Generation 7 (not the series)

Last week my pair of Generation 7 ver 1.4 PCBs for my RepStrap project arrived from Germany. They are isolation milled by Markus ‘traumflug’ Hitter on a wooden WolfStrap (or similar) using a “cheap V-bit for some 2  Euros” that leaves a amazing quality cut.  The edges of the copper are so smooth under good magnification that it was hard to tell that the board was not etched except for the fact that the cut was deeper than the copper thickness.  Rigid mounting was cited as one of the reasons for the good cut quality.

I dug out parts from my parts boxes and most of the bits were in stock.  I had to buy low profile crystals and the MOSFETs and still need to determine the connectors I will be using.

I hope my LaOS laser brain transplant (just received the new brain) goes well so it will assist with the building of the RepStrap.  I expect a bit of a DIY build made from surplus parts for my first.

Posted in 3D printing, CNC, DIY | Tagged | Leave a comment

LaOS controller, Ahoy!

The Laos kit arrived in good order from The Netherlands.

It took about 2 hours to assemble the bulk of the components. I still need to decide on what I want to do with the connectors, use ones that fit the laser wires or the new connectors in the bag that came with the kit. Some changes will need to be made as the LaOS will support power  control.  Im not sure if my laser can support high speed power control but as most engraving is done with one power level at a time and just on-off control it will be hooked up like the factory controller.

The assembly guide is pretty well done and was easy to follow, no big suprises.  It makes sence to read through it fisrst and determine what options you have planned so you know what parts to leave out if you are not planning to use them.  I put in pretty much everything that will not cause conflict but a bit can be eliminated with fore-thought.

The display board is a simple one with one twist, the switches and the LCD go on the solder side of the board and the holes  are tight so don’t jump in and solder until you have checked the assembly guide online.  Put off soldering the LCD module until you have decided on your mounting arrangement, it is likely you will want to mount it close to the PCB if it is behind a clear panel or a bit proud if mounted in a panel cutout.

Posted in CNC, LaOS, laser | Leave a comment

Laser Power Supply

All 4 of my PSU boards have the laser output as well as the 5 and 24 V controller supply.  It looks like a pretty standard switch-mode setup with a 220V mains input stage.  The model number is JK-KZJ3 and distributed and possibly made by Shandong Jinan Hongyuan Electric, it is mentioned on the net in a few places including a differently labelled one for sale on eBay and on a Italian language site with some reverse engineering details, I only found this site now after I had traced out the circuit myself.  It is rated at 20mA and 25kV and listed as 30W though hints of 40W are also out there though perhaps not quite up to the task.  The switcher IC is one of a range of similar ICs but the only one I could find that has possible pinouts is the SG3525A or compatibles.  It is wired up in a pretty standard way with output drive pulse-by-pulse current feedback and the trim pot and external power setting voltage are used to adjust the switching frequency.  I believe it runs the drive inductor or the output transformer to saturation, and then switches off.  The power is controllerd by how often it does this.  There seems to be no open circuit, short circuit or thermal protection of any kind.

The test button fires the laser in parallel with the controller TTL input, both are opto coupled though the test button is supplied from the onboard supply directly and hence not isolated in any real way.  These controls operates the shut-down input of the IC and so forces a soft start where needed (after some time interval) and might be cause for slow beam on in some case but I’m not sure how slow the ramp up would be.  I don’t think this would work well as a PWM input given the fact that it is controlling the shutdown circuitry and will be liable to work in a non-linear fashion.

Others have indicated that it is better to supply external laser power control to the POT input, on this design, and perhaps others, it feeds into a voltage adaptable input (though a resistor controlled current source charging pin in reality) with a minimal RC filtering that might need more added if using a slow PWM output from some micros.  I think it will work ok with the LaOS boards.

The laser enable input supplies power (24V) to the IC and should be connected through interlocks to prevent any inadvertent beam escape.

The pictures of the supplies for sale do not include the two extra connectors that also appear on the Italian site which shows an identical supply to mine.  The diagrams show the basic interface connections pretty well and I will perhaps re-draw my tracing of the IC circuitry one day and upload it if there is interest.  A Google translation is rather strange to read as it usually does a much better job but I can follow the gist of the posting, he traced the controls because the available documentation was inadequate. These extra connectors allow for connecting the interlock on the 24V line and having the test button on the JST connector while the boards without the extra connectors seem to support the interlock on the JST connector and need an externally wired test circuit and leave the laser switching IC powered all the time.


Posted in CNC, DIY, laser | Leave a comment

Laser controllers

The software is made for a MoshiLaser controller board but my board looks like nothing they have advertised on their web site or that I have seen pictured on the web.  It has 4 H-Bridge ICs (PBL3717A) to drive the 2 steppers and while my laser has the home switches on a separate connector there are holes on the PCB to fit a flex cable socket that is used in some lasers to connect to the limits and X-motor with one flat cable.  The processor part number is ground off but it is a pretty simple micro controller circuit with a bit of external buffer RAM.

From looking at pictures my small laser looks identical to DW40 mini laser engraver on the Jinan Dwin Technology website except it is black where the pictured unit is white.  The bigger laser is very similar to the DW400 unit but without the LCD controller screen and it has an orange paint job.

The larger unit has a more interesting controller, it is surface mount, all the part numbers are missing and it also has no mention on manufacturer on the PCB and sports a V8.1 in one corner.  It has silkscreen legends for DAC-out next to unpopulated PCB so could perhaps be a more upmarket controller family.  It has holes for a 20 way header that might connect to an LCD user interface.  It has holes for a D9 connector that could support a serial host port.  Already soldered on the board is the connector for the pretty common flex cable that merges the connections for the home switches and the X motor though not used on my laser.  It sports 3 out of 5 opto isolators but not sure if more than one is used for laser enable on my setup.  I have not powered the board up yet as the PSU with the missing laser tube had the fuse pulled so will test it before I apply mains when connected to the controller.  However the most interesting find on the controller board was the presence of a SD-card socket on the underside, further the find of an actual (128M Kingston) SD-card in the socket (with lock selector enabled).  I dutifully tried to read it on my regular card reader and it shows up as a totally empty FAT (16 probably) file system.  I will see if there is secret data hidden in raw format on unallocated sectors for fun one day but it is pretty interesting.  It uses dual H-bridge driver ICs in the MultiWatt-15 style with tidy heat-sinks.  The pin-outs correspond pretty well to an L298 with different current sense resistor values for the X and Y drivers and a full complement of 16 protection diodes.  There are a full 16 ICs (including the drivers) and all others are SMD (one is in a THP J-lead socket).  Pads for 3 more SMD and one 8 pin THP ICs are around the DAC-out area.

There are also 5 jumpers on the board with nothing other than a number next to them to hint at function.

If anyone knows more of this board I would be happy to find out details, just for fun.  Again it is not the same as any controllers currently offered on the MoshiLaser site but is supposed to be compatible with the software.

Posted in CNC, DIY, laser | Leave a comment

Moshi Draw

After some hunting the missing software dongle for my Chinese lasers was found.  There was just the one but the same software is supposed to run on both.  The controller cards are very different, one is an old Centronics parallel type interface (that was supplied with a USB to parallel converter) and the other is a USB connected board.

I tried the parallel one first as it has the tube installed.

I downloaded MoshiDraw 9.92 from the MoshiLaser web site.

I searched the net for drivers for the dongle, found none even though the dongle (a SafeNet MicroDog) maker is still around. I searched for a driver for the parallel interface and found none though it looked normal enough.

Throwing paranoia and caution to the winds I decided to install the MoshiDraw software and this went smoothly (Win XP Home) on my old laptop.  The dongle and interface drivers went in automatically and now both devices showed up as happy in the hardware Device Manager.  I ran the software and found my prejudiced expectations of the software were very accurate.  After having investigated Chinese lasers (as an affordable hobby option for myself) I had read many bad and a few satisfied reports of the available software and can testify than in the couple of hours I have spent on it so far it is rather poor software.  I will not say unusable as I expect mostly it is in fact made use of, however the unusual user interface, marginal English translations and strange way of displaying and changing settings it will take time to get used to it.  I will use it for testing and upgrade both lasers to LaOS in time.

