[19:14:41] <PCW> rats
[19:14:56] <Tom_itx> Valen, i don't think so but you never know
[19:15:02] <Jymm> zeeshan: http://www.alcoelectrical.com/Creative%20Concepts%20036.jpg
[19:15:12] <andypugh> It occurred to me that the cheapest way to get feedback of the hands is probably a web-cam at the other side of the Quadrangle :-)
[19:15:13] <Tom_itx> i'm gonna have to go thru the whole wiring and make sure it's all good
[19:15:14] <zeeshan> sexy!
[19:15:21] <zeeshan> why is that disconnect switch so huge
[19:15:27] <PCW> sense a power failure and write EEPROM only then
[19:15:39] <_methods> have limit switches on the hands
[19:15:45] <_methods> like those rotor switches
[19:15:56] <Jymm> andypugh: battery back?
[19:16:11] <_methods> or optical encoders
[19:16:21] <Valen> why are they in 2 boxes is what I want to know ;->
[19:16:33] * Valen loans andypugh a supercap
[19:16:35] <Jymm> andypugh: http://www.hobbytronics.co.uk/real-time-clock-module-ds1302
[19:16:52] <andypugh> So, my thinking is that I have 100,000 writes but I have 1k of eeprom, so I can write the 3 x MSBs to addresses 0, 1 and 2, then use byte 4 to indicate where byte 4 is stored….
[19:17:11] <Tom_itx> PCW the FAN2001 those 3 legged SOP chips?
[19:17:25] <andypugh> (Sorry, store byte 4 at the address pointed to by byte 3).
[19:17:33] <PCW> 6 pin QFNs I think
[19:17:36] <andypugh> Anyone see a flaw with this plan?
[19:17:52] <Tom_itx> oh i see em
[19:18:04] <Jymm> andypugh: http://playground.arduino.cc/Main/DS1302
[19:19:06] <Tom_itx> andypugh why not use flash?
[19:19:15] <andypugh> Clock is easy. It is remembering where the hands were last in the event that someone presses reset, plugs in a serial terminal, pulls the plug etc
[19:19:20] <Tom_itx> i suppose they have about the same write cycle
[19:20:14] <andypugh> With 1k of eeprom I can easily have 50 years at 100,000 writes. I think I will be beyond caring by then. :-)
[19:20:39] <Tom_itx> or you'll be working on rev 2
[19:20:47] <PCW> is it really EEPROM or flash EEPROM
[19:21:03] <Jymm> andypugh: Why not the battery backed RTc I linked to? You COULD use a SUPERCAP with it as well
[19:21:05] <Tom_itx> use one of those serial eeprom
[19:21:16] <PCW> (can you erase a single byte/word?)
[19:21:26] <Tom_itx> i doubt it
[19:21:31] <andypugh> It isn’t the actual time that is the problem.
[19:21:33] <Tom_itx> they're probably block or chip erase
[19:21:48] <Tom_itx> SD is 512byte blocks
[19:22:08] <andypugh> This is only a 1k eeprom, but that is a good question.
[19:22:38] <PCW> for flash you can just write the count plua a serial number and find the largest serial numberwith valid checksum, at startup
[19:22:54] <andypugh> Jymmm I am not storing the time, I am storing the last position the hands were left in
[19:23:18] <Tom_itx> yeah just increment the eeprom address and read the last one written
[19:23:18] <_methods> zero out the hands after a power off
[19:23:30] <_methods> then have them move to the correct time
[19:23:51] <andypugh> _methods: And how do I know where the hands are to zero them?
[19:24:14] <_methods> add a sensor or switch to the hands
[19:24:30] <_methods> hall sensor
[19:24:57] <Jymm> andypugh: In school, all the mechical clocks were slaves to a MASTER mechclock in the schools office, sync to the recess bells. I remember after a power failure, they would spin like crazy to the correct time. Maybe look into what they did?
