#linuxcnc | Logs for 2013-10-05

Back
[00:34:15] <MacGalempsy> hello? anyone funning a grizzley g0704?
[01:30:12] <mrsun_> hmm need some system to tighten the wires in the machine :/
[02:08:38] <DJ9DJ> moin
[02:13:14] <Jymmm> Howdy
[02:23:57] <DJ9DJ> hello Jymmm :-)
[05:29:34] <Loetmichel> mornin'
[06:21:16] <MacGalempsy> morning
[06:21:38] <MacGalempsy> hey Loetmichel have you heard anything about the Syil cnc products?
[06:21:50] <Loetmichel> never
[06:22:25] <MacGalempsy> it is interesting because they have the exact same machine that Bolton Hardware sells, but yellow
[06:22:32] <MacGalempsy> http://www.boltonhardware.com/category/bolton-tools/metal-lathes-wood-lathes-milling-machines/cnc-lathes-cnc-milling-machines-cnc-machine-center/4-axis-cnc-milling-machine.php
[06:22:38] <MacGalempsy> http://syilamerica.com/machine_x5_vfd.php?view=ts
[06:23:12] <MacGalempsy> except the boltonhardware machine comes with all the accessories and is 700usd less
[06:24:36] <Loetmichel> MacGalempsy: i would suggest both machines fall in china off the same foundry
[06:25:02] <Loetmichel> just the qualitycheck is more or less good ;-)
[06:25:48] <MacGalempsy> heh. still trying to figure out if I should get one of these two, or just settle for the TAIG. it would be nice to find something used with a ATC
[06:25:58] <Loetmichel> but i have not seen any of the machines live.
[06:26:23] <MacGalempsy> me neither, the bolton sales people are fucking idiots, they dont call back
[06:26:39] <MacGalempsy> I tried to get an idea of how square the machine is, but they did not know
[06:26:47] <Loetmichel> wrong
[06:26:53] <Loetmichel> they dont want to tell ;-)
[06:27:23] <MacGalempsy> from the people I talked to, Ithink they are just idiot order takers, not skilled sailes people
[06:27:40] <Loetmichel> ...or the company has intentionally not told them ;-)
[06:28:27] <MacGalempsy> lol. yeah you are probably right about that
[06:28:38] <MacGalempsy> hence the reason no call backs...
[06:28:53] <Loetmichel> i hanve seen HBM BF20 machines wehre they have forgotten to mill the T-slots out of the table...
[06:29:11] <MacGalempsy> holy crap
[06:29:13] <Loetmichel> just reguar cast slots in there ;-)
[06:29:59] <MacGalempsy> just keeping any eye out. the accountant told me to buy it before the end of the year
[06:30:03] <Loetmichel> so much for quality checks ;.-)
[06:30:23] <MacGalempsy> have you heard anything good about the grizzly g0704?
[06:30:29] <Loetmichel> no, sorry
[06:30:42] <MacGalempsy> heh. im shootin zero today
[06:30:56] <Loetmichel> i am german, maybe i have seen machines with similar castings, but the company names would be different over here
[06:31:16] <MacGalempsy> in all the cnc stuff you do, how often do you use an ATC?
[06:31:36] <Loetmichel> that depemds on what an atc is ;-)
[06:31:47] <MacGalempsy> automatic tool changer
[06:31:49] <Loetmichel> never
[06:32:06] <Loetmichel> i have seen some and worked on some less but never build one myself
[06:32:09] <ReadError> thats a pretty big jump
[06:32:17] <ReadError> a 1k to 7k in price
[06:32:18] <MacGalempsy> do you think it would be the time and effort?
[06:32:38] <Loetmichel> for hobby use its not necessary
[06:32:59] <Loetmichel> for production runs it will be required because of "man cost"
[06:33:27] <MacGalempsy> this project is so that I can start a small business
[06:33:47] <MacGalempsy> something to do in between hours of my normal work
[06:34:14] <Loetmichel> depends on what you want to do in that business: single or small run parts dont requre an ATC, mass production does
[06:35:00] <MacGalempsy> I was thinking it was more a time saver while doing more complex custome pieces
[06:36:07] <Loetmichel> for a single part you will stand beside the machine anyway, so you can change the tool by hand
[06:36:36] <Loetmichel> just at mass production one has to supervides a dozen or so machines: no time to change tools
[06:36:47] <Loetmichel> supervize
[06:36:53] <Loetmichel> supervise
[06:37:45] <MacGalempsy> here is the one I am inquiring about http://www.exapro.com/benchman-xt-cnc-vertical-milling-machine-p30801007/#!prettyPhoto
[06:38:33] <MacGalempsy> it is nice how it is contained, but it makes milling longer pieces difficult
[06:55:51] <JT_> Morning
[06:59:07] <mrsun_> damn, that wire did a number on the machine =)
[06:59:19] <mrsun_> steel cable ..
