cradek, You there?
hmmmm.... quiet this morning
Good morning everyone!
John, you there? I seriously need your help!
I got those new steppers, and they sent me 2 drivers to try but they use Quadrature.
and you want to convert from step/direction to quadrature outputs?
not too hard
save your .hal file under a different name before you start editing
Right, I learned that lessen the hard way!
ok, now near the top of core_stepper.hal, is a line like "loadrt stepgen cfg="0 0 0"
change to "loadrt stepgen cfg="2 2 2"
down further is
newsig Xstep bit
newsig Xdir bit
change from Xstep to X_A, and from Xdir to X_B, same for Y and Z
Ok, I was told to change to XPHaseA bit. Thats probably why it won't start.
doubt it... the choice of names for that signal is arbitrary
you didn't tell me you had already attempted this
if you already tried, then we should start with what error messages you got, rather than starting from scracth
Ok, error Im getting is HAL: ERROR: pin 'stepgen.0.XPhaseA' not found
ok... that isn't the proper name for the pin
there are two different names involved here... the pin and the singal
the newsig command creates a signal, you can name it anything you want
I suggested X_A, but XPhaseA is fine too
then you use a linkps line to connect that signal to a pin on the stepgenerator
I'll use your conventions!
that pin has a name that is fixed by the step generator component
I believe it would be "stepgen.0.PhaseA"
for X... for Y it would be stepgen.1.PhaseA, etc
so the linksp line(s) below the newsig lines should read something like:
linkps stepgen.0.PhaseA X_A
linkps stepgen.0.PhaseB X_B
linkps stepgen.1.PhaseA Y_A
(linkps means link pin to signal, so the pin names comes first)
Ok, I'm with you so far.
oops... the actual pin names are stepgen.0.phase-A
lowercase P in phase, and a dash before the A, B, etc
there may be another problem....
I recently discovered that insmod on BDI-4 doesn't properly handle arguments like cfg="2 2 2"
it should result in a string "2 2 2" being assigned to the argument
I'm using emc2 off cvs
but apparently even with quotes around it, insmod or the shell is seeing the spaces and breaking the string at the whitespace
result is cfg becomes a string like "2" instead of "2 2 2"
that has nothing to do with emc, it's the way the operating system parses the strings
for another driver that I was working on at the time I had to change the strings to something like "2_2_2" to eliminate the whitespace
which also requires the code that parses the string to be changed
I'll try the name conventions first.
ok... I'll test out the "2 2 2" thing here
but what about standard_pinout.hal?
that will need to be changed too, to send X_A out the pin where X_step was before
jmkasunich: do you kow what the current plans for the tp are?
nope... paul has been working on that.
I was just curious as I was tinking of conveting the second laser bed to emc
it lead to a revisit of the NML vocabulary, which has since taken center stage so to speak
but anonimasu thought that the current tp would give trouble at 10m/minute
John, linksp.X_A parport.0.pin-02-out?
no "." after linksp, just a space, but otherwise, yes
ok great, is ther a way to tell emc how long a pulse should be?
for step pulses, yes... for quadrature, the output state stays the same until the next step, there is no such thing as pulse length
* robin_sz nods
ah well, I have to make some sort of decision on what to do with this second laser ...
Ok, Thanks I'm out to garage to try this new info. will report back in a few minutes!
it will be either emc or baldor
strange... cfg="2 2 2" works when it is in a .hal file, but not from the commmand line
must be bash that's fscking around with it
ok, seemed to work fine accept for this EMC_SET_GAINS
emctaskmain.cc 1984: ignoring issue of unknown command 103
did you get that message before (when you were using step and direction drives)?
Not that I noticed!
ok... it's safe to ignore that message, I'll investigate from here
Ok, great I have to do some re-wiring to try out new drivers, Thanks again for your help. Will you be around for a while today?
on-and-off.... I have some shelves to build, and some parts to machine, but both jobs are in the basement where the computer is, so I'll check it from time to time
Ok, Thanks John. See ya later:)
that was quick
Sorry John, one more thing.
To hook up driver, X_A would be CW+ and X_B would be Ccw+. Would I then just tie Cw- and Ccw- to ground?
don't know... I know nothing about your driver.
that sounds reasonable, but you need to consult the driver documentation
jmkasunich: bash probably won't pass along the quotes
Ok, Thanks. Later:)
you would need to say 'cfg="2 2 2"' (inside single quotes) to pass them along
yeah... it doesn't respect quites withing a token...
yep - did that and it worked
bpmw_ said his drive is quadrature, but he talks about cw and ccw signals
I think maybe he doesn't understand quadrature
hmmm... you're right... he might be back soon ;-)
stepgen also supports "up/down" outputs (type 1 instead of type 2) so shouldn't be a problem
jmkasunich: Hi John, are you arround ?
that was fast :-)
there was a mail from Ray about the retrofit Fest
I was doing something brainless
[15:28:52] <jmkasunich> http://www.schlockmercenary.com
i design at the moment te breakout boards for the mesa card
btw. do you know the mesa FPGA card ?
I don't think so
[15:30:38] <Imperator_> http://www.mesanet.com/motioncardinfo.html
5I20 FPGA based PCI Anything I/O card
Lawrance and Stephen had one, and we are buying about six for us and some german guys
you had problems with the digital IOs of the mazak, don't you had the optocuppler board for the Vital card ?
the main problem with the Mazak was too many things to do and not enough time
the card is very interesting
the usual problem
we mostly used Opto-22 modules for IO, they work fine
A-L-P-H-A: 200$ for one, if you buy 5 then each is 155$
we did use one of the Vital digital breakout boards, the input section didn't seem to work
didn't bother trying to troubleshoot it, we had other things to deal with.
if we ran out of opto-22 inputs, then we would have to figure out the motenc breakout board
what is the opto-22 module ?? http://www.opto22.com/
[15:40:00] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak009.jpg
the red and white modules at the top of the cabinet
each white module is an opto-isolated input (24VDC) and each red one is an opto-isolated output (12-60VDC, 3A)
how can you connect them with emc ?
with 5V TTL input
5 v interface to dio card
each board holds 24 modules, and connects with a 50 pin ribbon cable
directly connectel with the motenc card
the motenc card supports two cables
(motenc-Lite that is, the motenc-100 supports 4)
and we also had another I/O card that supports 2 ribbon cables (48 IO)
this 50pin connectors with 24 IOs that is more or less a standard, right ???
