<-- this PDF, page 73 onwards seems to have a good way to do it. unfortunately, blender doesn't run too well on my laptop or I'd try it now. this should probably be compulsory reading for anyone modelling reprappables in blenderby Triffid_Hunter - RepRap Host
looks like a nema23 with a 1600 count per rev encoder on the back.. encoder looks like an avago aeds-9641-P10 or similar, see what's the winding resistance and inductance on that stepper? all up an extremely nice find I thinkby Triffid_Hunter - Mechanics
actually, telephone cables are simply parallel wires, and only certain uncommon types of ethernet cables (STP rather than UTP) have a shield. the twisted pair itself regales most noise to common-mode, while the signal of interest is transmitted differentially, making them fairly easy to differentiate at the receiving end. ethernet cables can carry more power than phone cables simply by stint ofby Triffid_Hunter - Controllers
apparently the easystep only goes to 750mA, which from what I've read doesn't quite cut it. However, a similar module with the A3977 (rated 2.5A) would be perfectby Triffid_Hunter - Controllers
I'm wondering if there's an easy way to print out 2D views of various parts at 1:1 scale, so I can glue the print-out to some wood and cut the part out to get my repstrap on? Have poked around in both AoI and blender but haven't found such an option yet.by Triffid_Hunter - RepRap Host
as far as I know, the main issue with steppers is heat, which is closely related to current. A higher available drive voltage just means you'll have better torque at high speeds. I don't foresee any problems as long as you hook things up properly and make doubly sure that your current limiter is working correctly.by Triffid_Hunter - Controllers
the motherboard<->stepper channel in gen3 doesn't use rj45 which seems odd to me, maybe because 4 twisted pairs don't cut it? However, I'm more interested in the chip with built-in 1/8th stepping and quieter current control than cabling concerns. Zach Smith replied to my enquiry, saying: "Good question. We're still ironing out the last few bugs on that board (gotta tweak the r/c circuitby Triffid_Hunter - Controllers
Has anyone else spotted this at ? I'd personally greatly prefer the smaller connectors for end-stops, and the 1/8th stepping driver chip sounds excellent. I'd also probably replace the motor connector with screw terminals, although a crimp connector might be more mechanically sound. would be nice to see this up on the wiki somewhere tooby Triffid_Hunter - Controllers
only 25mA output current? hm, your mosfets will get very hot at high speeds. Mosfet gates "look like" capacitors, so while it doesn't take any current to keep them on, it does take current to switch them on and off quickly. Ask google about calculating switching times given Qc (total gate charge, should be in datasheet) and gate drive current. power dissipated is voltage * current (P=VI). When aby Triffid_Hunter - Controllers
also check out which seems to be a far cheaper source of suitable motors than oceancontrols. I haven't tried them just yet thoughby Triffid_Hunter - General
1. 3v3 goes to the sd-card socket. eagle will highlight all tracks and pins if you use the 'name' tool 2. the matching KK-156 header plug, although I would put screw terminals in the board instead of that to keep things easy. 3. you may get some interference from the motor drivers, so keep motor wires well away from everything else, and ask google about interference suppression. I'd also suggeby Triffid_Hunter - Controllers
yeah looks good. Since they don't supply Rds(on,Vgs=4.5v,Tj=175C), I'd estimate it at Rds(on,Vgs=10,Tj=175)/Rds(on,Vgs=10)*Rds(on,Vgs=4.5) or 24/15*18.5 = 29.6mR. With Tj-amb=100C per Watt, Tj(max)=175C and ambient of 50C, you get Tdelta=(175 - 50)=125C 125C/(100C per watt)= 1.066W. P=I^2.R, so your max current is sqrt(1.066W / 29.6mR) = about 6 amps. This means, if you put 6 amps through yourby Triffid_Hunter - Controllers
then take into account Rds(on) at Vgs=4.5v, 175C which is usually listed in the datasheet if the Rds(on) is usably low at that voltage.by Triffid_Hunter - Controllers