[06:59:23] <mrsun_> maybe its named
[07:19:16] <Tom_itx> MacGalempsy i wonder if the side pannels open up. the left one doesn't look like it does. we've run a few parts with those doors open when they othewise wouldn't fit
[07:40:57] <MacGalempsy> Tom_itx: I think the door on the left will open
[07:41:11] <MacGalempsy> oops, I mean right
[07:41:51] <MacGalempsy> however, the window on the left could be popped out with the rubber seal
[08:00:19] <ReadError> MacGalempsy you never machined but want to start a business doing it?
[08:01:14] <MacGalempsy> ReadError: i machined in high school, and have been getting pretty good with designs on the 3d printer.
[08:01:54] <MacGalempsy> ive never run a cnc machine, but think that over time I could perfect the skills
[08:02:44] <MacGalempsy> im maxed out on my current income with my job and have been wanting to start something on the side. this seemed like a way to get into that
[08:05:24] <MacGalempsy> in some ways I think it would be easier than 3d printing because you dont have to worry as much about temp flux
[08:05:49] <ReadError> lol, 3d printing is way easier
[08:05:53] <ReadError> you dont have to toolpath etc
[08:06:46] <MacGalempsy> true, but the CAM software does that part
[08:07:05] <ReadError> erm
[08:07:08] <ReadError> no
[08:09:15] <The_Ball> Is there a Fedora build how-to somewhere?
[08:20:37] <Einar> Any suggestion on an easy way to get LinuxCNC install for a lathe? I've been banging my head against the wall the whole day! Now it seems none of my PC's have a parport!
[08:21:22] <Einar> What hardware can be suggested that I just stick in a PCI slot, pick a ready made config, hook it up and run?
[08:22:07] <Einar> Oh! My PC's did have parports earlier. I have no clue why not today.
[08:26:04] <Tom_itx> mesa makes good hardware for pci
[08:28:23] <Tom_itx> it won't be plug n go but it's not that difficult to set one up
[08:28:56] <Tom_itx> install from the live cd and pick a lathe configuration
[08:29:04] <Einar> I looked at those, but there are so many cards, breakout boards etc. that I'm afraid I would be in the same situation. Having a combination tahat does not work well.
[08:29:13] <Tom_itx> 5i25
[08:29:22] <Tom_itx> and possibly a daughter card for it
[08:29:38] <Tom_itx> ask pcw_home when he wakes up
[08:30:32] <Einar> Yeah.. it's that "possibly" that scares me. Their website is not much help. At least not to someone that does not have a clue what will work with LinuxCNC.
[08:30:39] <Tom_itx> would depend somewhat if you have steppers or servos
[08:30:47] <Einar> Steppers.
[08:32:03] <Einar> Any idea why on a PC that worked some days ago, now lspci does not mention any parport?
[08:32:05] <Tom_itx> i think there is a combo you can purchase already for that
[08:32:14] <Tom_itx> the 5i25 bundled with another card
[08:32:40] <Tom_itx> nope
[08:32:59] <Tom_itx> dmesg may show a clue
[08:34:12] <Tom_itx> you should first run a latency test to make sure the MB is a viable option before you go too far
[08:35:42] <Einar> I found 3 PC's where the latency is around or less than 50000. For USC that should be fine. But it is very frustrating moving back to square one after just spending a few days wiring up the machine.
[08:36:15] <Einar> Then, when getting back to the PC, I'm lost again. SIGH!
[08:36:19] <Tom_itx> once a few wake up they may be able to help with the parport
[08:38:11] <Tom_itx> you would get better performance from the mesa card anyway though
[08:54:25] <Einar> Now I put in a PCI parport board. It shows in lspci. Could not find USC there either.
[08:55:26] <Einar> Disconnected USC. Powered it off, on again and connected to this PCIboard. Then it works.
[08:56:33] <Einar> Then comes the really strange part: Connected it to the builtin parport and ran it at 0x278. And it works!! That parport is however not shown by lspci.
[08:59:05] <The_Ball> Einar, the mesa cards work very well with linuxcnc
[09:14:20] <Einar> In what file is the address of my parport set?
[09:19:37] <CaptHindsight> Einar: check the BIOS settings
[09:20:35] <Einar> I know the parport is at 0x278. dmesg bitches because it was not found at 0x378. I need to find where to change the expected address.