the mesa card has three
you can see the ribbon cables above Dave's left shoulder, going to the PC mounted on the cabinet door
ah, ok, and the PC is at the far left corner of the door
at the top
the PC is the entire top part of the door
the blue glow top left is LEDs in the power supply fan
hard disk top right
motherboard below the hard disk
real industrial quality :-D
(yeah, the only milling machine with a gamer PC power supply)
din strip below the power supply
the PC is strictly consumer grade stuff
the only expensive part is the motenc card
and the opto-22
yeah.. but I wasn't counting that as part of the PC
expecially if you buy new
hm, hope not to design such a PC for EMC
hi Dave, guess you survived the trip back,
yep...one one 700 mi day ;-)
I had a good trip, but am still fighting off the head cold that I came down with the last day of the show
that was a very tiring week
you worked so hard that you hardly got to see the rest of the show
would hate t o keep that schedule all the time
yeah... it took me until this weekend to get caught up on sleep
I like to work late and sleep late... but the day job doesn't let me do that
got home on Wed and am still dragging
thats understandable... that drive would have anybody dragging
I was looking at flying instead of driving the next time
if I do that again I will fly... and ship stuff
found a $199 fare from Cleveland to Peoria
think it may be cheaper
SW had a $160 rt into Midway in Sept.
could be if you don't have to rent a car, but is that realistic?
well take a look at 6 days of driving with gas and food...
sharing the trip with Matt made it reasonable for me, otherwise not sure I would have gone
we want to use the three connectors of the mesa card like this (one breakoutborad for each connector) :
1. 4x encoder inputs and for each encoder 2x limit- and 1x refswitch input ( optodecuppled )
2. 4x DAC 16-bit output with 8x relay output and 8x optodecuppeld inputs (for the amp ! amp-enable, amp-fault, ....)
3. Digital IO 16 digital outputs (relay) 16 digital inputs (optocuppler). On this port we want to use 16 lines for data, some for adress and some chip selects, so there can be used much more digital IO boards.
I spent serious $$ for that trip...think of it in terms of tooling
yeah... it adds up
motel, gas, food
the $$ hurt, but the time away from my business hurt even more, got to work through the holiday to make a delivery
for me, the time away from my day job was the best part ;-)
steve...that does cut into things
I must say we had a great group to work with
Imperator - does Mesa offer the breakout board, or do you have to make them
I just wish we could have spent more time actually working to get EMC on the machine, rather than simply doing machine re-wiring and mechanical stuff
the lesson learned was a valuable one, just not the one planned
we almost needed a group to do wiring and the turn the emc group loose
steve_stallings: they have some boards, but they use PWM for analoge outputs, i want to have a real DAC chip
yeah... but the re-wiring group also had to do some re-engineering
looking back I think we should have pulled the air change stuff out of the cpu cabinet....leaves lots more room
almost 15" deep
it would have been cool to have a webcam take a photo every five minutes or so, and made a movie
Imperator - ultra precision or low noise? don't forget that most modern servo amps are PWM inside anyway
we really didn't need any more depth
except that we could have tucked the cpu away and mounted the opto where it was just as accessible
not knowing that we were going to pull out the power supply until after the IO panel was made certianly didn't help
I wouldn't want to bury the PC
steve_stallings: 16bit and noise level as good as possible
16 bits is asking a lot
the way I see it, space is measured in square inches, not cubic inches...
layering things makes for horrible serviceibility
certainly in terms of accessibility
sure... like a door ;-)
like the door
ok I give up. :-)
I'm looking at my own setup ... which is a real mess right now
did the Mazak get fitted in the end?
robin: ran out of time
too much time spent "fixing" the machine, rather than retrofitting it
did they get it moving at all?
btw.... I seem to have arrived home without my docs... 8.5 x 11 x 4 box ...probably with a Longbottom Coffee label
not under control
have you contacted Roland?
not yet ...
main problem wrt motion...
amps are velocity mode, expect tach inputs
motors do not have analog tachs
the old control had encoder to analog conversion onboard
which was not clearly documented and not discovered until about Friday
with 20:20 hindsight ... never attempt in public gaze something you aren't sure about :)
things we did that we weren't planning on:
lots of speed bumps
re-wire estop chain (original overtravel limits were NO contacts in parallel, we rewired to NC in series, and added the other ESTOP switches as well
the project machine had old SCR phase angle control servo amps, the almost identical machine sitting beside it had PWM ams, like Ray said no two Mazaks are the same
ain't that the truth
built differential to single-ended converter board for signals from the spindle orientation sensor
(a sensor we didn't think even existed, we thought it had mechanical spindle orientation)
the spindle drive had been changed and it had a really nasty lash-up to attempt spindle orientation
the spindle tach connections had been ripped loose at the motor end and the tach cover was missing
(somebody machined a new one while we were there)
so .. how long got spent on it? 3 days?
6+ for the main crew...longer for a few
long days too.
one of the flex-conduits with limit switch and other wiring was flattened (they set the machine on it when moving it), we cut it back and landed it on a closer cabinet, re-routed all those wires
some good came from the spindle interface converter board, I learned about a neat high common mode range instrumentation amp, thanks JMK
whats the number on the amp?
AD629 or INA117
unity gain, over 100V common mode range
629's are nice
I can already imagine some posts on "other" places ... emc crew coulent get emc working eeven in a week for 8 people, I got Mach2 (or whatever) running in an hour .. an unfair comparison, but I expect we'll see it somewhere
like to see them get it running on a Mazak
this was not an emc problem...it was a machine problem
I think that the Mach folks who were there gained a lot of respect for EMC just seeing what was being attempted
at most 1 day was spent doing "EMC" stuff
the rest was things that would need to be done for any conversion
Im just guessing how the news will filter out ...
an huge amount of time was spend just identifying wires
emc comes up pretty easily once the other stuff .... e-stop, amps, etc are working
we need a webpage for the project that describes just what we did and didn't do
the wiki ....
good a place as any
that's OK...XX brand is welcome to tackle a Mazak in the same condition and see how they do.
trouble is ... publicity and politics aint fair or truthful
still, we know the core
[16:16:51] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak008.jpg
was Art there?
nice pendant arm, eh?
yeah, art was there
well the Mach folks showed up with NO machine to run, not even for demos
nice guy, didn't resemble my mental picture of him at all
loaned them my MicroMill so the crowd could see something other than CRT images
that arm was also built at the fest BTW
jmkasunich: yeah, Ive spoken to him on the phone .. seemd a decent guy
so, you were thinking something like a cross between Bill Gates & Satan? ;)
note the shiny cap on the tach on the lower end of the spindle motor...also the new front panel...all of this takes time
whose the guy in the reverese cap?
no, not at all... just didn't fit my mental image
My machine lost it's fiberglas shround yesterday
dotn know him
good edm and plc guy
and did Ray show?