On testing the laser which does not have water connected yet I disconnected the laser on line to the PSU so could just play with the motion control to start with.  As we know there is only laser on and off in the software and power is set with the multi-turn pot on the dash.  There is no feedback from the plotter and it looks like there is no easy way to interrupt a job from the software side.  The work area settings seem to be set in the templates (which is ok if you prepare one) so it is possible to drive the laser outside the work envelope without any limit checks if you are sleeping on the job.  Raster engraving of text worked ok and importing of monochrome files is possible, I could only get this to work with the tool button on the left side menu.  My X-axis stepper is not working correctly at rastering speeds and it is not the driver as I tried with swapping the motors around.  As some speeds it gets going and then works but usually it makes tragic sounds and jerks about or rasters for a quarter of the expected travel like it was missing 3 out of 4 steps.  Even with the belt of it continued to do this.  I will locate a suitable similar motor and see how that behaves.

Note that the software comes in left and right versions and there is a 2012 version also (downloaded but a bit fearful of installing with the other one) to accommodate the top-left or top-right home/limit switch position.

The software is made for a MoshiLaser controller board but my board looks like nothing they have advertised on their web site or that I have seen pictured on the web.  It has 4 H-Bridge ICs (PBL3717A) to drive the steppers and while my laser has the limit switches on a separate connector there are holes on the PCB to fit a flex cable socket that is used in some lasers to connect to the limits and X-motor with one flat cable.  The processor part number is ground off but it is a pretty simple micro controller circuit with a bit of external buffer RAM.


Posted in CNC, DIY, laser, software | Leave a comment

First light

I received my pre-owned CO2 laser engravers yesterday.  A pair with mixed prospects.  These are typical of the cheaper Chinese units with no interlocks, external water pumps without flow sensors and a cryptic controller board.

The smaller one with a 20×24 cm working area, a 30W tube and vintage Parallel port controller interface, no software.  The Z height is controlled by a surprisingly crude scissor jack that could never have worked well.  It has a side feed air assist and a digital tube current display. This unit has worked hard but the controller homed the lens on power up and tapping the test button burnt a hole in a sheet of paper with the unfocussed beam.

The second unit is larger with 40 x 50 cm working area, has a fixed T-slot bed with no Z height control and missing the tube, slightly newer controller with USB interface, also no software. This unit seems to have done very little work and is clean.

These units are destined to be put to work and will likely receive the LaOS brain transplants unless I find suitable software that does not require a dongle. I have spare PSU boards so should be able to get the other unit working as well.  The power supplies are only rated at 30W but the larger unit might fit a 40W or 45W tube in the case, so I might get it a bigger tube for it if I can find a slightly bigger power supply that will suit.

Posted in CNC, DIY, laser, Plotters | Leave a comment

Working from Home

The new cottage industry

IT has been said (and will be said many more times) that the 3D printing industry will revolutionise the means of production.  This is perhaps a bit misdirected because 3D printing as compared to regular manufacturing  is a similar comparison to what has happened with direct digital priting in the commercial prinitng sector.