[19:25:51] <andypugh> _methods: It’s this clock: http://farm1.static.flickr.com/3/4657514_53908e6bb7.jpg
[19:26:23] <andypugh> Jymm: Yes, this was exactly that type of clock, but the Master was in a building that has been demolished....
[19:26:25] <_methods> heheh well that's one way to tell me to STFU lol
[19:26:26] <Jymm> andypugh: is the black area metal?
[19:26:45] <andypugh> Jymm: Yes
[19:27:30] <Jymm> http://www.instructables.com/id/Arduino-based-master-clock-for-schools/
[19:27:34] <pfred1> andypugh just get a bunch of Raspberry Pis and run ntpd on them all :)
[19:27:58] <andypugh> I thought about Pi and NTP, but I want something that will run for decades.
[19:28:10] <pfred1> LInux will do that
[19:28:19] <Jymm> andypugh: http://www.hvtesla.com/masters/masters_index.html
[19:28:33] <pfred1> donno how long a Raspberry Pi will last though
[19:28:42] <andypugh> Yeah, I have seen all these pages :-)
[19:29:07] <pfred1> recap it with solid state caps and it'll probably last
[19:29:09] <Jymm> andypugh: http://www.hvtesla.com/masters/slaves_intro.html
[19:30:55] <_methods> http://playground.arduino.cc/Code/I2CEEPROM
[19:31:04] <_methods> add a secondary eeprom
[19:31:13] <pfred1> http://www.10000yearclock.net/learnmore.html
[19:33:29] <andypugh> _methods: That might help. I guess it rather depends on if that is block-erase and if the Arduino eeprom is block-erase…
[19:33:52] <_methods> well i wonder if you could spread it betwee multiple eeproms
[19:34:20] <_methods> have multiple sacrificial eeproms to spread the writes between
[19:34:21] <andypugh> Was the suggestion of searching for the highest serial that the eeprom chips know which cells are empty?
[19:34:23] <_methods> cheap too
[19:35:04] <_methods> not sure how i'd tag the latest write location
[19:35:16] <_methods> yeah keep them all empty
[19:35:21] <_methods> just shift it
[19:35:38] <Jymm> andypugh: Just realize that no matter how much you fuck up the clock, it' be right twice a day =)
[19:35:58] <pfred1> Jymm that is not true with digital displays
[19:36:13] <Jymm> pfred1: It aint digital though (yet)
[19:36:42] <pfred1> the digital display atthe gas station I go to is unintelligible
[19:36:57] <pfred1> it is always like 72:99
[19:37:14] <andypugh> So, perhaps I read the MSBs, then just look for the highest value in any other location. I think this will need a deeper understanding of exactly how the eeprom is managed.
[19:37:20] <Jymm> andypugh: Do you have no way of attahing anything to the clock arm axels ?
[19:37:47] <andypugh> Not without modifying the antique mechanism :-)
[19:38:01] <Jymm> andypugh: any pics in the internal mechs?
[19:38:20] <andypugh> Yes, I am uploading as we speak.
[19:38:25] <Jymm> k
[19:39:54] <pfred1> andypugh replace the guts with steppers and use LinuxCNC!
[19:40:21] <Jymm> andypugh: I'm thinking there AHD to be some way to know it's position previously.
[19:40:25] <Jymm> HAD*
[19:41:58] <andypugh> No, they jut trusted them to work, and a little man had a full-time job looking after every clock on campus.
[19:42:19] <Jymm> andypugh: (seriously?)
[19:42:25] <andypugh> Yes
[19:42:42] <Jymm> lol, ok. I'll wiat for the photos =)
[19:42:46] <Jymm> wait*
[19:43:23] <andypugh> pfred1: I did like the idea of having two servos and making the hands always take the long way round….
[19:43:57] <andypugh> But another fun idea I have is to make the clock always show Sidereal time. It is a science college, after all. :-)
[19:45:37] <andypugh> (You would think that the second-best University in the world could look after its own clocks, wouldn’t you?
[19:46:11] <Jymm> andypugh: Page 5 http://www.ats-usa.com/pdf_apnotes/MCMODES.pdf
[19:46:34] <Tom_itx> PCW are the FAN2001 the 3.3v regs?
[19:46:48] <Tom_itx> smps
[19:46:58] <PCW> one is 3.3 other is 1.2
[19:47:09] <Jymm> andypugh: http://sound.westhost.com/clocks/alternate.html
[19:47:22] <PCW> bbl Dinner!
[19:47:26] <Tom_itx> does the 3.3 feed the 1.2 or both from 5v?
[19:47:34] <zeeshan> http://www.homedepot.ca/product/1-1-4-in-emt-set-screw-connector/910107
[19:47:40] <zeeshan> anyone know what size hole this needs to fit in
[19:47:43] <zeeshan> it cant be 1-1/4"
[19:47:46] <zeeshan> cause thats the tube passing!
[19:47:57] <andypugh> Jymm: that’s a different style of slave clock..
[19:48:29] <Jymm> andypugh: K, just throwing things your way do give some ideas/resources =)
[19:48:57] <andypugh> http://youtu.be/LwpqX9QK1zU
[19:48:59] <PCW> Tom_itx: just turned computer off so you will need an Ohmmeter...
[19:49:08] <Tom_itx> np
[19:49:45] <Jymm> andypugh: is that YOUR video?
[19:49:50] <andypugh> Yes
[19:50:03] <Jymm> andypugh: SPEAK UP next time, your voice is quiet =)
[19:52:05] <Tom_itx> zeeshan, http://amftgs.com/techlibrary/EMT%20Fittings%20Product%20Cut%20Sheets%20Online.pdf
[19:52:11] <Jymm> andypugh: Hey, I see number markers on the gear. Can't you paint it all black except for a shiny spot then use an opticla interupter to know the position of the hands?
[19:52:30] <andypugh> That’s good for minutes, but not for hours.
[19:52:32] <zeeshan> http://brassmein.com/tech/holesaw.htm
[19:52:37] <zeeshan> thats what i needed
[19:52:49] <Jymm> andypugh: there's no hour gear?
[19:53:14] <andypugh> Yes. But It is rather small and buried.
[19:53:47] <andypugh> I will see how it goes. These things always used to run open-loop. It should be fine.
[19:53:56] <Jymm> andypugh: oh, that tiny 1" thing?
[19:54:22] <andypugh> That’s the “Canon gear"
[19:54:26] <Jymm> ah
[19:54:43] <Jymm> I guess I need to see the pics to have other ideas
[19:54:56] <Jymm> andypugh: what voltage are the coils?
[19:55:40] <andypugh> 24V
[19:55:47] <Jymm> AC ?
[19:55:51] <andypugh> DC
[19:55:53] <Jymm> k
[19:55:58] <Jymm> (just curious)
[19:56:17] <Jymm> andypugh: interesting project
[19:56:26] <andypugh> It would be interesting to see what they did with 240V AC..
[19:56:37] <Jymm> lol
[19:56:55] <Jymm> andypugh: Oh, does this clock have a bell/chime?
[19:57:02] <andypugh> Quite a few coils work on 24V DC or 240V AC.
[19:57:22] <andypugh> No, no chime. But we could add one now.
[19:58:02] <Jymm> heh, just make it sound realistic, the fake ones sound SOOOOOOOOOO bad
[19:58:23] <Jymm> andypugh: how the pics coming along?
[20:00:05] <andypugh> I could wire it to these: http://youtu.be/RjjfJe469As?t=1m3s
[20:00:29] <andypugh> Ah, no still pics at the moment, just the unexciting video
[20:00:32] <Jymm> andypugh: YEEEEEEEESSSSSSSSSSS!!!!!!!!!!!!!!!
[20:00:56] <Jymm> andypugh: that would be awesome!!1
[20:01:18] <Jymm> but after 90s annoying as hell =)
[20:01:23] <andypugh> And difficult
[20:01:39] <Jymm> Just one bad ass bell will do tyvm =)
[20:02:03] <Jymm> hells bells style
[20:02:27] <Jymm> andypugh: https://www.youtube.com/watch?v=etAIpkdhU9Q
[20:03:26] <Jymm> andypugh: AC/DC Every hour on the hour across all of the campus! lmao
[20:04:17] <Jymm> andypugh: Hells Bells for midnight, Highway to hell at 9am, Back in black at 1700
[20:05:03] <andypugh> I saw AC/DC playing Hells Bells at Donington in 1982
[20:05:09] <Jymm> andypugh: "School of Rock" =)
[20:05:16] <Jymm> nice
[20:05:55] <Jymm> dirty deeds at 2100
[20:06:19] <andypugh> (Actually, musy have been 1981 or 1984 :-)
[20:06:41] <Jymm> Hells Bells. (C) 1981
[20:11:59] <andypugh> It seems that Arduino EEPROM erase is byte-level.
[20:12:11] <andypugh> So, Plan A ought to work fine.
[20:12:35] <andypugh> (change locations every 512 minutes)
[20:13:41] <andypugh> http://electronics.stackexchange.com/questions/21232/100k-eeprom-writes-per-bit-or-as-a-whole
[20:14:05] <andypugh> And I should be asleep.
[20:14:07] <andypugh> Night all.
[20:52:36] <Valen> so dad is boosting up the power in his wifes hand held hedge trimmer https://www.youtube.com/watch?v=S0udr_yhmi0
[23:24:06] <ssi> hi
[23:24:21] <pfred1> hello
[23:24:21] <XXCoder> yo
[23:24:36] <ssi> first heat treating experience went quite well
[23:24:50] <pfred1> any caveman can do it huh?
[23:25:00] <ssi> I suppose
[23:25:13] <ssi> I've never tried to do it the caveman way, but doing it the sophisticated way came out quite nice
[23:25:23] <ssi> parts didn't seem to move much at all
[23:25:26] <ssi> maybe a couple tenths, maybe
[23:25:27] <pfred1> I had a job noce where i heat treated printing press dies
[23:25:46] <pfred1> we warped a few of those real bad
[23:25:55] <ssi> lol
[23:26:12] <pfred1> the super didn't think it was so funny
[23:26:21] <ssi> I made some tapered bushings that I would like to have for riflesmithing
[23:26:23] <pfred1> I mean potato chip
[23:26:25] <ssi> and I'm about to heat treat them
[23:26:28] <ssi> but I cut them wrong :(
[23:26:33] <ssi> so I have to remake them anyway
[23:26:38] <ssi> but I'm interestd to see what the ID does
[23:26:43] <pfred1> we used an arsenic oven?
[23:26:45] <ssi> cause they're reamed to .375
[23:26:55] <pfred1> in the morning it looked like this rusty bucket of white powder
[23:26:59] <ssi> and it needs to be a hydraulic fit on a piece of thompson shaft
[23:27:12] <pfred1> the nwe'd fire it up and it would turn into this molten glowing red goo after a while
[23:27:17] <ssi> lol
[23:27:21] <ssi> sounds dangerous
[23:27:24] <pfred1> it was pretty funny
[23:27:29] <pfred1> oh unreal dangerous
[23:27:30] <ssi> I have a little fiber oven
[23:27:36] <ssi> does pretty well
[23:27:49] <pfred1> we'd wire up the dies with tie wire and dip them into that glowing goo
[23:27:58] <pfred1> we had like a rod over the furnace we'd tie them to
[23:28:14] <pfred1> then we'd fish them out and quench them in barrels of salt water
[23:28:27] <pfred1> whoosh! the water would go shooting up the wall!