Ray lived there for almost two weeks
[16:20:20] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak003.jpg
blue shirt, facing away from the camera, on the left (seated)
matt is to his left
the far left, facing camera, is Matt
that machine looks a bit of a beast
then ray, I think the standing person is Jon E, then seated is Tom,
* mshaver is going to look at this picture...
the standing blur is Dave I think
too nervous to stand still
and seated right in striped shirt is Rick of Rick-o-matic I think
[16:22:10] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak001.jpg
your's truly on top of the machine, rewiring Z limit switches
hi dave - yep, that's me sitting at the far left
mazak001 - Dave E at far right
does give you a sense of the size of the machine
pic 1 shows the old front panel
and the head is NOT all the way up!
it is .. without a doubt, a big mother
[16:23:31] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak002.jpg
has about a foot to go
shows the new panel
and the toolchanger
and the Mazak was not the largest CNC there, two Giddings and Lewis take that honor
the big fiberglass cowl/housing in pic 1 came off to get access to the Z limit wiring
will the refit get done?
thats good ...
some of us are planning to return in a couple months
ray to talking about going back in Sep
yes - even if i have to go back there...
gonna replace the SCR amps with modern torque mode amps to solve the velocity feedback issue
yeah, it was hot
jmkasunich: som of thos amps you rescued?
I've already looked at 1525br's and SD's newer amps
no, the ones I have are AC, and not complete amps, just power sections
plenty of space though
I didn't notice the dust...just the humidity
do they already have something running emc?
would make it easier if they did
that G&L would be an interesting conversion
neither of the big Giddings mills is in working order
Rowland had been using Turner Controls, looking toward a future without them
[16:27:42] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak010.jpg
you can see the top of one of the G&Ls in that pic
as well as the servo amps for the Mazak
Z on top, X and Y below
hey john - ray gave me this housing that holds the spindle orientation sensor - i'm supposed to mount an encoder in it's place - will that work with the existing spindle drive? - i remember your explanation of the ramp voltage that causes the spindle to auto-drive itself to the resting equilibrium - just wondered if we can do somewthing similat in software...
hydraulic pumps and such behind the opened door of the amp cabinet
[16:28:46] <robin_sz> http://www.linuxcnc.org/Dropbox/mazak012.jpg
btw- Fanuc does the disconnect differently for those amp and my old ones
what a lot of wires:)
robin: that's only about half of them
tracing out wires was a big part of the battle
have to go downstairs & give a pill to a cat - back shortly
re the sensor, we'll make it work
one only needs about 20 inputs and 14 outputs to do the tool change
mshaver is now known as mshaver_catmeds
all the rest are to keep one out of trouble
matt...if things get too tought....get a catapult
things like "hydraulic pressure OK"
jmkasunich: sure, but all that needs to go in eventually ...
btw...someone needs to tackle the air leak in the hydraulics
some I/O probably won't go in at all... cost/benefit isn't there
one more little problem
the original control had feedbacks for every fuse and breaker in the machine
only need enough I/O to make the machine safe and provide for some troubleshooting
spindle controller needs an Al mounting plate to replace the rusty steel
[16:33:11] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak012.jpg
power cabinet, spindle drive is green
yes about the plate, but IMHO Roland's guys should do that before we arrive
also, that spindle contactor seems to be borked
and the hydraulics
either that or not connected correctly
the bank of 24V->120VAC solid state relays in the upper right corner handle most of the toolchanger outputs
how is that going to be implemented .. in HAL ? or classicladder?
CL most likely
using HAL for I/O mapping
those ssr will handle tool change easily
did paul_c attend?
nope... tho there was one guy (not emc related) who came from australia
most attendees were from the US
no Paul, though John Printice of the UK who did the Mach docs was there
so .. a good CNC holiday was had by all :)
how are hsi eyes?
John was happy with the results of his surgury
it was non-certain he would be OK when I spoke to him
dave_e: I just realized, if we replace the spindle orient sensor with an encoder, that entire +/-15V supply can go away
use PC +5V for all the encoders and be done with it
one more simplification
what servo card you using?
john...how easily will the front panel encoder stuff transfer to emc1
motenc for the Cardinal machine
stg for mine
front panel encoders stuff? you mean jogwheels?
dave: for emc1 you are on your own
good reason to go to emc2
the jogwheel is using the HAL encoder counter
getting the counts into the PC is the easy part (with HAL)
what remains is to write a GUI component that reads the wheel and sends NML jog messages
but then they need to diddle the counters
ray and I talked briefly about it at lunch one day
i wwouldnt imagien the jog wheel to go through the gUI
jog commands come from the user interface
the fact that the interface happens to be using a hardware device (wheel) instead of a keyboard is irrelevant
but shouldn't the jogwheel just update the position counter and let the gui display it
of did I miss something
sounds more like a plan
I was thinking a small window, with buttons for X, Y, Z (to select which axis the wheel moves) and x1, x10, x100, etc (to select how much one click of the wheel move it)
id have thought 3 push buttons was a more intuitive interface
mabe with a smal display
that can (I think) run totaly independent of the main GUI, it sends jog messages, the machine moves, and the main GUI displays where it is
robin: you mean 3 physical buttons, near the wheel, instead of 3 screen buttons?
* robin_sz nods
how about probe sensing?
you could do that, but it would need more I/O
and more mechanical work and wiring
must that be connected to the realtime part or is it possible (useable) also to connect it to the userspacepart like the gui
always more i/o
since we were limited by i/o and manpower, screen seemed a better way to go
i would have thought .. what, 120 channels of I, 60 channels of O would do it ?
looks that sort of size of task to me ...
but it all needs breakout boards and wiring
or some sort of pre-made HMI
what about using a ATMEL that is connected to emc via a serial cable
the motenc-lite gives 32 in, 16 out
the other I/O board gives 48 total, configurable in 4 and 8 bit chunks
I think the plc on my mazak had 55 in and 33 out
I'll count up on my laser .. I think its 240 in, 64 out
some insanely large number for a 2.5 axis machine
we also had a PMDX-122 board connected to the parport, mostly for the charge-pump relay output (since I saw some disturbing power-up results with the motenc outputs)
that gives us about another 12 bits, but 5V only, useless for most stuff
right .. I must be away ...
we did bring the jogwheel in thru that board, and I suppose some front panel buttons could come in there
some of that 55 in can go away when one passes tool numbers thru emc
but 5V I/O in an industrial environment sucks
a crapload of the I/O was used to access the front panel switches
[16:51:30] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak011.jpg
back of the front panel
and a gui takes care of a lot of that
gotta love rotary switches used as encoders 8-)
absolute encoders that is
I'm gong to run...I'm starting to dummy up I/O for the tool changer...