In the printing world there was a very slow progress in the very early days, wood block printing was cumbersome and slow, corrections required a complete new block to be carved (or tricky correction inserts to be made) and the printing was manual.  Then the demand was very limited as literacy was not that common except amongst the ruling classes (preachers, merchants, monarchs and associates).  As the literacy increased and the effort required in printing came down mass production of printed matter became the norm, that famous fellow Gutenberg had a hand in that, this caused the continuous increase in the ‘replication’ rate of the written word until around the 1980′s.  This was a turning point when small independent computers with attached printers appeared and individuals were able to publish for themselves, and not just reply on copied workd from the printing houses, the quality was poor and the cost was high to start with.  How ever here in the 21st century we are seeing the turn around in the cost and quality.  The latest sheet and web fed digital presses are approaching the speed and quality of the last generation of commercial litho presses.  This has now allowed the industry to have variable data on each page making a custom book or an edition of just one possible for no extra cost.  This should drive down the cost of books as warehousing should be reduced (and it will be reduced where the finishing is also automated) but special cover styles will have to be batch processed even in the print fun can be made smaller.  The benefit of flexible publishing that was in the hands of the individuals has now been handed back to the industry.  The ramifications of this are yet to be seen but it will become harder to justify ownership of printing means when an email can print and bind whatever one chooses.

With the 3D printing model similar things have happened but much later.  Products were initially hand made by craftsmen, then by skilled workers in teams.  Still later the work was simplified into manual batch production by semi-skilled workers.  Then came the industrial revolution and machinery started to replace the human element and production volumes started to increase allowing for specialised machinery to manufacture articles at much lower costs for mass marketing, this was the product of capitalism, those who could afford to build a factory and keep it going were able to ‘print money’ just by having it running.  The side effect here that the labour movement struggles with is that workers become impersonal and are merely parts in the greater production machine and the capital does not need to have any personal investment in their lives.  All of a sudden computers came onto the scene and allowed for customised and automated production of variable parts, parts with differences and parts in small quantities, this made it harder even for the craftsmen to compete and they struggle to occupy the diminishing niches that cannot be fulfilled by automation.  A further blow is felt when the additive manufacturing technologies start to emerge, companies can now have models, samples and patterns made that previously could only be done by craftsmen.  The pool of skills in design is shrinking as less designers are required because they no longer have to craft their design and move onto the next job while a machine completes the fabrication.  Now comes the current tide of individual 3D printing, it started a little earlier with the DIY CNC followers who would use subtractive automation at home or in small shops, it has kept the competition alive in that field.  This is analogous to the desk top printer in the history of printing.  It has given the individual designer back some of the power that has been missing for so long, he is able to produce to his needs without having to have the large amounts of capital to finance a factory. He can now with growing confidence manufacture parts that are usable as well as automate the model building of his designs and become competitive again.

That was the good news.  Look forward a decade or two and we will be seeing the trend swing the other way.  The capital sponsored industry will keep developing at a rate always ahead of the independant users and with more marketand volume will be able to stay in front technologically.  The turning point will occur when it is cheaper to make a part in any volume than it is to mould the part using expensive moulds.  At this point the industry will have got to the point of the full digital printing press we are at today where there is no quantity barrier for those that can afford the large expensive machinery.  Competition would be high and costs would fall and the power in individual hands will be eroded again.

However there is a silver lining here. In the printing industry where the individual is trapped into desktop printers that will never approach the economy of commercial presses and is no longer unique in beeing able to print variable data the 3D printer user has an edge.

He can make his own 3D printer, while the author cannot make a better printer.  He can improve his equipment, follow the latest trends, and even logarithmicly scale up production to outstrip the large capital industry if hehas a good product.  This is the legacy that the RepRap idea has given to the individual, it gives the opotunity to be in control of the source of production like never before since the first automated machine was purchased.

So if you want to be part of the first true global revolution in individual independance you owe it to yourself to gain access to a 3D printer, you can always make parts for your own printer before it is too late.

The above thoughs are simplistic on many levels and gloss over a lot of big hurdles that need to be overcome but they do indicate the great reliance we have on centralised mass production and how we can limit this.

You too can join the 3D printing revolution.

Posted in 3D printing, DIY | Tagged | Leave a comment

The new Steam engine

Here is a quote by Jim Pinto

Nobody could have predicted the impact of the printing press in 1450, the steam engine in 1750, or the transistor in 1950. It’s impossible to forecast the long-term impact of 3D printing. But the technology is coming, and it is likely to cause a significant disruption. This will be a major inflection point of progress.

