Idle musings on mass production

In another topic:

Quote
dmould

Yes, 3D printing is really great and all, but it is definitely not (yet) suitable for any sort of mass production. Apart from the time, there is zero economy of scale and quality control is more problematic. As demand increases, I feel that RepRapPro will have to abandon the idea of printing, even though it might go somewhat against their ideals - you eventually find it a lot better if you use the correct tool for the job!
(#106)

Dave got me thinking; maybe it's been thought of and knocked down already but how about making it a pyramid game? Everyone who buys a printer is given the opportunity to sell back printed parts to RRP ...kind of like getting a deposit back. It strikes me that a significant part of the purchase cost could be discounted in a scheme such as this.

Anyway, probably unworkable but I'm just curious to know what the issues are in mass-producing 3D printers the way RRP do it.

---

RS Components Reprap Ormerod No. 481

I think the problem with that Radian is you'd have to set the price by the weight of the item I think.
I'm sure I have some Bit-Coins at home..... I put them in the Desktop as they were getting heavy....

critical step ..
production delegated / outsourced or....
free production?

just the beginning ..

Open Source Project

Imagine the fun Ian would have troubleshooting crowd-sourced parts I guess that a major issue would be QC'ing the received parts before shipping them out, if they control production it's considerably easier to check I think. There'd also be the issue of telling people that their parts aren't acceptable etc etc.

Ray

Quote
Radian
Dave got me thinking; maybe it's been thought of and knocked down already but how about making it a pyramid game? Everyone who buys a printer is given the opportunity to sell back printed parts to RRP ...kind of like getting a deposit back. It strikes me that a significant part of the purchase cost could be discounted in a scheme such as this.

Anyway, probably unworkable but I'm just curious to know what the issues are in mass-producing 3D printers the way RRP do it.

That's a perfectly good model for a hobbyist or very small scale niche manufacture, but in the commercial World I foresee many issues with quality control and associated complaints/returns from both the producers and the customers, and 3D printing is fast moving out of the hobbyist/niche market. Above a threshold quantity parts supply becomes a viable way to make real money - and that's when shortcuts and scams start creeping in. If a person was being paid a reasonable amount for a kit of printed parts, some people would want to churn out as many kits per day as they could, sacrificing quality for filament cost and printing speed. There is a good reason why commercial manufacturers keep their outsourcing to a select few companies rather than buying from every Tom, Dick and Harry who claims to be able to supply. Would you be happy knowing that the parts in a new washing machine you bought were made by a variety of unknown, untraceable and unqualified people in uncontrolled conditions in various garden sheds? Every part coming in would need to be checked for flaws, measured for size and weighed, and still a high percentage of duff parts would get through. There would be arguments with producers as to why their parts were rejected, and demands for compensation from frustrated customers that would end up in expensive litigation before too long.

Right now RRP appear to be getting away with adopting a "We're all in this together," approach so that we customers are content to do our own repairs and modifications, (though I suspect that there are also people who bought from RS who are uninterested in joining a "community" and simply demanded a refund). That approach won't last for long in the commercial World. Think about a situation where in every flatpack you buy from Ikea, you discover that 10% of the parts don't fit or have a design flaw. How long do you think Ikea would remain in business?

Dave
(#106)

Well it was just a musing. Still, there's a healthy hinterland where hobbyists do business on ebay selling quality built clones of things (e.g. stepsticks). Ebay reputation is one example of a loosely coupled quality control system that might be adequate. Whatever, I think sub-£1000 3D printing will remain a niche business until the big names see it as being a truly commercial proposition with mass market appeal.

On a related theme, reading about the problems with printing ABS with various bits of Ormorod melting in the process made me wonder if someone who has converted the vulnerable parts might "set up shop" making them available to others.

Quote
dmould
Think about a situation where in every flatpack you buy from Ikea, you discover that 10% of the parts don't fit or have a design flaw. How long do you think Ikea would remain in business?

That made me chuckle. I've never bought anything from Ikea but most other self assembly stuff I've owned has had some kind of issue.

---

RS Components Reprap Ormerod No. 481

Hi Guys

I would like to offer an idea..

What about, like me down in South Africa.... R9 000.00 / $900.00 is way more than the regular Joe can spend on anything..
I bought the Kit from RS, purely because with shipping and all the other issues around that 3D Printing was just out of reach..

Now, having my own printer, the plan is to start printing parts for friends and family, or who ever want some... a small cost for the effort...
But if you think you will become rich like that, you have another thing comming...
Some retailers (No Name) here startselling this nice cube printer... cost R20 000.00 / $2000.00, price of a cartrige R800.00 / $80.00
Did not see the next door kids comming home with one this christmas... thats my problem..

3D Printing should be in every home, every country in the world..

I believe we all have a part to play in making this work, as it should...

I will buy from anyone once, if the quality is crap, you will know, and that person will have to pick up his / her game... or stop...
Simple as that..

Madmob

I was going to comment yesterday but didn't get a chance. I must apologize, I often use my tablet to post and things go wongkey typing sometimes!