I think that front panel stuff has to go over a bus, otherwhise you won't get happy
the vast majority of the front panel stuff is gone... on screen as part of the GUI
[16:55:46] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak001.jpg
old front panel
[16:55:59] <jmkasunich> http://www.linuxcnc.org/Dropbox/mazak002.jpg
new front panel
front panel "physical" controls:
tool unclamp (for manual toolchanges)
thats a big change
and I think there is gonna be a feedrate override knob
g'day .... I am working on a Hal driver for the Mesa card... is anyone else?
not at present
LawrenceG: if i get my card, then I#m with you
ok... I'll try to get it to a point where it does something and get it into cvs for review
have some practice with the drivers
LawrenceG try to follow the pin naming conventions of the motenc card where practical
yes... it is very close actually....
I started with the motenc driver and have been stripping and adding where necessary
one thing that working on the Mazak brought up... mapping "TRUE" and "FALSE" HAL pin values to actual logic levels on physical pins
the original motenc driver assumed TRUE was high on the 5V logic outputs
maybe it is a good thing not to download every driver to the cvs, otherwhise it looks in a year like the drivers Dir of EMC1
however, when connected to opto-22 output modules or the vital breakout board, ON is LOW on the 5V signal
and opto assumes low is TRUE
imperator: I disagree... I want lots of drivers
ok, but then well organised
as long as they mature to something working....
lawrence: right... and its important that the output remain off until told to turn on
there are several stages...
that was why i said at Fest that mybe we set up a webside for drivers or so
power up, before any driver is loaded
after the driver is loaded, but before the realtime code runs
and finally once the realtime code is running
the power up stage is determined strictly by the hardware design of the card
computer stuff probably needs to be up and stable before control power is energized
is the breakout board watchdog relay usually wired into estop?
we used the estop chain for that
estop chain on the mazak:
starts with +24V
goes thru all the axis overtravel switches (NC contacts)
goes thru all the red estop buttons
to ESTOP IN (one opto-22 input)
and also goes to one side of the relay on the PMDX board
good.... does hal currently drive an output for the PMDX board?
the other side of the NO relay contact goes to the coil of a small contactor and from there to ground
Hi John, could not get AC- Quadrature driver to work. On to other driver which is back to DC step/dir. This is the drive that needs 5us step pulse duration, 10us direction hold time, and 20us turnon/off time. Are these within Emc's parameters?
when EMC sees ESTOP-IN go true, and the user hits F1, it pulls in the relay
bpwm: let me finish a thought, be right with you
if ESTOP-IN goes away, even momentarly, EMC turns off ESTOP-OUT (the relay)
in addition, the PMDX board has a charge pump output that is ANDed with the relay
EMC constantly toggles the charge pump output
that small contactor is the main ESTOP relay, 115V control power goes thru it to operate things like the contactors on the servo amps, the contactor on the spindle, the contactors for cooling and hydraulic pumps, all the solenoid valves for the toolchanger, etc, etc
I need to build a charge pump circuit here, as I am not Using the PMDX board
bpmw: the default configuration for stepgen is 1 "period" of direction setup and hold time, and step pulses 1 "period" long
those can all be extended if needed
"period" is the value of PERIOD in the ini file
Ok, I believe mine is set to 0.000025
so then you have 25uS for all those times
should be good enough
bpwm: earlier you said something about CW and CCW inputs to your drives?
that's not quadrature, that sounds like up/down
That was for the AC drive.
right... but that would explain why the AC drive didn't work
Maybe thats why ?
up/down is stepping type 1, quadrature is type 2
mshaver_catmeds is now known as mshaver
now need to do some honey-do items...
mshaver is now known as mshaver_honeydo
oh, i did start reading emc2.hh with ps, jmk & aj comments - i'll try to write something up tonight or tomorrow night.
John, am I wasting my time on these drives should I just go buy some Gecko's?
no idea... that depends on how valuable your time is... three geckos is something liek $400
if it was me, I'd figure out how to make them work (I don't spend $400 casually)
Or did I buy wrong steppers, they are 4 phase.
I dunno... I can help you with the software, but I don't know anything about your machine, drives, motors, etc
These drives are going to cost me about the same
"going to", or "already did"? if bought and paid for, makes sense to do whatever it takes to make them work
They are on 2 week trial.
got a URL for specs, etc?
or should I be spending my time trying to get my xylotex drive to work better.
I don't know
xylotex and gecko are completely different in terms of power
if your machine is small enough to use xylotex, gecko is overkill
if you need geckos, then xylotex is a waste of time
My mill is 3hp gearhead about 850 lbs.
what size motors? frame size, oz-in, volts, and amps?
Motors I bought for X & Y are 1282oz/in 11v 1.42A
that's very high... they won't go very fast
especially with Xylotex, that is limited to 24-30V
Thats maybe where im having most problem, I'm trying to get them to do 20ipm rapid.
the rule of thumb is 10-20x voltage
have you read the gecko whitepaper about steppers?
Thats why I thought I'd try the ac drive.
what exactly is this "AC drive"
www.nyden.com then checkout IMS20 drive.
the other drive they sent me the dc one is SD230 drive.
the INS20-210 sounds like it would work, but it says it can be configured for STEP/DIR or CW/CCW
why don't you just run it in STEP/DIR and keep things simple?
by the way, the extreme microstepping they are doing is silly
5x, 8x, or maybe 10x is about as far a you should go
The ac dive is only full step i believe
the dc one will do up to 8
the INS20 page on their website says 1,2,2,5,5,8,10.....1000 uicro-steps per full step
ok, full or half step
still configurable for S/D or CW/CCW
Could I try a config for CW and CCW? step/dir didnt work for some reason coulnt get motors to reverse direction.
step dir went one way OK but wouldn't reverse?
did you try swapping step and dir signals?
Didn't try that
Do you want me to?
are you sure the drives themsleves are configured for step/dir?
need more info:
when it doesn't reverse, do you mean jog forward turns one way, and jog reverse turns the same way, or do you mean jog forward turns one way, and jog reverse does nothing?
they both turn the same way
that is exactly what would happen if the drive was configured for cw/ccw and you sent it a step/dir signal...
it's also what would happen if the dir part of step/dir wasn't connected right
if you swap step and dir and get the same results, except it turns the other direciton, then the drive is configured for cw/ccw, and you need to read the drive manual and figure out how to change it
Manual doesn't say. So could I make a .hal file for CW/CCW?
you have a drive where the webpage says "STEP and DIRECTION or CW/CCW pulse (selectable)" but the manual doesn't tell you how to select one or the other?