We live in changing times, things are moving forward at ever faster rates, quantum computing may be just over the horizon.  3D Printing is here to stay.  The fact that a hobby/kit style 3D printer costs less than the early laser printers is certainly a trend indicator.  You can now buy a laser printer for under ZAR700 (less than US$100)  and it is much better than the early 300 dpi giants.  The same will happen to the 3D printer but because it can be done at home it will spread from there.

The one local importer of budget 3D printers mentionedon a forum that they had shipped around 150 of their budget printers, more by now for sure, and they retail for about ZAR11000 (US$1400)  or there about.  The DIY kits like the RepRap can already be had for half that from a number of suppliers and the price is falling.  Collecting all the parts yourself should bring it all in under US$500 and the Printrbot project has gained pledges (they received over 30 times the project amount) for around 800 kits world wide (some local too I hope) for about US$440 on the Kickstarter special with a projected retail price of US$499 estimated.

This is the cost of a medium laptop these days. and it can make copies of itself (well partly, but most of the tricky bits at any rate) so it can earn its keep.

For those looking to roll your own a neat place to get bits is old flat bed plotters (Roland DXY series) and some of the older 9-pin impact printers.  These are sometimes still found lying around in office and factory corners, in company storerooms taking up space and sometimes advertised on Junk Mail or Gum Tree.

The circuitry to run the printers is being simplified all the time and most of it can even be made at home if desired.

Keep plotting


Posted in 3D printing, CNC, Fun stuff, Plotters | Leave a comment

An idle pulley

Well, not even an idler pulley, but an idle moment to reflect on some of the new jargon that 3D printing will add to our language.

An anagram for RepRap is “rapper“, perhaps they can be used to print out bling one day

Darwin is “rad Win” though linux is often used.

Mendel can be “LED men” though gamers are more fond of pimping their boxes.

With the advent of Prusa Mendel we now get the “resumed plan” to make things easier.

RepStrap folds to “star prep” for printing in zero gravity, already tried on the vomit comet.

Filament extruder hides some advice when “unflexed, trim later

P.S. The longest words in english that have no duplicate letters are “dermatoglyphics” and “uncopyrightable“. Now how neat is that, one is the study of fingerprints and the other says they are in the public domain.

Thats all for now, plot on.

Posted in Fun stuff | Leave a comment

The purpose of a Wiki

A Wiki can be many things, one of the more famous is Wikipedia, the free to use, free to edit (within guidelines) and free to reuse Encyclopedia.  My reason for installing MediaWiki software was to make use of the knowledge management and note taking capabilities.  I have a large store of collected knowledge and information that I want to consolidate and hope to do it in a manner that lets me get to the data and hopefully will give others some benefit as well.

These blog posts are going to be more in the way of passing thoughts, some will be very topical, others not so much.  The consolidated data will be in the wiki pages and should keep growing with time as I have something I want to add.  My plan is to transcribe all my notes and collected clippings and web links and add them in so that I can find the information should I need it at a later stage.  Some of the information is of use only to researchers and historians as it will relate to things from a bygone era that does not mesh neatly with the modern high speed world.

I selected MediaWiki as my wiki engine even though it was not available as a single click install on my web hosting service.  I actually had to download the file using linux commands to fetch the compresses arcchive file, unpack it, move it into the correct place nad fiddle with a couple of permission bits.  Using the cPanel console on the selected domain I then created the required (shared with a prefix) MySQL database and user so that the MediaWiki configuration script could to the resto of the work.  All told it took an hour or two including the reading and pondering and anguishing about having pretty URLs when done.  Using a shared hosting service is very convenient but has a couple of minor limitations when it comes to setting wanting control of the global httpd.conf file for the Apache web server.

All the above to explain that there was a burden for me in selecting the particular engine, the reason that I chose it was that Wikipedia uses the same (slightly modified) engine and as a small time editor on Wikipedia I was familiar with the basic editing techniques.  I will still have to study the permissions, moderation and other advanced features.


Posted in Hosting | Leave a comment