As most of you know, one way or another I make a crust running light industrial lasers, and pretty much this means stuff made using highest quality chinese components, which are as good any anything except the very best european components, but at a literal fraction of the price.
Just as laser and inkjet printers have “print engines” that are a combination of software in the pc and hardware in the printer itself, much the same sort of thing is true of lasers, where it differs is there is no “Canon print engine” for lasers.
What you get is *probably* going to be RDWorks for any plotter based machine and EZCAD for any galvo based machines, I say probably because it is usually the default option for everything from lightweight hobby lasers up to light industrial stuff.
Forget any ideas about useful help manuals or amazing video tutorials, it’s like Autodesk users, they appear fully formed out of the ether…lol… but training / learning is an issue, it is not really addressed, so you get some people like me who can and do use the stuff commercially ever day who just use keyboard shortcuts and stuff without thinking about it, and others who start clicking menus and wondering what the various options mean and do, and if I alter this will it alter the job or break the laser?
There is a guy on youtube with a channel going by the name of something like sarbarmultimedia, I haven’t checked the spelling, he is a retired guy who allegedly worked in engineering who bought himself a cheap chinese hobby laser, and his channel is dedicated to it, and to the software that came with it, RDWorks.
The channel itself and all the comments therein show that lots of people who did not materialise fully formed and fully skilled out of the ether find the channel incredibly helpful and useful.
At this point we could mutter aphorisms about “the blind, leading the blind” and “in the land of the blind, the one eyed man is king” but it’s distracting, the facts speak for themselves, to lots and lots and lots of people, this guy is an incredibly helpful and informative resource.
My problem is that most of what he says is crap, 95% of this guys problems have sweet fuck all to do with RDWorks, even though he titles his videos the RDWorks learning lab, 95% of this guys problems are based on the fact he bought a bargain basement hobby chinese laser, instead of spending the extra cash and buying a proper light industrial machine.
Yes, you can make changes in software and have observable effects in the finished product, but like JPEG artifacts, you wouldn’t have any of that shit in the first place is you had a proper sensor and electronics package that output RAW images.
For my sins I reached out once when I saw one of his more egregious errors early in his series and pointed out that the problem was mischaracterized, the problem wasn’t the use of a toothed belt, the problem was cheapo lightweight toothed belt + cheapo nasty sticky slides + cheapo nasty steppers + cheapo nasty stepper controllers + cheapo nasty machine XY frame, you can’t “fix” that in software, and you can’t “compensate” for it in software (unidirectional passes, slower speeds and feeds etc) unless you identify it correctly, just blaming the default settings in the software itself is missing the fundamental problem.
Of course I got told I was an elitist asshole and don’t let the door slam my ass on the way out.
Fair enough, no skin off my nose if you all want to follow the one eyed king in the land of the blind.
The ISSUE here is we have a lovely real world concrete example of the problems with a so called knowledge based economy.
If sarbar rebranded his channel and called it something like “tweaking rdworks to compensate for buying a shit chinese laser” with a tagline of various experiments and mods on said laser I’d have a lot less issues with it.
So anyway a few weeks ago someone sent me a link, sarbar has done a video that is basically a copy of a video I did a couple of years ago, or rather it is an attempted copy, me, I just did the video and let it speak for itself, because all I was demonstrating was that 40 watts of beam power on either a 2.5″ or 4″ lens will eventually pierce a 4″ / 100 mm thick block of acrylic, to debunk the idea that you absolutely had to have 200 watts of beam power to do anything “useful”.
So sarbar has done a copy video in a much thinner piece of clear acrylic, it looks about 60 mm, and he doesn’t bother piercing it, and the void is all over the fucking place because his optics and shit and because his beam profile is shit, and then he studiously ignores that and starts speculating about the processes going on when the beam sublimates the acrylic… which you can see in my video, and btw mine were shot at 1080p and because I pay for vimeo you can view them at 1080p and no adverts so everything is clearer… no monetisation and ad revenue there for me…
Now this piece is starting to risk being a “sarbar is an idiot” piece, and it isn’t, the guy is feeling his way around in the dark, experimenting, getting answers however good or bad, and providing lots of people with what they consider to be valuable help, so no, I’m *not* saying sarbar is an idiot, sarbar is doing a damn sight better job than anyone else out there (including me) when it comes to RDWorks video tutorials.
He is doing a good job.
He is doing it himself, and I strongly suspect that he would not want my help or input, even were it offered freely, so I come not to malign sarbar, nor to praise him, but to use him as an example.
There is also a lot to be said for the validity of those who would say I was an elitist asshole, there are those who could afford a proper light industrial laser, they just do not want one, they want a cheap hobby one (parallels here to certain readers and 3d printers) and so what they want is to learn how to get the most out of the cheap hobby one, not how crap their cheap hobby one is compared to a proper light industrial one.
I get that 100% and I am on board with it 100%
My issue is this, when I attempt to do some thing myself on a cheap hobby scale, I try to factor in and identify and separate the issues that are 100% solely down to the fact that I am using some cheap hobby tool or method, as opposed to the issues that would apply to anyone trying to do the same thing, because to me that is usually where all the problems arise and all the quality disappears.
I have seen some *lovely* work produced on 300 buck cheap as you can get chinese “K40” lasers, simply because the person involved accepts all the shortcomings of the hardware and tries to accommodate them and work around them, rather than trying to “fix” them in software.
If you have an old mill or lathe with ACME screws that you have CNC’d up and thrown a copy of Mach3 at (been there, done that, got the tee shirt, many times over) you’re a complete fucking cunt if you try and compensate for the backlash in the system by going into Mach3 and trying to map out the backlash and leadscrew profiles.. trust me, you have either tried and know *exactly* what I mean, or you haven’t, and don’t see why it should be a problem.