Not that I can see!
The manual is online if you'd like to check it out
one moment and I'll find the url.
I'm reading it now
pretty crappy manual
what is the model number for the drives you have?
[17:50:54] <bpmw_> http://www.nyden.com/nyden05b/product/IMS20.pdf
yeah, that's the one I'm reading... all six nearly worthless pages of it
The Ac drive is IMS20-210
The dc drive is the SD230
ok... IF I'm reading that cruddy manual right, the IMS20-210 is CW/CCW only, and the IMS20-211 is STEP/DIR
so it looks like you need a UP/DOWN (CW/CCW) hal file
not a problem
no harder than quadrature
quadrature is step type 2, hence 'cfg="2 2 2"'
up/down is step type 1
so change that to cfg="1 1 1"
then you'll have to change the lines that refer to stepgen.0.phase-A to stepgen.0.up
and stepgen.0.phase-B to stepgen.0.down
etc, etc for stepgen.1 and stepgen.2
can the rest stay the same
for now yes... later you might want to rename X_A to X_up, etc
bpmw_: got that quadrature running?
Ok I'll go and try that. It'll take a little while to rewire everything
it's not quadrature... the drives want CW/CCW (aka UP/DOWN) signals
Sorry Alex, John's right
the manual for those drives is pretty crappy
jmkasunich: short questions for you ;)
Ok, I'll be back in a while
do you still remember what the dataflow was?
I mean if the HMI sends a teleop jog command
it eventually gets to motion
bpmw_ is now known as bpmw_[A]
* bpmw_[A] is now away - Reason : Trying new config.
set_teleop_vector is a sort of jog for machines without trivial kinemaitcs
normal jog moves only one motor
btw.. seen my machine work?
doing that on a hexapod probably doesn't do anything usefull
nope, last I saw it was just under construction
jogging one joint could even break the machine ;)
[18:02:45] <alex_joni> http://dsplabs.utt.ro/~juve/emc/mytoy/
anyway, teleop lets you say "move along +X" and it should move all motors to allow you to make that move
but it doesn't ;)
damn... no video viewers on this box
* jmkasunich looks at jpegs
the movies are the nice thing ;)
you can see my emc computer there ;)
guess I gotta fire up the doze box
or apt-get a mpeg viewer
I thought teleop was just the opposite - that you could jog a single joint, regardless of the machine kinematics
I could have sworn that's what Fred said at Fest
SWP_Away: normal jogging does joints
teleop is for carthesian
SWP_Away is now known as SWPadnos
"free" mode is where each motor is controlled individually
surely normal jogging does axes
normal jogging is in free mode
the problem is when motors and axis don't map 1-to-1
right - on a hexapod, for example
but the GUI would still allow you to jog in cartesian space
it uses TELEOP mode to do that
oh - odd.
and the teleop mode code in the motion controller isn't debugged
not so odd.... you need kinematics to jog along X with a non-Cartesean machine
well - as long as the GUI does what I expect, regardless of the machine, I guess it doesn't really matter
(which is called what)
the problem is that cartesean machines let you get lazy
since axis map 1-to-1 with motors, you can write GUIs or controllers that don't make the distinction... then when its time to do a hexapod or robot or something, you have issues to deal with
ahh.. I see ;)
jmkasunich: had a few to twiddle with before I could move my machine
but it works pretty ok now, except the jogging
* jmkasunich is looking for a movie viewer
heh - I'll soon be in the same boat as matt - SWP_honeydo :)
lunchtime coming up here...
for some reason I lost my appetite this morning
I got out of bed, and before I had even had coffee, my wife says "Honey, I've got this fun thing for you to do" ...
Strangely, I don't find fileting a Salmon a "fun thing" before coffee
(or after, for that matter)
alex: video working.. that's too cool!
my wife made me clean som of my crap out of her side of the basement yesterday
we were working on our garage yesterday
worked out well tho.. I've been avoiding that even tho it needed done,
sheetrock, shelving, moving crap around
jmkasunich: like it?
at least I'll have more space for the milling machine once we're done (and maybe even a car or two)
we wound up getting two new shelf units, one for her, one for me
I have to assemble mine today
Gorilla Racks at Costco - just about the best thing going
got a URL? gorilla rack search at cosco.com came up blank
hmmm - I'll look
the racks are 48 x 18 inches, 5 shelves, 6' tall, 1500 pounds/shelf capacity
$65 or so
* alex_joni is packing
I got similar from home depot
48" x 24", 5 shelves, 6' tall
I'm on vacation the next 2 weeks ;)
same thing, but I think the Costco ones are a little stronger
have fun Alex
SWPadnos: surely will do ;)
particleboard shelves, steel frame, knock together with a hammer (no bolts)
yep - www.gorillarack.com
yeah, that kind of stuff
this one has a different cross section for the beams
(instead of simple angle)
yep - Costco has the GRZ 8460-5
jmkasunich, have you seen any of the discussions about the Mesa 5I20 FPGA board?
only a little this morning
ok - it looks pretty darned cool
LawrenceG, and I have one, Martin has been trying to get one
It's the least expensive FPGA card I've seen
see ya - enjoy
SWPadnos is now known as SWP_Away
bpmw_[A] is now known as bpmw_
Well that didn't work! couldnt get motor to turn at all.
hmm.. why not?
It was like both coils were firing at the same time and the motor just vibrated
hmm.. sounds like the sequence is not right
but if it moved before.. it's probably not
I even tried unhooking 1 coil and that didn't even work.
Alex how was the storm the other night?
pretty ok ;)
right.. going for some dinner
and more packing
alex_joni is now known as alex_joni_away
Well I think I'll give the DC drive a try now.
See ya later Alex!
steve_stallings is now known as steves_logging
alex_joni_away is now known as alex_joni
hmm.. quiet today
* ValarQ disappears again
wie geht's ?
gut, danke und selber
Du hast Ferien !!
wo geht's hin ??
zu erst mal campen
hab mein Zelt schon gepackt
na dann hoff ich mal es regnet nicht !!!
ist mir egal
das ist immer gail, Zelten und Regen !! nach nem halben Tag ist nicht nur drausen, sondern auch im Zelt alles voll mit Schlamm und Dreck
du musst aber die boots draussen lassen
dann ist es nicht soooo schlimm ;)
wir hatten mal einen Anh�nger, der ein Zelt enthielt, das war recht gut
mit festen Betten
that's pretty cool
hast Du ein richtiges Zelt oder nur so ein drei Mann teil
anyways.. I'll be about 50 km from my house in wolfsberg away
so if I have enough of camping *g*...
so if it beginns to rain you can easyly go home
or even if I have enough of it, and I need a good bed
I hate camping
it's nice.. once a year
in the summer ;)
near a lake
Im to oald for that
heh.. don't really think so ;)
if the weather is fine that makes fun
oh.. next week I'll be getting that STG
but I won't be here :(
matt sent it a while ago
but shipping to .ro is problematic
would have had problems with customs
do you get the LS7266
so my cousin is bringing it in
I'll be in Stuttgart between 19 and 22
this month ?
cool, hope we can meet
will you be around?
start my new job tomorrow
that is the middle of Schwabenland
about 20 km from Stuttgart
where the "Kehrwoche" was invented
an der fils?
where do you know that ??
"Wir k�nnen alles ausser Hochdeutsch"
[20:12:18] <Imperator_> http://www.wirkoennenalles.de/
tja, ich kann nur hochdeutsch ;)
so.. how's your new job?
any idea yet?
* alex_joni loves technology
I will get a toolmaker
I just VPN'ed into the work network
and I just found out that the hotel I'll be staying at has a wireless hotspot ;)
[20:16:48] <alex_joni> http://www.hansa-stuttgart.de/html/hotspot/index.html
although.. preiswert doesn't sound like free ;)
but not that expensive
and a good location
about 30 km from ebersbach to stuttgart
jep, ant the B10 is nearly a Autobahn, about 20min to Stuttgart
well then.. maybe I'll get to meet you in stuttgart
what are you doing there ?
I'm going to IPA
cool, i was there half a year
which department ?
I think it's Technische Informationsverarbeitung
hm, don't know them
but I don't know the most of them
It's cool there
I'm just going to a meeting
Hi folks, I give up!!!!!
All I can get motor to turn is at feed of 2 with xylotex drive. Mycom drive only works one way- something wrong with drive.
What is the best drive to buy for nema34 motors?
ALL I want to achive is 20 Ipm rapid, I'd be very happy with that!
bpmw_: one more thing
you said once that after doing that stepping type 1
the motor just humms on the place
that also can be wrong timings for the motor
speed too high
or accel too high
if a stepper start losing steps it'll halt
are the motors connected to anything?
Mounted on machine.
right.. so accel too high might make it stall
try lowering accel values
max velocity is set to 0.34 which should gives me 20Ipm.
what's your steps / unit?
32000 for that one 28800 for the other
Xylotex is in 1/8 step mode.
accel i've tried everthing from 1 to 200
make it low
1 should be ok
still won't move?
start with slow G1X1F1
I can jog 110oz/in at feed of 3 on X 4 on z and 1282oz/in on Y only 2
Havent mounted other 1282oz/in on X yet
heh.. found a nice quote
if violence is not getting results, you're probably not using enough.
alex_joni: sounds like my old teacher :)
I dont know do you think a 5lb sledge hammer would help:)
we unix nerds usually call that a LART ;)
usually works on more than just lusers
I did find another nice quote.. but .. *g*
it might offend some ;)
alex_joni: so what? this is irc, we can't hit you :P
Define scene whore. I would say someone who is a scene whore is something who doesn't contribute a damn thing yet befriends all the'right people' in an attempt to stay recognised. Geez I wonder who does that.
if you don't get it = not you I was talking about ;)
maybe i was offended, maybe i'm off to code a cad system now...
hmmm.. there are plenty of cad's ;)
yeah, but nothing like holycrap :D
heh, ein half translation :)
helig = holy
I can even pronounce it correctly.. I think
heilig in german
hmm, i guess it can't be directly translated to swedish...
that doesn't make as much sense as holy crap
heilige scheisse sure sounds ok in german ;)
that's the old spelling ;)
or the new one?
I can't remember.. they keep changing them ;)
dunno, i'm no expert at german
well I am ;)
or should be ..
you speak german?
not really expert
I have a Sprachdiplom Stufe 2
same here, i know enough to read german (with some trouble thought)
you are very good in german, Alex
that's C1 Niveau.. whatever that might be ;)
yur german is much better than my english
alex_joni: my earlier line was a response to �a 'bit'�, not �I have a Sprachdiplom Stufe 2�
ValarQ: Dann m�ssen wir ein wenig �ben, dann wird das schon !!
or was that french quote marks? i keep messing them toghether...
Imperator_: was m�ssen wir? was ist eing �wenig �ben�?
Die deutsche Sprache !
einwenig = a little bit (or so)
can't read that
alex_joni: did you learn anything from the swedish lessons? ;)
Imperator_: ok, i use iso-8859-1
Imperator_: it will be better after the UTF-8 revolution
don't know which one is which, but that seams better :-)
Imperator_: have you heard http://download.slayradio.org/mastering_swedish_-_lesson_3.mp3
Anybody here driving Nema34 Steppers?
ValarQ, Imperator_: after 3 years of kindergarten and 12 years of school in german you might expect I learned some ;)
german that is
Imperator_: that mastering swedish is great ;)
I'm at C1 it seems
yeah, alex speaks flawless swedish now :)
alex_joni: r�ck mig en sill och r�ksm�rg�s imorgon klockan sju
Hm, I think my english is then B1 or B2 level
alex_joni: you have visited a german school in romenia ?
can't be worse than mine thought...
ValarQ: I'm not getting up that early
usually way after 7
10-11 would be more ok
Imperator_: yup, http://www.lenauschule.de/
alex_joni: ok, i'm happy with that :)
but is it imorgon? or should it be i morgon ?
alex_joni: thats short for "i morgon"
ahh... ok ;)
I'm not that far yet
Imperator_: getting this?
"imorgon" is short for "i morgon" in swedish ;)
not sure that makes a lot of sense ;)
my spellchecker accepted imorgon but lexin didn't...
oh, maybe it did... ?...
well.. my french is pretty rusty
i don't even know one french word (i think i know one canadian-french word thought)
oh how so
mais.. francais c'est ne pas tres dificile
tryed to learn french at school but i give up after half a year
hm, alex is pretty good in languages
i'm of to bed... g'dnite folks
Imperator_ : heh ;) not really
bonne nuit ;)
Dobro veche (serbian ;)
hm, have to drive now every day one hour to my new company and one hour back
Imperator, thats not so bad I've done it for years!
Imperator_: it'll be bad when you drive one hour to work, and 2 back
Ah yes the good old winter snowstorm.
hm, do you have to drive so long Alex ?
nah.. I have to walk about 15 minutes
although I don't walk
so I drive about 5-10 mins
something good to know
[21:26:18] <alex_joni> http://www.elite.net/~runner/jennifers/english.htm
I'm also used to walk to school, university, work, the last 20 years
Alex, how much does fuel cost there?
I know diesel
it's 3.1 lei =))
in Germany 2,23 EUR at the moment
3.6 lei ~ 1 EURO
1,23 I hope
Imperator_: not 2,23
what that in usd.
about 1 $ / litre
2,7 $ in germany
wow thats not bad, ours is about the same. I was under the impression that fuel was vey exoensive in europe.
but in the USA that is per galone
Imperator_: can't be
1,7$ in germany
our is in litres
1 EUR=1,20 $
and fuel is 2,23 EUR?
last time I checked it was 1,23
I put some diesel last week in austria and it was 0,954
1,23 EUR is right
we always convert it in the good old DM
robin_sz: Wan shang hao
there it is at the moment 2,40 DM !! when i started driving it was 0,98 DM
alex_joni: Zao chen hao!
morning is still pretty far away
but.. you may call it that too ;)
[00:00] <alex_joni> but.. you may call it that too ;)
* alex_joni is preparing for bed
last night in a nice bed ;)
two sperm swimming along ..
one says to the other:
alex_joni: Wan an !
"Is it far to the falopian tubes?"
the other replies:
"yeah, miles yet .. we only just got past the tonsils"
alex_joni: Zhu ni luu tu yu kuai!
I give up
thank you ;)
robin_sz: two elephants were flying through the air, the one says to the other:
let me fly in the middle
See ya guys, I'm going to research some drives.
* alex_joni goes to bed too
* robin_sz ponders what to do with his no2 laser bed ...
another look at it tonight didn't give any more answers
It has BIG AC servos ...
turn it into a piece of "living artwork sculpture"
do you have BIG AC drives to go with the servos?
the whole thing is all there ... drives, servos, encoders, IO
in the middle is a 6502 based CNC control with no ducumantation
and is a bit "crap"
sounds like a retrofit just waiting to be done
question is .. with what?
EMC of course
well .. .thats what I thought
but anonimasu thinks I might have trouble with the tp
robin_sz: Plug the romset from a Vic-20 into it?
Phydbleep: for all I know, it might be that way already ;)
robin_sz : Look for the RF Adapter output! lol
jmkasunich: this one is fairly slow, 10m/minute max
5in/sec, more or less
are those sort of velocities "doable" with emc?
WOW! lots of info http://www.google.com/search?q=6502
under EMC2, you could conduct some tests without hardware
yeah... the core motion module simply spits out position commands
PID and drives and motors and such is all external
so feed the position command back to the feedback directly to prevent following errors
sounds ... modular :)
then use halscope to look at the outputs
there are also HAL differentiator blocks, so you could generate signals to represent velocity, accel, and even jerk if you want
I might give that a go
if I wasn't buried under other stuff I might do it myself
dont worry ...
trying to clear some of the basement mess
concentrate on the other stuff, I'll handle my end
(IRC is an avoidance tactic)
jmkasunich (isn't it always)
have 40 sq ft of beautiful new shelving
and about 60 sq ft of messy piled with crap floor
so I'm already doomed to failure
uh huh ...
you are going to need what we call "a garbage bin"
yeah, a little of that
but some of this stuff is hard to bin
18" LCD panels
(two of those)
then just tie it to the ceiling ;)
jmkasunich where are you at?
why do you think I have a 1500 foot factory?
half of it is filled with toys :)
jmkasunich: Toss it in the FREE section --> http://cleveland.craigslist.org/
1500 ft? long, wide, or square?
jmkasunich : http://cleveland.craigslist.org/zip/
going to make an effort to get that 2nd laser up and running
looks like I have work for it :)
the only snag is ....
we only have a 100amp 3pahse supply ...
dunno if thats enough to run 2
one way to find out...
yeah, insert finger in ears ... press the ON button
must remember to get hold of some hydrochloric too
there is a very fine gauze filter ... thats all scaled up
I was thinking more of measuring amps draw with unit 1 running, then with unit 2 running
but your way sounds like fun
how many volts is your 100A at?
at the previous factory we had to do a "modification" to the fuse
thats some decent kw
just short of 72KW
going to be close then
laser 1 is 30Kw input
plus a chiller and compressor
laser 2 has a 25kw o/p RF stage ... plus a chiller
it gets "quite warm" in summer :)
looks to be past "close" and well into "not quite"
could be right .. but we have a handy fuse mod
worked before ...
until the wires turn into fuses
I'm not so fond of fuse mods
depends how close you are to cable limits ...
they tend to be bigger than needed
what percent of the time do the lasers spend actually cutting (vs loading/unloading, etc)
but one job has some LONG cuts ..
25% cutting? then just arrange it so that laser 1 cuts while 2 unloads, etc
that could be up near 80% at times
could do that, true ...
I have a job thats possibly going to occupy 40 hours a week of machine time for several months ...
I'd like to put the 2nd machine online to just do that, day in, day out
how much babysitting do you think that job would need?
not a lot once its running
just part changes
how often do you need to load/unload?
once every 45 minutes
3 x 15 minute parts per sheet
end plates .. for motors
what about a second shift?
could be ...
do you have off-peak electric rates over there>
probably do that at first while I bring the 2nd machine on line ...
hey, good point!
yes, we do
* robin_sz hadnt thought of that
1am to 6am
are there other consumables, or is electricty your largest variable cost?
only cost really
lamps are 1$ hr
load/unload labor is probably the highest - even if the guy sits on his duff for 45 mins and works for 5 you still have to pay him
(or course, if "he" is you....)
probably be that way
hey, youd like these motors though
your kinda motor
do the parts need any post processing (debur, cleaning, packing)
a little, not much
get this ..
stator pack is 850mm dia
rotor is 500mm dia
nice sized motors
we cut 120 odd holes inthe stator plate
and all the little "fingers" in the centre
"plate" as in core laminations?
and hold 0.006"
its the end laminations
the rest are punched
from 0.35mm stock
end ones are extra thick?
we cut 3mm and 6mm end plates
but same profile as the laminations?
strange... I would expect eddy current losses to be a problem in the thick ones
most motors I've seen have the whole stack the same thickness
these are axle motors on trains
Hi John, I bit the bullet and just ordered 3 Gecko G210's. Will you be able to help tune them when they arrive?
right... "I don't ask questions, I just make the parts"
oh i ask questions :)
but they have been doing this for 15 years,
bpmw: sure... shouldn't be much tuning required
I figure they would have it sorted now
anyway .. we submitted smaples and they are OK
bpmw: what are you going to use for a power supply with the geckos
get a QC inspection some time this week.
those 11V motors aren't gonna set any speed records no matter what drive you use... but the higher the power supply voltage the better
* robin_sz crosses fingers, legs, arms and other bits
I'm going to try and get a 48v PS from the place i used to work
60-70 volts would be better
Where do you find those?
usually you build them
you really need to read the geckodrive white paper in detail, and peruse other online sources
* robin_sz nods
bpmw_: are they 8 wire?