@dmould
What you talk about is how the chinese changed the world. They don't worry about quality of supply or supply chain audits. I import a lot and high standards and quality control don't make china cheap as a supplier! After all it comes from china, quality what? If I sold you a crappy part from south africa, you would probably say, ah it's from africa, not surprised. When it comes from the UK or the US, where is the quality control? Send it back! My point is there is an "expected norm". 3D printers are not quite at production levels yet, one quality of prints, I must say at 0.10 layer height it looks good and the obvious one is speed. By the way I see HP want to get into the "home 3D printer" market. Sure to be a game changer and set standards, which is good.

Raw material supply is second and this is where it gets interesting. I was in the plastics industry for years. Worked for a company called Plasticwrap Pty Ltd. We did sheet extrusion, bags, film, you name it. Sasol polymers supplied us raw material in the form of 50kg bags of pellets back then, HDPE, PET, ABS or whatever you wanted. SASOL is very big and exports huge amounts in raw pellet form. Now they don't do PLA but every grade of ABS you can think of.

The bottom line is, I started looking at extruding my own ABS filament. I buy ABS filament that probably came from south africa in raw form. Its crazy to think that we export to china in pellet form and then re-import it in filament form with cost of manufacture, shipping, profits and so on. Now there is money to be made and not so much in printing parts for commercial use. Its not because it would be more expensive to produce locally, just no person is doing it. Most of the filament out of china on ebay is produced by some guy in his "shed".

If plan on printing for commercial purposes, like all things, you will build a name over time through quality supply and it is possible for sure. My problem is filament costs and while PLA is great, the commercial world is going to be after ABS parts.

Dieter

#257

A major thing I like about the Ormerod and RepRap in general is that the products are close to the Human Scale. Consequently it should be possible for an individual anywhere in the world with limited resources to be able to make and replicate the product. (You can argue about the details of what minimum resources are required). As soon as you require to sell a large number of units to recover costs you start to become alienated from that. The ability to quickly evolve the machine in different directions also suffers. A plastic part produced by injection moulding doesnt have the same properties as one done on a 3D printer so you will start to get problems as people start to find that printing their own bits doesnt work, because the maufacturer isnt going through the same production process. The manufacturer starts to become alienated from the builder/user.

regards
Andy

---

Ormerod #318
www.zoomworks.org - Free and Open Source Stuff

Quote
dieterzar
I was going to comment yesterday but didn't get a chance. I must apologize, I often use my tablet to post and things go wongkey typing sometimes!

@dmould
What you talk about is how the chinese changed the world. They don't worry about quality of supply or supply chain audits. I import a lot and high standards and quality control don't make china cheap as a supplier! After all it comes from china, quality what? If I sold you a crappy part from south africa, you would probably say, ah it's from africa, not surprised. When it comes from the UK or the US, where is the quality control? Send it back! My point is there is an "expected norm". 3D printers are not quite at production levels yet, one quality of prints, I must say at 0.10 layer height it looks good and the obvious one is speed. By the way I see HP want to get into the "home 3D printer" market. Sure to be a game changer and set standards, which is good.

Raw material supply is second and this is where it gets interesting. I was in the plastics industry for years. Worked for a company called Plasticwrap Pty Ltd. We did sheet extrusion, bags, film, you name it. Sasol polymers supplied us raw material in the form of 50kg bags of pellets back then, HDPE, PET, ABS or whatever you wanted. SASOL is very big and exports huge amounts in raw pellet form. Now they don't do PLA but every grade of ABS you can think of.

The bottom line is, I started looking at extruding my own ABS filament. I buy ABS filament that probably came from south africa in raw form. Its crazy to think that we export to china in pellet form and then re-import it in filament form with cost of manufacture, shipping, profits and so on. Now there is money to be made and not so much in printing parts for commercial use. Its not because it would be more expensive to produce locally, just no person is doing it. Most of the filament out of china on ebay is produced by some guy in his "shed".

If plan on printing for commercial purposes, like all things, you will build a name over time through quality supply and it is possible for sure. My problem is filament costs and while PLA is great, the commercial world is going to be after ABS parts.

Dieter

#257

You might be interested in building one of the many DIY filament extruders - just Google. Here's one for example - [hackaday.com]
Your filament costs could even get down to close to zero if you scavenge your ABS from the local rubbish tip!

On the other point - while expectation of quality is certainly a factor as you say, probably a greater issue is the legal protection afforded to consumers in Europe and N. America for goods bought from a domestic supplier. If I buy something cheap direct from China and find that it's a load of junk, I'll probably chalk it up to experience and bin it. If I buy the same thing from a UK supplier (who may have imported it from China), I'll probably return it for a refund - and the supplier would be very foolish to refuse. That makes it unprofitable for suppliers in Europe/America to resell low quality goods. In addition, if a UK/US supplier sells something that does not comply with local safety standards and it results in an injury, the consequences for the supplier could be pretty serious.

Druk lekker, en mooi ontwikkeling Andy!
Dave
(#106)

Quote
Radian
Well it was just a musing. Still, there's a healthy hinterland where hobbyists do business on ebay selling quality built clones of things (e.g. stepsticks). Ebay reputation is one example of a loosely coupled quality control system that might be adequate. Whatever, I think sub-£1000 3D printing will remain a niche business until the big names see it as being a truly commercial proposition with mass market appeal.

On a related theme, reading about the problems with printing ABS with various bits of Ormorod melting in the process made me wonder if someone who has converted the vulnerable parts might "set up shop" making them available to others.

Note that others have already gone down the crowd printing/distributed manufacturing route, e.g. [reprap.org] . