lordpil's blog

rechuck to square everything, youre like between one and two hours in now

Posted by renesis at 15:55 | permalink | 0 comments

zeeshan: you know what sucks about small machine tools?
they still want big shop tools to prepare stock
like, big mill is neat, but big mill plus a metal band saw is like, useful
little mill is kind of the same way, except still want a big bandsaw to cut down raw material
the small space alternative is a power jigsaw, which works but then you spend a lot of time on the mill squaring up stock
and im impatient, i always fuck that up, feed to fast or use a barely good enough fixture
right and then cleanup is a single pass, if you cant get away with just chucking the raw cut stock and machining that
jigsaw on a micromill is like, hmm gonna take 20min to get that side straight

Posted by renesis at 15:50 | permalink | 0 comments

like, if you accidentally homed instead of zero'd, in the middle of your work envelope, you could lock yourself out of part of the envelope
and there was no way to reset the limit unless you reset axis gui
if you were in the middle of a program, you basically fucked yourself, had to reference if possible

Posted by renesis at 15:44 | permalink | 0 comments

which to me is like, might as well use them to home the machine
do they still use #emc on freenode?
or are they using #linuxcnc now?
the devs used to be extremely active in there
i dont know anymore
guys smart, dunno if he was a lead dev or what, but he worked on it and knew the system, was a machinist
i would ask in there about hominh details with limit switches
because my guess is theres more than one way to do it
and you can probably use series/parallel switches to reference one axis at a time
if you cant, ask for
im pretty sure the unhoming feature happened because i bitched a lot on irc as the problem was happening to me

Posted by renesis at 15:39 | permalink | 0 comments

rab: found a newer cable, jack is fine, pushing the plug both ways and holding
seriously like a dozen cable failures like this, this phone
no, soft limitss
your shit is nig, no?
big, sorry
but thats actually an industry term =\
i wont say what its short for
yeah do put switches
zeeshan: + rig
im not sure if you can home like that
thats touching off, not homing
i always home so my entire g53 space is negative
and i usually g54 positive
so i can just look and coords and know if im in machine or work space
linuxcnc wont even let you MDI without homing
soft limits are based on your home
yeah you want hard limits

Posted by renesis at 15:34 | permalink | 0 comments

yeah that sucks, it looks like something needs very tight tolerances
they know marketing and relations, wont ever deny that

Posted by renesis at 14:26 | permalink | 0 comments

"5,000 cycles is a very poor rating in the connector industry."
yeah i dunno about that
max cycles on pretty much everything ive ever checked seems low for application
for internal connectors, sometimes you dont even gets hundreds
rab: yeah thats kind of typical for low end board to board stuff
its mean to be assembled, hopefully glued, and left
10, 50, sucks, but its prob better than ide cables
like, working on computers from like 2001 to i dunno 2006, like 75% of my pc problems were ide cables
thunderbolt almost looks to simple to fuckup if they manage to keep tolerances tight
its like inside out usb

Posted by renesis at 14:19 | permalink | 0 comments

yeah but its breaking in the right direction
theyre $5 at frys for 6ft
its annoying but ultimately this is more acceptable then junking my phone
rab: my Samsung Seek slider phones both did the same thing
i dont think ive had a jack failure
i assumed early on the jack was fucking up, but mostly because of mini usb experiences

Posted by renesis at 14:13 | permalink | 0 comments

rab: do you know why microusb cables fail so much?
prob like a dozen microusb cables go intermittent in my phone
i thought it was the jack, but new cables click in nice and work fine
eventually, i have to halg the cable off the desk so it holds the plug in at an angle and makes contact
multiple brands, multiple plug types
i keepy freaking out its the jack, but its solid when i poke at it, and its a galaxyt nexus, between samsung and google i figure they pick out a decent jack
is that like a design goal?
because they hit it
and im frustrated but ultimately happy with the decision
sucks ass but i buy new cable and my shit is fine
cool, mini usb was a heart breaker
the samsung ones dont fail but i lose them

Posted by renesis at 14:08 | permalink | 0 comments

timecop: yeah so were kind of assuming hes attached to you
all that pasted trash and my terminal isnt fucked up at all
pretty weak troll