Has anyone tried fitting a conveyor belt to the wedge's Y-axis instead of the moving table? Looks to me from the videos that it wouldn't be too difficult, and would allow it to drop printed parts into a bucket for very long runs, and/or print very long parts if the conveyor was extended out the front. Are there any flexible materials that the extruded plastic will stick to enough for a solid priby Triffid_Hunter - Mechanics
Wade: put the quadrature encoder on the motor shaft, rather than the output of the gearbox and see how you goby Triffid_Hunter - Controllers
1) easier for battery powered rigs (think solar) 2) much better noise rejection when the regulator is onboard, rather than at the other end of a bunch of wires running close to other wires carrying motor currentsby Triffid_Hunter - Controllers
computer hard drives use a magnetic servo, which resembles a two step motor and is entirely microstepped to access the thousands of tracks on the platter. The shaft will always be drawn to the point where the flux matches the average current, so microstepping should give great improvements in accuracy with light loads, and good improvement with medium ones. Of course, you need feedback for highby Triffid_Hunter - General
Hi, I'm in Toowong. My day job is a firmware engineer with CSIRO, I do electronics in my spare time and all my computers run linux, so I have plenty to lend to the project I've already started collecting parts, namely a bunch of threaded rod and an arduino. Keep me postedby Triffid_Hunter - Australia, Brisbane RepRap User Group
seattlerobotics.org, arduino.cc, a whole lot of other robotics sites discovered through googling various topics. I find that trade journals spend too much time discussing what other people are doing, and not nearly enough discussing how you can do those things for yourself, so not much good for learning.by Triffid_Hunter - General
700 celsius at 24v,3A,300KHz.. certainly sounds like a feasible heating method to me. It would allow us to put an air gap between the barrel and coil so the wires don't need any special heat insulation, and a few extra turns near the bottom could keep the viscosity low all the way to the nozzle. My only concern is attaching the thermistor in such a way that it doesn't get fried by the magnetic fby Triffid_Hunter - Plastic Extruder Working Group
re: roller bearings check out the rolamiteby Triffid_Hunter - Mechanics
Mistakes occur. There's simply no getting around it. The only way to deal effectively with this fact is to build your projects assuming that something's wrong which you can only track down when powered, but without frying anything. This is why putting car tail light bulbs or other series resistances in series with your +12v at power-on is such a common practice - if there's a fault, it lights upby Triffid_Hunter - Controllers
the arduino is an atmega168 on a breakout board. its code should be easily adapted to any atmel you care to useby Triffid_Hunter - RepRap Host
industry practice is that a soft stop is a good idea, but cannot replace a big red normally closed button in an easy to reach place connected in a fail-safe manner. Perhaps have separate buttons for stopping extrusion and lifting the head, and stopping all movement and power.by Triffid_Hunter - RepRap Host
Wow, the TO-220 package with no heatsink at all does 65C/W into ambient air with convection cooling. Even bolting a couple of bits of meccano will drop that to some 20C/W. The recommended heatsink must be tiny to achieve just 35C/W. Does the board layout support getting a bigger heatsink (like an old cpu cooler, usually 1 to 0.2 C/W) and bolting all the hot devices to it? If you want to know hoby Triffid_Hunter - Controllers
Audio guys know lots about ground bounce and related issues. I find the easiest way is star earthing - each ground from each device is wired separately to one point, even if this means you have 5 or more ground wires running side by side. At this point, you have a big power filter, usually a big electrolytic capacitor with a few ceramics in parallel for high frequency response, and a choke fromby Triffid_Hunter - Controllers
Twisting that way causes the wire to loop and coil when you release the tension though.. I much prefer doing it by hand - not twisting, but crossing the wire over and over again so neither wire is twisted, but they coil around each other. This method will leave you with a cable that stays as straight as the wires it's made of.by Triffid_Hunter - Controllers