Faced with the option of curing it by spending time and money buying ballscrews to replace all the acme shit, I said fuck it, and set everything up to always retract to one side and then machine in one direction only, same as old school turners compensated, and I got/get good work out of 50 year old machinery.
Old / skilled manual machinsts eliminate backlash by this method… the machine is still worn and sloppy, but their skill and working practices mitigate that.
Software cannot compensate for either operator skill, or sane working practices.
This then is my issue with the likes of sarbar and the so called knowledge economy.
There is a majority subset of that economy that DOES NOT WANT to learn the correct factual truth about something, they WANT the approximated shortcut, they are not trying to do a thing RIGHT, they are trying to do a think subjectively BETTER or EASIER or just feel slightly more enlightened or in control.
There is only a small minority that want to learn and do it right, and once a month I will get an email from someone asking me how to do X in RDWorks on such and such a machine, and sometimes my answer is you can’t, not to commercial quality, but usually it is something they did not know, yeah, config, page setting, you can set the adjust distance to anything you like, so if you always want two object offset by *exactly* 11.375 mm set that as your adjust distance, align them with one another, then select one and tap the required arrow key once et voila, yes this included rotational angles etc, use ctrl+arrow key.
But anyway, back the the so called knowledge economy and the failing and weaknesses of same.
Software is a tool, be it ever so funky or crap, you can either sit there and complain that it is a bug ridden piece of crap that crashes for a pastime, or you can learn study and master it, so you will experience only 0.1% of the “problems” that the guy sitting next to you does who is trying to do the same thing with the same software.
So we are at the gap between actual knowledge, and “bodging” and as any skilled tradesman will attest, for 99% of the world the bodge is the way to do it, and skilled tradesmen are seen as being skilled bodgers, not skilled tradesmen.
Which means that the so called knowledge economy is 99% the bodge economy, and 1% the knowledge economy, and like sarbar and I, never the twain shall meet.
So now we have to identify when it matters.
As mentioned before, we have readers here into cheapo desktop 3d printing, it is unabashed and honest bodge economy, and being brutally honest trying to apply the true knowledge economy to it would probably kill it, outside of the subset of the very few trying to push their desktop 3d printer to the limits.
It all starts to fall over badly when you go out and buy a commercial grade product, and it is really a bodge product, and nobody is the wiser…. that shit ends careers and economies and lives and civilisations.
Which brings us neatly to Kobe steel, which has been forging steel quality certificates for 40 years, which will not surprise ANYONE who worked on 1970’s japanese motorcycle frames, and found that the speed of the hacksaw through the steel varies randomly… so we called it shit steel.
We got ignored, and everyone else just pointed at the certificate as proof that it was good steel (ref previous stories here about Sheffield steel and the smelter who used a glass plug from the glassworks down the road to tell when to pour) and nobody gave a fuck, and if there was a problem when we computerize and automate it, why, we will fix that sarbar style, we will NOT go anywhere near addressing the issue that software cannot compensate for operator skill or sane working practices.
Suddenly we have opened the door to a literal whole world of problems, I do not fly any more, if I did, one of the planes I would avoid like the plague is the 787 dreamliner, tesla is allegedly having the same issues with model 3 production, apple is having them with iphone production, grenfell tower had it with cladding, and the beat goes on and on and the rabbit hole goes deeper and deeper.
So, to sum up.
I have nothing whatsoever against sarbar or his followers, because nobody is claiming to be industry standard and commercial quality, it is strictly hobby fucking around at home in their own time, and more power to them one and all, they are having fun.
I’m distinctly uneasy at the notion exemplified by sarbar and 99% of the population, that the differences between what they are doing and what constitutes proper engineering practice and knowledge are not that great, or are at least related, yeah yeah, my 900 quid hobby chinese laser is a distant cousin to the 500,000 quid bystronic… no, it fucking isn’t, it is more closely related to a cheese sandwich…
Kobe steel, to give just ONE example, makes the ENTIRETY of BS5750 and ISO9000/9001 etc utterly redundant and useless, now you either have a product that contains steel that was independently tested and verified and certificated, or you have mystery metal with a certificate that says it is something else.
Seems all those old Sheffield steel makers complaining that they could not compete against the foreigners and still make quality steel were right.
We did not modernise the machinery of industry, and we did not prize the operator skills and working practices of those in the industry, and we needed to do BOTH to stay alive.
Doing one or the other is to fail, only a matter or time, and historically the operator skills and working practices were the first fucking thing to be discarded.
So now it is 2017 and we are facing a robot revolution, the problem remains the same, the person teaching the robot has to teach it the skills and working practices above all else, the people making and buying and installing and programming them are all about how they negate the need for operator skills and working practices, and sick days and maternity leave and strikes and boredom.
It’s not a robot revolution, or an AI revolution, or anything else, it is the last gasps of the so called knowledge economy, as the model has spread so far that it can no longer rely on true knowledge to bridge the gaps in the bodge cheap economy, and things are starting to fall through the gaps.
The glue in those gaps is seen as the problem.
The truth is that everything worthwhile and serviceable ever made by mankind was an alloy, comprising of a mix of basic product material, and operator skills and working practices.
This is why in my own small business, I may never set the world on fire or make a million, but I will never be OUT of work, nor will the diligent and skilled plumber, electrician, welder and fabricator, auto engineer, etc etc etc… we will all always be able to earn a crust at least.
Like this:
Like Loading...