Posted by renesis at 13:44 | permalink | 0 comments

the vfd is the big plastic thing?
heh probably
just keep it against the chassis
its not as bad as it used to be
or im not around as much to notice, shrug
anyway, laters

Posted by renesis at 09:37 | permalink | 0 comments

use the lock washers and lock nuts like in the image
and you want the power wires to be shorter than the earth wire is possible
in the case the power entry module or cord gets pulled
you want power and neutral to disco before earth
heh, yeah drilling into assembled shit sucks

Posted by renesis at 09:32 | permalink | 0 comments

no grounding = electrocution
this is #electronics
nothing is more important that grounding topics
your system is probably pretty immune to noise
current is the scary part
read that
mount all module grounds to that
why the fuck wouldnt you?
what if the VFD terminal melts?
sorry we dont want you to kill people

Posted by renesis at 09:27 | permalink | 0 comments

isolate the chassis, and star ground them
if you want to know the best way to do it
find a white paper on how to mount earth wires to chassis
theres a pretty standard nut and lockwasher stackup
yeah but you maybe dont need them in isolation
right it has to do with safety complicance
all isolation distances and insulation thicknesses are specified
like, the length of the different power input wires are spec'd
right, located right near live mains power
if youre doing it right
why do you think car fuses blow for no fuckin reason like all the time
bad grounding = fire

Posted by renesis at 09:22 | permalink | 0 comments

yeah this isnt #cars
worse case youre going to get a local current loop that may couple EM fields from one part of your system to another
your shit isnt audio, it doesnt matter
synth: if the screws become loose, you could create resistive coupling
which in a fault may cause a fire
wlel if it makes you feel better, it prob wouldnt be a problem for months or years
long enough for you to build confidence in the system and not take as many precautions while operating it

Posted by renesis at 09:17 | permalink | 0 comments

so yeah in that case you probably want to pull, heh
i dont think it would be difficult to find some sort of fan in a pipe with a shaft for coupling
and a motor outside the pipe
belt, shaft, whatever
shut off all the things
what was the name of the guy an hero'd on the light pole cable?

Posted by renesis at 09:04 | permalink | 0 comments

in a push system youre adding the heat of the fan to the system
but im thinking there are crack rocks involved if this is somehow an issue
zeeshan: you could always reposition the psu so the the fan intake are mated to the wall of the chassis
would just try and make sure the cpu sink heat isnt getting sucked directly into the psus
wtf @ 840W motor

Posted by renesis at 08:58 | permalink | 0 comments

and the lower pressure will suck more air in
like any restrictions from high pressure (or low pressure if the restriction is on the intake side) will eventually be counteracted
assuming the fans move the same amount of air on average
i usually think it has to do with flow around components
with a pull system, air comes from everywhere and your components arent in this path, youre just moving air instead of cooling
when you do push systems, you can put the air directly where it needs to be, and the turbulence from the localized high pressure areas will help to surround components
also, the higher pressure will be created in front of the fans, so if youre components are in this higher pressure area, you benefit from the (slightly? insignificantly?) higher thermal transfer
i dont think push/pull matters nearly as much as internal flow

Posted by renesis at 08:51 | permalink | 0 comments

i dont understand why optimal heat transfer would be at 1 atmosphere
plot there kind of backs up what i guess, at higher desity you get conduction, because shit is closer
so i think the efficiency of a push pull system would just be less stable but would average out over time
and the buffeting might help cooling by getting air movement in places that normally wouldnt
yeah but im saying that higher pressure will turn into lower pressure

Posted by renesis at 08:46 | permalink | 0 comments

and im not sure if airflow suffers, one would think it would be aided as much as it suffers when pressure and fan loads are equalizing

Posted by renesis at 08:35 | permalink | 0 comments

rab: yeah but on average you still get twice the flow

Posted by renesis at 08:28 | permalink | 0 comments

timecop: wtf

Posted by renesis at 01:08 | permalink | 0 comments

Top | Add to Technorati Favorites

© 2007 lordpil.   XHTML 1.0! CSS! Site design by GNAA  Blog Engine by pbx | MULTI2 | ian hanschen | lolwat