[09:28:40] <CaptHindsight> Einar: http://linuxcnc.org/docs/EMC2_Integrator_Manual.pdf page #10
[09:29:27] <Einar> I found that, but not the file where it's supposed to be. I found one, but it seems that cannot be the one LinuxCNC is actually using.
[09:30:30] <CaptHindsight> page #10 .hal
[09:31:39] <jp_mill> what hal file is your ini calling
[09:31:52] <Einar> I'm runnning from linuxcnc-dev and I just found it is using linuxcnc/configs/univstep.
[09:32:26] <Einar> Not linuxcnc-dev/configs/univstep.
[09:36:39] <jp_mill> how are you starting your config?
[09:37:26] <Einar> Startting my config? I edit it in the text editor.
[09:42:24] <Einar> So I fixed it in all existances of the config. But now suddenly the test program cannot find the board either. So I am dealing with both hardware and software problems it seems.
[09:44:49] <DJ9DJ> re-hi
[09:47:45] <Einar> Powered all down cold. Restarted. Ran univstepdiags, OK. Started linuxcnc. dmesg says in there "linux parport parport 0 does not support mode 4" Then it informs it cannot find the board. Parport address is now correct.
[09:48:12] <jp_mill> Einar: have you tried just using halrun via terminal?
[09:48:13] <Einar> After this I run univstepdiags again, and now it cannot find the board!?
[09:48:44] <Einar> halrun? I have no idea. I run linuxcnc.
[09:48:50] <jp_mill> http://www.linuxcnc.org/index.php/english/forum/24-hal-components/27024-8255-based-io-cards
[09:49:08] <jp_mill> that was for a different car but same idea^
[09:49:40] <jp_mill> look down and andy had directions in yellow to test via terminal
[09:56:23] <Einar> I did. And it seems a lot of PPMC cards are found. But I have only one card, and it's a USC. So my conclusion is it's reading random data from the parport.
[09:58:02] <Einar> Even though the univstepdiags is running fine. But after the attempt at running LinuxCNC or halrun it will not be found bu univstepdiags anymore.
[09:59:00] <Einar> A full powerdown of PC and USC will fix this, and univstepdiags will run and declare everything is OK.
[10:00:29] <Einar> Disconnecting and power cycling just the USC will not clear the problem.
[10:08:45] <pcw_home> Are you running the USC from the built-in port or PCI port?
[10:09:12] <Einar> The built-in. I just put in a PCI card and will try that.
[10:17:56] <pcw_home> Hmm I would expect the built in port to be less trouble :-(
[10:18:36] <pcw_home> but the built in port might be bad/marginal in some way
[10:36:28] <Einar> Lost my Internet connection temporarily. What I said (into a non existing network) was: And that works. At least it does at the moment. So it seems the univstepdiags is not catching the problem I have.
[10:39:32] <Einar> Now trying to prune down the mill config to a lathe config. Dealing with the errors as they are reported.
[11:58:06] <IchGuckLive> hi all B)
[12:12:40] <jp_mill> hey
[12:14:00] <IchGuckLive> hi jp
[12:14:10] <IchGuckLive> how is the weather
[12:14:16] <IchGuckLive> its rianing here ;-8
[12:16:54] <jp_mill> not bad 15degC
[12:19:01] <IchGuckLive> iv seen 35cm of frech snow on CNN in colorado
[12:19:49] <IchGuckLive> not unusual at this time of year
[12:21:55] <jp_mill> yeah i can wait for that we get enough of that dec - mar
[12:22:31] <IchGuckLive> im off BBQ here today last oktober fest weekend celebrations
[12:22:32] <IchGuckLive> BY
[14:29:30] <archivist> andypugh, your gearbox measurement may cockup where there are two ways to get a ratio
[14:42:06] <ries> hey guys, let's say I am at G0A0X0Z10 and then issue G1A6000X50 what does LinuxCNC do with F? Eg, what should I take into consideration if I want to keep my mill over wood speed at 2400mm/minute?
[14:42:19] <ries> when I just issue F2400 the angular speed is way to fast
[14:43:28] <archivist> the angular axis currently has no clue of radius
[14:44:11] <archivist> I fiddle the f number for some sense, others use the inverse time mode
[14:46:23] <ries> archivist: I have been 'fiddleing this to '40' so it looked right (and looking at the chip) but I was wondering for a calc
[14:46:54] <ries> G93 seems to make sense though, since I can calculate the length aroun dmy stock, then I know how long it should take to complete
[15:05:33] <archivist> hmm strange time of day for other self to be timing out
[15:18:04] <ries> archivist: Programming G93 works very nice, just tested it and works as expected
[15:20:11] <archivist> needs fixing one day so it is more like the css mode of a lathe
[15:24:08] <Einar> The Z DRO in axis always shows zero. When program runs the graphics seems right. No machine connected, so I don't know what the step/dir outputs do.