Oh, I read the whole thing before I orderd drives. most of it is foriegn to me!
Robin steppers are 4 wire.
when I said read it, I meant read it and understand it
it is pretty much written for the layman
bpmw_: are you SURE they are 11V?
strange for 4 wire to be such high voltage
11V, 1.something amps, 1282 oz-in IIRC
where did you get them? removed from some used equipment, bought surplus, bought new?
1280 oz-in seems a LOT for nema 34 ...
There new from Nyden Corp.
I'm guessing they came from something that needed a lot of torque at very low speed
if new, why did you buy such a high voltage motor?
bpmw_: how long is the body?
6" or so
2.5, 5 or 7.5"
6 ... could be 2 or 3 stack
lets assume its triple stack ...
thats the onlyway you'll get 1280 oz-in
thats what? 9Nm?
I hope they weren't too expensive
They have another model lover voltage ,but higher Amps
lower not lover
that is almost always a better choice
well, for geckos at least
bpmw_: model number?
their amps run direct from the 120V line, so the motors might be a good fit for their amps
I bought these ones hoping they would work with 2 amp Xylotex drive, I didnt' expect to have 7 Amps to play with!
xylotex is only 30V max tho... lousy speed with a 11V motor
they seem to build their motors to match their drives ... they will be low torque on a "normal" drive
well, low torque when moving anyway
I tried to get two of there drives to work today to no availl.
bpmw: the hard truth is that those motors will only work will with the Nyden drives
only work well with Nyden drives
* robin_sz nods
Why is that?
they need around 250V DC on the drive side to get decent speed
normally you want the power supply to the drive to be 10 to 25 times the motor rating
normal motors are 2-6 volts, so you use 20-70 volt supplies
actual power available with 70V ... well,, you could only be getting 1/10th the power that a nyden drive will
So I just wasted $450 US on the Gecko's
he already has the Nyden motors and drives
he has Nyden drives?
IMHO, the answer is to get the Nyden drives to work
I'm very confused!!!!
bpmw_: good news, geckos sell for *almost* full price on ebay!
why did you not use the Nyden drives int he first place?
Can't get them to work!
they are new?
They sent me 2 different drives to test.
IMS20-210 and SD230.
WTF is with their website... the pages won't scroll
It's not the best thats for sure!
fricken frames or some such sh!t
I can send these back and they will let me try other ones.
ah-ha... opened the frame content in another window
slow down before you send anything or order anything
IMS20-210 is what you need
did they recommend any particular drive/motor combo?
they do 2 versions
because the SD230 drive looks like a horrible fit for the motors you have
No I bought the steppers beforehand
IMS20-211 is what you need
211 is step and direction
ok, you bought motors from them, and later went back and bought drives from them. Did you ask them what drive they recommend with the motors you have?
210 is weird ass
robin: weird, but should still be usable
yeah, but WEIRD
Ok. But i did try the 210 in stepping mode "1 1 1"
if they will do him the IMS20-211 hes home and dry
this is so frustrating...
if I was there, I'd pull out the scope and meter, and probably have things going without too much trouble
but troubleshooting something like this by remote control sucks
bpmw_: and you appreciate they are opto isolated imputs?
jmkasunich: they look pretty nice drives, decent interface
jmkasunich: whats the 'co' input?
when he was trying S/D signals into them, he got movement the same direction whichever way he jogged
which means that the step signal was successfully driving one of the two optos (either CW or CCW)
Ok. I'll call Gecko in the Morning and try to cancel the order.
so it should be possible to drive the other as well, then use step type 1 which is UP/DOWN
dunno what the CO is, gotta look at the manual again
jmkasunich: yeah, should be
bpmw_: Mariss is a decent guy .. if he aitn shipped, he will
cancle I expect
yeah, sounds like he just needs a different steppign type
Ok, should i try to get the IMS20-211 model?
bpmw: are there intellignet folks on the phone at NYDEN, or just stupid sales drones?
my first choice would be to make what you have work... there is no good reason it shouldn't
bpmw_: wait there ... as jmkasunich says, you might be OK as you are
why spend money on shipping, etc, returning things, plus waste time
I actually talked to singh their engineer about 5 times on Friday
singh... does he speak english?
(sorry, my prejudices are showing)
211 will me more "universal" if you have a choice .. but 210 will work just fine as well .. e ahve stepping types for every occasion :)
need to figure out what the CO input is for, and whether it needs to be tied active, tied inactive, or driven with some signal
Ok, Im game for trying Step mode 1 again.
is it 1?
did you wire them according to the manual, with a +5V supply to CW+ and CCW+, then connect CW- and CCW- to the parport?
robin: he's using emc2 stepgen, 1 is up/down
one sec have to get manual from garage!
jmkasunich: sikhs are usually doctors, engineers, lawyers ... its quite probably mr Singh knew his beans
no doubt there
problem is communication
but the guys in Mumbai do pretty well ... sorta
ok, page 4 of the manual has a wiring diagram
sorry, page 5
does that picture makes sense to you?
I think I wired it wrong last time
do you have a source of +5V?
(preferably the PC)
pin 10 on parport is what i used
can be pulled from the PC power supply connectors (an unused drive connector)
there are no parport pins that supply +5V
* robin_sz agrees with jmkasunich
you need real 5V, not some signal on the pport
it measured 4.9v i thought was close enough.
there may be pins that are at or near 5V when outputing a logic 1, but that does not mean they can supply the current needed
only a few ma worth
5 to 10 at most
you need 40 to 60 ma
5V off a pare drive connector insdie the PC OK?
you have a meter, right?
shouldn't be a problem, and yes
I think the red wire on a drive connector is +5V, but measure it
the black ones are ground
yellow is +12 IIRC
and I think the yellow is +12
red=5V, yellow=12V, black=ground
SWP_Away is now known as SWPadnos
wouldn't want to get 5 and 12 crossed
ok, I can do that, and then wire as per diagram.
then wire that +5V to Cw+ and CCw+
* robin_sz nods
yes... CW- and CCW- go to the parport
phew ... problem solved :)
oh, it will be ...
in addition, you probably need to invert the parport outputs
you can do that in the HAL
see the timing diagram on sheet 3 of the manual
active low outputs
You mean X