[23:28:31] <ssi> I be.t
[23:28:35] <pfred1> these dies were liek an inch thick
[23:28:37] <ssi> I'm hardening O1
[23:28:39] <XXCoder> http://hackaday.com/2014/12/04/cutting-glass-with-cnc/
[23:28:44] <ssi> and I bought some honest to god quench oil
[23:28:49] <ssi> and the quenching is pretty uneventful, thankfully
[23:28:56] <ssi> there's a little flame the instant it's dunked, but goes right out
[23:28:57] <pfred1> and when we fished them out they were like translucent almost as if you could see right through them
[23:28:59] <ssi> minimum smoke and smell
[23:29:08] <pfred1> it was weird
[23:29:31] <tjtr33> skunkworks, maybe your opencv can track andypugh's clock hands by doing angle recognition https://www.youtube.com/watch?v=njab2bBps6U
[23:29:35] <pfred1> I mean I've heated up some steel but nothing like what we were doing at this joint
[23:30:22] <pfred1> well before I got there they used to oil quench but they had some problems i heard
[23:30:24] <unfy> i've been meaning to fiddle with opencv, just haven't yet ._.
[23:30:42] <pfred1> I didn't lasti n that place too long it was kind of messed up
[23:31:01] <pfred1> we also hard chrome plated stuff there
[23:31:11] <ssi> that stuff is horrible for you
[23:31:15] <pfred1> yeah
[23:31:33] <pfred1> we had a plastic apron that was so filthy dirty no one wanted to wear it
[23:31:39] <pfred1> that was our safety gear
[23:32:02] <pfred1> there were some masks hanging on hooks on the wall I can't tell you how long they'd been hanging there
[23:32:13] <pfred1> no one wore thoseeither
[23:32:26] <pfred1> you kinda held your breath when you went into the vat room
[23:33:56] <pfred1> place cuold have used better management let me tell you
[23:34:39] <pfred1> the guy that was there before i was went out to lunch one day and never came back
[23:35:00] <pfred1> a week later i quit myself
[23:41:46] <ssi> heh I don't balme you
[23:42:08] <ssi> I need to get my surface grinder wired so I can grind these test blocks
[23:42:32] <pfred1> I've run a surface grinder
[23:42:46] <pfred1> man when I got taught how to do that I thought it was the silliest thing I'd ever done
[23:43:02] <pfred1> the way you have to spin the two handles
[23:43:14] <pfred1> I was good at it though
[23:43:27] <pfred1> man i could make a finish like a magnetic tape
[23:43:40] <ssi> I wouldn't say I'm good at it
[23:43:41] <pfred1> you know how some tapes sort of look like a rainbow?
[23:43:49] <ssi> I still haven't gotten to where I can make a really nice finish
[23:43:56] <pfred1> yeah you'll get it
[23:43:59] <ssi> and I don't have the intuition for setups on the grinder like I do on the mill
[23:44:00] <pfred1> dress the wheel a lot
[23:44:07] <ssi> I need to figure out where my diamond is
[23:44:20] <pfred1> yeah you'll never get a good finish if the wheel ain't dressed
[23:44:36] <pfred1> place i worked for that's what they made wheel dressers
[23:44:42] <pfred1> so we always had 'em
[23:45:08] <pfred1> we did a big business in industrial diamonds too
[23:45:15] <pfred1> that was what kept the shop going when I worked there
[23:45:45] <pfred1> other than that the place was a joke
[23:45:57] <pfred1> but we had soem nice machines to play with
[23:46:30] <pfred1> in its hey day that shop had 40 people on the floor when I was there there were 6 of us
[23:54:05] <ssi> I have a wheel dresser
[23:54:09] <ssi> I'm just not sure where it is at the moment
[23:54:13] <ssi> moving sucks :(
[23:54:42] <XXCoder> always.
[23:59:31] <ssi> I found it
[23:59:32] <ssi> https://pbs.twimg.com/media/B4EaIbZIQAEPSwK.jpg:large