[15:50:10] <andypugh> archivist: How can there be two ways to get a ratio?
[16:03:57] <jp_mill> not being a huge user of steppers. what is a typical max vel for a mini mill?
[16:06:00] <andypugh> You can reckon on steppers running up to something between 300 and 500 rpm in my limited expereience.
[16:10:26] <jp_mill> ok thanks got the mini mill setup now and i can get about 50 ipm out of it
[16:11:31] <jp_mill> probably good enough for anything anyone would want to do on a small mill
[16:27:54] <andypugh> That sounds about right, actually.
[16:31:55] <jp_mill> cool. I always tried to say away from steppers but they do work quite nicely
[16:45:14] <DJ9DJ> gn8
[16:51:07] <Loetmichel> my personal experience is that every stepper with 200 steps/rev and size below 60mm square skips out at about 10 rev/sec when driven with 40V or less
[16:52:06] <Einar> What is SCALE in the .ini file?
[16:56:39] <jp_mill> Einar: Step_Scale?
[16:57:58] <jp_mill> oops scale is step scale either name can be used
[16:58:06] <Einar> No, just SCALE. And then there is OUTPUT_SCALE which is set to 1.000
[16:58:12] <jp_mill> so Scale is number of counts per machine unit
[16:58:49] <jp_mill> i just finished setting up my mill
[17:00:11] <jp_mill> So 200deg motor at 5X microstepping with a 3:1 gear ratio and a .100" per rev screw gets a scale value of 30000
[17:00:46] <Einar> So with a 200 step/rev motor running half step and a 5mm screw its should be 200*2/5= 80 ?
[17:00:53] <jp_mill> for one inch of travel
[17:01:02] <jp_mill> my machine is in inches
[17:01:58] <jp_mill> is you motor connected straight to the screw?
[17:02:16] <Einar> No, there's a cogged belt.
[17:02:23] <jp_mill> 1:1 ratio?
[17:02:27] <Einar> Yes
[17:02:33] <jp_mill> that sucks
[17:02:43] <Einar> Uhh?
[17:02:53] <Einar> Why?
[17:03:16] <jp_mill> low resolution just joking around
[17:04:25] <jp_mill> what units are you planning on?
[17:04:50] <jp_mill> millimetres?
[17:07:50] <jp_mill> you will also hae to check your max velocity and acceleration parameters. If acceleration is set too high the motor will just stall and cause you to run around scratching your head
[17:16:42] <ries> archivist: I did a couple of more test, but in my case it will work just fine. I am writing a little tool that makes from a square stock of wood rond in a efficient manner, starting at corners first. Un the end I am going around 960 degrees (keep rotating, like a lathe) soo I needed to calculate a different F when using A20000 for example
[17:18:16] <ries> Einar: I run my machine direct drive without issues, and I don't even have a gear, just rack and pinions.
[17:19:22] <Einar> Ries: I do the same with my mill. But then it runs Mach3, not LinuxCNC.
[17:19:35] <ries> You are missing something then :D
[17:20:01] <ries> I do wood working only, so it's perfectly fine in my situation...
[17:20:37] <Einar> Oh, I don't run rack, but direct drive to ballscrew.
[17:23:01] <ries> Einar: on your questions, this is perfeclty explain here : http://www.linuxcnc.org/docs/devel/html/config/ini_config.html
[17:23:11] <ries> Go to 2.10.3. Stepper
[17:23:49] <Einar> Been there. But I use the USC driver.
[17:26:59] <Einar> I'm trying to put in the right numbers but get a following error as soon as I try moving!
[17:31:06] <ries> out of curiocity, did you take one of the sample config's?
[17:33:02] <Einar> I took the sample config for mill. There is none for lathe. After a lot of mucking around it ran. But when trying to put in the right speeds for my lathe everything blew up!
[17:35:10] <ries> oops
[18:27:18] * FinboySlick hehs, cute: http://www.youtube.com/watch?v=d5pGN6pqkyY
[19:21:59] <archivist> andypugh, think of a screw cutting gearbox or other type of gear box that has high/low range and a reversal
[19:23:47] <andypugh> archivist: Well, this is a spindle gearbox, and the gears are all unique in ratio. (and, if too gears have the same ratio close enough to be detected as the same, then they can surely share the multiplier?)
[19:28:05] <archivist> I think classic ladder like skunkworks used on the K&T spindle, but he has automated it, I know what you are thinking but I expect some exceptions
[19:28:51] <andypugh> it works pretty much perfectly on cars.
[19:30:03] <archivist> if one chose gears for a sensible range
[19:31:00] <andypugh> Spindle gears do tend to be deliberately distributed.
[19:31:36] <MacGalempsy> evening guys
[19:31:52] <archivist> except users look at a comp and use it for another job, I am thinking of my southbend screw cutting gearbox
[19:35:34] <archivist> it has a two way large divider input, a 3 way range selector, and about 10 output gears, one has to select the reverse tumbler for infeed on the cross slide, and the fine feed clamp in the saddle box is different to the leadscrew feed nut
[19:37:04] <t12> anyone happen to know the name of this kind of connector block? https://www.dropbox.com/sc/d4q5hkjfhxu8m7t/gN84lRBwwk
[19:38:09] <DaViruz> pluggable terminal block
[19:38:26] <archivist> one of my stepper drives has that
[19:38:26] <DaViruz> or you mean the model name of that specific model?
[19:38:43] <t12> something to point me in the right direction for finding the mating part
[19:39:05] <t12> i just got this hc-kfs43 + mr-j2s controller working
[19:39:13] <Jp11> Phoenix contact digikey
[19:39:22] <DaViruz> a head on photo and a measurement of the pin pitch would help
[19:39:25] <t12> i thought i has a busted controller until i realized that you need two seperate mains connections for the controller
[19:41:11] <andypugh> Can be a real pain. The come in 5mm and 5.08mm and 3mm 2.54mm
[19:42:18] <andypugh> Your current solution would be OK if you had used square sockets on the square pine.
[19:43:01] <DaViruz> funny how the block seems obscured by the case
[19:43:06] <t12> yeah
[19:43:12] <t12> the last pin isnt in the manual either
[19:43:22] <t12> nor is the connector type, but maybne i jsut havent found it
[19:43:39] <t12> pitch is 5mm or .2", since its a .jp part i assume 5mm
[19:44:04] <DaViruz> i can just barely see a logo and number on top of it
[19:44:29] <DaViruz> id say you can't guess pin pitch by origin
[19:44:48] <t12> oh i didnt notice if there was writing on the block checking
[19:45:06] <andypugh> Yeah, Mesa use metric pitch, and I can't buy the metric versions in the UK..
[19:45:30] <DaViruz> surely farnell stocks them?
[19:45:47] <andypugh> You would have thought so, wouldn't you?
[19:45:56] <DaViruz> i would :-)
[19:46:47] <DaViruz> on a different note, i can get parts faster and with cheaper shipping from farnell in the uk then i can get them from my swedish suppliers
[19:47:49] <andypugh> Have a look at rapidonline too
[19:48:12] <andypugh> Range isn't as good, but they are cheaper
[19:49:35] <t12> ah here we go
[19:49:43] <t12> yeah phoenix, mtsb 2.5, 5.08mm pitch
[19:51:08] <MacGalempsy> repeatibility of .001 seems pretty good. does that sounds right for a grizzly with ballscrews?
[19:51:31] <andypugh> mm or inches?
[19:51:40] <DaViruz> 0.001 feet sounds about right
[19:51:52] <MacGalempsy> inches
[19:52:19] <andypugh> Actually, I would hope for better than that with ballscrews
[19:52:35] <MacGalempsy> hoe much better?
[19:53:59] <andypugh> Depends on how you are measuring. if the indicator is at 0, then G0 X50 / G0 X0 should be as good as the indicator can measure.
[19:55:00] <andypugh> If you are measuring from different approach directions than 0.001 is OK, if not stellar.
[19:55:17] <archivist> depends if your screw has backlash and is the slides are well adjusted
[19:55:53] <archivist> you can get an axis to rotate a bit with some slide types
[19:56:07] <andypugh> Well, _repeatability_ to me means doing exactly the same thing every time, and all those variables should disappear.
[19:56:37] <archivist> not with loose gibs as fitted bo the overseas gents
[19:56:43] <archivist> by
[19:58:17] <MacGalempsy> I would prefer to buy made in USA, but how many machines really are?
[19:58:46] <andypugh> _I_ would prefer made in the UK...
[19:59:11] <MacGalempsy> of course, we love our motherlands
[20:00:05] <MacGalempsy> do are you guys pros or hobbyists?
