Bye-bye Unity and Arch

April 16, 2013

I tried, I really did.

For a more than a year I’ve been running Unity on my main PC but can’t say I ever “fell in love” with the interface. It kept getting in the way on the one hand but then would become stubborn and awkward on the other. Things like getting keyboard short-cuts; default applications; window-to-window drag-and-drop and 3D graphics to perform as I wanted them ate up my time. I can see that running Unity on a touch device (with limited screen space) could work well and might even be intuitive but on a desktop PC with a mouse and keyboard just seemed like too much hard work.

So, last week, I took gparted to my hard drive and installed Linux Mint 64-bit Debian Edition with Cinnamon. The install went OK, and once I’d updated all the packages and installed the nVidia drivers, I could get to transferring my home folder backup from the NAS. The only problems I’ve found so far is in using Super-? keyboard shortcuts require a double-tap to activate (this is a known bug: forum thread here ) and compiling the latest pronterface from git (but the eMaker/RepRapPro fork worked fine). Happy days.

That is, till my laptop that I use for work suddenly decided to stop launching a window manager. It was running ArchBang, which I had enjoyed at start but had recently found to be a bit too precious and fineky and easy to break. Considering that I could spend a few days mucking about with pacman to get openbox working again I spent an evening installing Mint on the laptop instead.

The upshot of all this change is that I hope to be able to sync the two machines to have the same settings/applications/files which should mean that switching between the two will be a painless experience… If nothing to do this exists then I’m sure some kind of script to drive aptitude from a Dropbox folder might do the trick.

Advertisements

Sometimes, root isn’t the answer

Although written regarding “rooting” Android phones still relevant in other areas where people insist on doing dangerous things when they don’t really understand the dangers. More often than not: as alluded to in the comments to the article, it ends up causing more “support” issues for the poor geek who gets pressed into servicing the users needs.

Printed out for the office wall…

Hot and Bothered

March 4, 2013

Finally got round to fitting a new heater to the Huxley today, not as easy a task as I had anticipated. I removed the hot-end from the X carriage yesterday only to find that the old resistor was stuck fast in it’s hole. I also had problems un-picking the crimps that I used to join the resistor to it’s wires.

In the end I had to cut the crimp in half; place the nozzle block in a vice and have at the old resistor with a hammer and a dowel. When it came out I could see that the enamel around the wire-wound resistor had worn off and probably shorted out on the block causing the burn-out.

Next challenge was getting the new resistor in, very tight and I hope that I won’t run into the same problem again in another few months. If I do need to replace the resistor again I may look at either rebuilding the whole hot-end or finding an alternative, see below. I did put a new crimp on the wire end only and slid the resistor lead into the crimp end so should be simpler to replace next time, but I do need to check the insulation around the crimp does not slip over time.

After re-fitting the hot-end to it’s carriage and re-levelling the X-Axis I fired up Pronterface and warmed up the resistor, starting from 50°C and going up to 200°C in steps of 50°C. All was well ’till I checked the Z-Axis end-stop! There was a clunk and the Extruder side Z-Axis threaded rod fell off it’s stepper motor.

So, dismantled the clamp; re-set the rod; re-re-level the X-Axis wind up for a test print and hit PRINT… Quarter of way through I notice that I probably don’t have enough PLA loaded. Stop the print; back up the PLA; load a new reel; let the partial print cool; remove from the bed; start again. Finally, I have a working printer again!

This has made me realise that I ought to make a list of things that I’d like to improve. Top five are:

  1. Power supply loom is messy – I’d like to add a 12v supply for the extruder fan; add an ON/OFF switch and shroud the inlet a bit more.
  2. Remove and re-build the bed. I’ve never been happy with it at it’s a pain to level and a pain to clean between prints.
  3. Add an out-of-filament alarm, or, if possible an out-of-filament emergency stop.
  4. Checking out some X-Carriage mounted Extruder as these will cut down on wasted filament. As it is, unless someone can tell me otherwise, once the filament goes past the hobbed bolt in the Extruder then it’s wasted as there’s nothing to push it into the hot-end or pull it back out again.
  5. Redo the hot-end wiring as it’s a pain to replace, as I know from today’s little demonstration! Whilst waiting for my test print to complete I saw a “J” hot-end which looked very neat and compact: might be worth a further look (http://www.youtube.com/watch?v=4mb9BjxOMFw).

Anyway, end of this week I hope to clean and level the bed and apply some of the Kapton tape I got to see what the results are like.

 

Nova

January 13, 2013

As 2013 gets going I notice that’s it been a while since the last post. To be honest, this is partly because I got far too excited about receiving my RepRap Huxley kit; distracted after building it and mostly because I got lazy after a few months of not posting. Now, it’s a new year and time to start again, I think. I’ll endeavour to keep posting at least once a month, even if it’s just to excuse myself for not posting!