[20:00:09] <archivist> an old bridgeport will hold better repeatability probably
[20:00:21] <MacGalempsy> i dont have room for an old bridgeport
[20:00:35] <andypugh> I am very much a hobbbyist.
[20:00:47] <MacGalempsy> I have a two car garage and its residential
[20:01:00] <t12> thats way plenty room for a bridgeport
[20:01:01] <MacGalempsy> I think an old bridgeport would be quite disturbing to the neighbors
[20:01:14] <archivist> mostly I fire up a machine to earn some money, but not often enough
[20:01:15] <t12> - the cars
[20:01:20] <MacGalempsy> not with two cars and washing/drying machines!
[20:01:28] <andypugh> But my mill was made in Heckmondwike. And is all the better for it.
[20:01:29] <Jp11> Gee I have a kitamura and a BP in a 2car
[20:01:52] <MacGalempsy> Jp11: how man cars in there at the same time?
[20:02:05] <MacGalempsy> I cannot part my porsche outside :P
[20:02:06] <archivist> we turfed the car and bikes out years ago from a single garage
[20:02:09] <Jp11> None never
[20:02:35] <Jp11> Only the kids bikes
[20:02:41] <andypugh> I have a one-car garage containing 2 motorcycles, a pushbike, a workbench, a Harrison miller, a 9x40 lathe/mill combo and a Rivett 608. It's a bit cramped.
[20:03:00] <MacGalempsy> andypugh: I would guess so!
[20:03:07] <archivist> andypugh, there is another rivett on fleabay...
[20:03:18] <andypugh> I saw it. Looks like a good one too.
[20:03:20] <MacGalempsy> trying to talk the old lady into stacking washer/dryer combo >)
[20:03:31] <Jp11> Yeah stepping over everything is getting old
[20:04:15] <archivist> I dont get nagged here, so little lathe in the kitchen, 5 axis upstairs in small bedroom, big stuff in the garage
[20:04:29] <t12> i've discovered that more space doesnt really work that well
[20:04:41] <archivist> too far to walk
[20:04:47] <t12> eventually you spend too much time walking around
[20:04:57] <t12> discipline with moving shit around and not leaving it around may work better
[20:05:02] <t12> not that i'm any good at taht
[20:05:05] <archivist> I hate the drill being in the garage when working upstairs
[20:05:28] <MacGalempsy> so the grizzly has a 14" z axis, I was thinking this little 5th axis may be fun http://www.ebay.com/itm/CNC-engraving-machine-Rotary-Axis-A-Axis-B-Axis-4th-5th-Axis-CNC-Rotary-Table-/181216504085?pt=BI_Tool_Work_Holding&hash=item2a31586915
[20:06:43] <archivist> seen that before, /me not going to get one
[20:07:14] <MacGalempsy> why not?
[20:07:19] <andypugh> MacGalempsy: With 6" height in the widget and the length that tools are you might find things a bit tighter than you expect.
[20:08:10] <archivist> the stepper under the table has no gearing therefore little to resist cutting forces
[20:08:28] <andypugh> If you can find one with the table nearer the pivot you might find it more useful.
[20:08:45] <archivist> I had serious height problems on my machine
[20:08:59] <Jp11> A mini 2 axis cradle would be cool
[20:09:41] <andypugh> I had to rescue my shortest (chipped) boring bar today to fit the boring head above the work.
[20:10:48] <andypugh> And my drill chuck is almost never usable, I have to put drills in collets.
[20:10:51] <archivist> I had to put a rectangular block under the column to get space to work on the two rotaries
[20:11:15] <andypugh> (There is no reason that the drill could not be an awful lot shorter, that I can see.
[20:11:25] <andypugh> (Drill chuck, that is)
[20:12:12] <Jp11> I use Collets to hold drills just to avoid that
[20:12:52] <Jp11> My drill chuck adds like 5"
[20:13:21] <andypugh> I think mine is one of these: http://www.ctctools.biz/servlet/the-325/13MM-BT40-KEYLESS-INDUSTRIAL/Detail
[20:13:39] <archivist> I find you need that length to get over the rotary to its centre http://www.archivist.info/cnc/stage8/IMG_0268.JPG
[20:13:41] <andypugh> I am sure a lot of the mechanism could hide up the shank.
[20:16:00] <andypugh> archivist: That would look so much better assembled from glued-together granite :-)
[20:16:24] <archivist> hehe, I know how to make fugly
[20:18:44] <archivist> contemplating a new structure for the rotaries at the moment to reduce vibration and make setting up easier
[20:24:50] <MacGalempsy> it is pretty difficult to find a mill with enough z to run the 5th axis
[20:25:31] <MacGalempsy> I thought that with 14" z, a 6" z, that give about 8" with tool and material. a very small 5th axis project
[20:25:47] <MacGalempsy> so maybe 5-6" material?
[20:26:14] <MacGalempsy> archivist: what kind of 5th axis you have in that bedroom?
[20:26:58] <archivist> home brew from two standard rotaries and an angle plate
[20:27:17] <MacGalempsy> pics?
[20:27:37] <archivist> it is on that picture I just linked http://www.archivist.info/cnc/stage8/IMG_0268.JPG
[20:28:17] <MacGalempsy> that looks pretty mad scientist bro
[20:28:51] <archivist> I call it fugly, but it works
[20:29:08] <MacGalempsy> I am hoping to dig deeper into my mad science. working is all that counts.
[20:29:32] <MacGalempsy> and to use this machine to make my enclosed 3d printer :)
[20:29:40] <archivist> the large rotary is a 6inch and the smaller is 5 inch ish
[20:29:46] <MacGalempsy> well, not so much make, but finish and improve
[20:29:52] <andypugh> I am too driven by aesthetics, hence having castings made for my conversion.
[20:30:09] <archivist> arsethetics
[20:30:32] <archivist> I am driven by cost
[20:31:36] <MacGalempsy> it is kind of frustrating trying to figure out the right machine. my budget is about 10k
[20:31:42] <andypugh> I spent thursday evening on "Bother, this casting moved between the two clamp settings with the face mill. I wonder if I can scrape it into a flat?
[20:32:19] <andypugh> MacGalempsy: 10k will buy you a fully working Tormach
[20:32:35] <archivist> a good make
[20:32:40] <MacGalempsy> you mentioned the tormach before.
[20:32:58] <Jp11> What about a syil?
[20:32:59] <MacGalempsy> their site looks good
[20:33:05] <MacGalempsy> I am curious about the syil
[20:33:18] <MacGalempsy> it appears to be the same machine as that bolton tools
[20:33:24] <andypugh> Well, you said you liked the idea of US manufacture.
[20:33:29] <Jp11> They seem all the same to me
[20:33:41] <MacGalempsy> I do, but syil is just sold in US, not made here
[20:34:01] <archivist> yes, it gets badged by all, and is less good (cheap sh...)
[20:34:03] <Jp11> One might spend more on screws and such
[20:34:12] <MacGalempsy> yeah. so with the bolton being 7k, and basically the same thing, I was have really been thinking about going that way
[20:34:32] <andypugh> Tormach don't seem to be quite the same as the others. They seem to be the only one with turcite slides, for example.
[20:34:49] <MacGalempsy> 7k gets machine, stand, oiler, cooling system and 4th axis
[20:34:53] <archivist> if you want accuracy and backup from the seller I would go elsewhere
[20:35:05] <Jp11> or buy a manual one for a grand and do the rest yoirself
[20:35:43] <MacGalempsy> since this is my first endevor into CNC machining, the goal is to get a turnkey model
[20:36:10] <andypugh> Part of the reason I like Tormach is that they are contributing back into LinuxCNC: http://git.linuxcnc.org/gitweb?p=linuxcnc.git;a=commit;h=b6b2c6343a700eb997fc57b648cc7b675beb8d82
[20:37:03] <andypugh> Of course Smithy are even more deeply involved, as far as I can see it was Matt @ Smithy who dragged EMC out of NIST and into the world.
[20:38:10] <andypugh> The new Tormach lathe is LinuxCNC only. The mills are LinuxCNC as an option.
[20:38:29] <archivist> here a seig as sold by a uk seller note the reassembly guide he provides up front http://www.arceurotrade.co.uk/Catalogue/Machines-Accessories/Milling-Machines/Model-Super-X3-Mill
[20:41:04] <andypugh> That's actually looking like quite a decent machine now.
[20:44:09] <MacGalempsy> how does that differ from all the other x3 derivtives out there?
[20:44:16] <andypugh> In the drilling video you can seen the that the drill is shaking the head, but that motor / drive is impressibely quiet.
[20:44:20] <archivist> I helped someone do the spindle rebuild
[20:45:37] <archivist> MacGalempsy, I am not sure it differs at all except for the honesty of the seller and the documentation he has had written to fix the supplied errors in manufacture
[20:45:39] <MacGalempsy> andypugh: remember that benchman xt I showed you the other day? I offered that dude 5k and he laughed. heh
[20:46:20] <MacGalempsy> but there is another guy with one who is checking with his partner about a price.