The Huxley has been printing well but I’ve been having a lot of problems with the heated aluminium bed collecting stray bits of PLA. which I could not get off in order to keep the bed clean. When I did some asking on IRC and forum hunting the best I could find was lots of advice on using acetone to remove ABS but nothing on what will remove PLA. I suspect that acetone won’t work but I may see if I can cajole the chemist at work to let me borrow some to try. Until then I’ve tried a kitchen scouring pad which works OK but tends to be abrasive and gives everything I print a silver coating of aluminium powder on it’s base: I also got it all over my desk and keyboard the first time as well.

I’m currently trying out a thin sheet of glass which works very well but as it’s too narrow and too long to fit on the bed it’s reduced my print size; is not fixed tightly to the bed; might crack under the heat (not tempered) and looks a bit silly. I also have some Kapton tape on order but Amazon are botching my deliveries so goodness knows when it’ll turn up.

Another problem I did encounter was using “Crystal Clear” PLA, the stuff seems to be so hydrophilic that it became unusable after about four months and kept popping the Bowden tube from the Extruder. I got some luck with drying it out in the oven a few times but in general I think I’ll have to write off the rest of the reel which is a shame as the prints looked awesome.

All this is besides the point as my hot-end stopped working before Christmas, whenever I turn it on the electronics resets. Checking with a multimeter the wire-wound power resistor that does the heating has a short across it but I’m not sure that this is down to the resistor itself being fused or if it’s the lead has worn through the heat-shrink/insulation and is touching the block or each other. I need to investigate a little further but there’s no rush as the replacement resistor (my favourite theory) is out of stock at RepRapPro and Mouser and I can’t find an alternative on FEC/CPC with a small enough diameter to fit in the block. I may email RepRapPro to ask if they have any suggestions. If I need a new resistor that I’ll probably also need to find some more of the high temperature heat-shrink and crimps.

Before I go, this afternoon I heard the sad news that my first boss passed away on Friday after a long illness.  He gave me my first job as a work placement when I was still at University and I’m still there today, 14 years on. Dr. Tom was an enthusiastic supporter of young entrepreneurs and was still touring schools and colleges telling everyone about his successes and failures until quite recently. In fact, I remember him visiting my Tertiary College to give the same talk with the same old slides and overheads many years ago but the truth and inspiration that he brought with him was the same then as it was the last time I saw him. A great loss.

 

Another date

February 13, 2012

Quick post today…

RaspberryPi have announced a date as well (here). With a bit of luck the boards will be available at their shop at the end of this month — with only 10,000 available in this first batch and more than 10,000 RiPi forum members I’m not holding my breath on getting one, but one never knows!

With the RepRap kit shipping this week I should have that up and running by the end of the month — I’ll be documenting the process (unboxing pics, anyone?) here.

I’ve got a shipping date.

February 6, 2012

I’ve received an email telling me that the Rep Rap kit may be shipped on the 15th. There seems to have been a problem with the electronics so it’s being shipped with a Sanguinolou board, rather than the Melzi. Here’s the relevant bit of the email:

We have had some issues with the Melzi electronics and so we will be shipping your kit with the tried and tested Sanguninololu electronics.

We will aim to dispatch your Huxley on the 15th February with RepRap Sanguinololu electronics with micro SD reader (http://reprap.org/wiki/Sanguinololu). Once despatched you will receive a tracking number for your kit.

Melzi and Sanguinololu are functionally identical – the differences are in their shape, and the fact that Melzi has screw connectors for all the wires, whereas the Sanguinololu requires sockets (supplied) to be fitted to the wires.

So, looks like I’ll be crimping Molex connectors after all — which is a pain. I might look into swapping the pin headers on the board for screw terminals but I’m not that sure that it can be done because the two connectors typically have different pitches (0.1″ and 0.25″). I could look at doing a fly-off distribution board with Molex inputs and screw terminal outputs…

Since getting the email I’ve also been reading up a bit on the Sanguinololu and can’t see any refernece to how the SD Card slot is connected. I’m guessing that it’s on the I2C bus but might just as easily be on the second UART. If it’s on the I2C then I might be able to enable the second UART and connect the RPi-B to it, the idea being to then telnet into the RPi-B to send G-Code to the RepRap.

We’ll see when we see — either way the RepRap wiki might get some editing once I get my machine.

Watchia?

January 23, 2012

First post. First blog. What to say!?

 

First of all, let’s just note that I’m waiting for a RepRap kit to arrive (mid-February) for which I’ve been making room in the spare room. I’ll need more room for the construction though so plenty to do before then. The plan is to set up my Mini-ITX PC, which has been idle for years, as a networked serial port for driving the printer.

 

Eventually I hope to get a RaspberryPi-B to do the same job as this will be more efficient… The printer is controlled by a Audrino type processor so a secondary project might be to ditch this in favour of the RaspberryPi-B so it has a network interface; SD card loading an maybe even a little display?

 

First thing’s first, though, is to make room for it all and get the kit.

TTFN.