[20:47:44] <andypugh> Loooking at the Sieg facing video, my little Harrison would _laugh_ at both those jobs :-(
[20:47:54] <andypugh> Wrong bracket :-)
[20:48:29] <andypugh> But then with a 45 rpm bottom gear and a 2hp motor it is not short of grunt.
[20:49:43] <archivist> my horizontal has similar grunt
[20:50:26] <andypugh> Mine was originally designed as a horixontal. I guess that they were intended to turn big cutters rather slowly.
[20:50:30] <archivist> the chinese machines tend to have columns far too small for hard work
[20:51:15] <andypugh> Yes, you can slap the side of the head of my Chinese one and it goes "boing". You slap the side of the head of the Harrison and _you_ go "Oww"
[20:51:17] <archivist> should go back to bed, steam engine driving tomorrow
[20:51:31] <andypugh> Where and what?
[20:51:34] <archivist> http://www.middleton-leawood.org.uk/leawood/whatson2013.html
[20:51:51] <jp_mill> Didn't someone fill theirs up with epoxy and aggregate
[20:52:24] <andypugh> Well, I filled mine with the ballscrew, so that would be incompatible.
[20:52:59] <andypugh> To be fair, my super-neat ballscrew installation did terrible things to the rigidity.
[20:53:32] <jp_mill> ah aesthetics had a price :)
[20:53:45] <andypugh> And bolting a large lump of steel to the back of the column would probably be a huge improvement.
[20:54:32] <andypugh> But the machine is basically a lathe with a milling head, and lathe cross-slides are not good milling tables.
[20:57:26] <andypugh> Compare and contrast: (The working envelopes are very similar) : https://picasaweb.google.com/lh/photo/Dk3wsS9o7XJ-N2kfUnESDtMTjNZETYmyPJy0liipFm0?feat=directlink and https://picasaweb.google.com/lh/photo/LcJ0c7JzGi1VAI-kJZaactMTjNZETYmyPJy0liipFm0?feat=directlink
[20:58:13] <andypugh> The 3 x 12
[20:58:47] <andypugh> The 3 x 1/2" T-slots v the 2 x 7mm T-slots is a huge improvement.
[20:59:05] <jp_mill> I like the lathe setup. How big are those buttons on the Harrison?
[21:00:41] <andypugh> The buttons are 50mm I think. Probably meant for fruit machines. But big and tough.
[21:03:46] <jp_mill> well got the mill running now just have to see if the 7i76 and an off shore dc speed controller play nicely, but that's tomorrows fun. Night
[21:04:14] <andypugh> Time I was making Zs too.
[21:10:22] <MacGalempsy> nighty nighty gents
[22:01:21] <Nekosan> can linuxcnc use a usb connection to the pc from a breakout?
[22:01:55] <pcw_home> for power, yes
[22:02:25] <Nekosan> still have to use a par port for coms?
[22:02:56] <pcw_home> or non real time I/O (PLC like things)
[22:03:12] <pcw_home> for step and dir, yes
[22:03:57] <Nekosan> non realtime is not usefull
[22:04:10] <pcw_home> There is some USB motion code around (Arais Robo) but not merged
[22:04:48] <Nekosan> would be nice to see it work with a ramboboard
[22:08:12] <pcw_home> Perhaps a better target would be some Ramps varient and a BeagleBone Black
[22:08:37] <pcw_home> (probably alreadty working with LinuxCNC)
[22:10:32] <Nekosan> how about a smoothieboard... that looks like a nice alternative
[22:10:43] <pcw_home> BBB is probably better in almost every way
[22:12:08] <pcw_home> PRU (200 MHz 32 bit coprocessor)stepgen, real time Linux on the BBB etc etc
[22:14:06] <Nekosan> looking at building a cnc router, have a frame in mind... now the electronics to drive the steppers from linuxcnc
[22:15:34] <pcw_home> Those little Ramps like boards are not going to drive the stepmotors of a router (unless its really tiny router)
[22:29:54] <Jymmm> http://www.youtube.com/watch?v=ONj4NpuJByY
[22:30:05] <Jymmm> pcw_home: ...in an altoids tin
[22:35:19] <Tom_itx> those drivers are only good for a couple amps
[22:35:25] <Tom_itx> at betst
[22:35:27] <Tom_itx> best*
[22:50:36] <t12> pwc: think theres room on any of the mesa fpga's to handle the mitsu encoder serial?
[22:50:50] <t12> i almost have everything together to do the LA
[22:51:19] <t12> pcw even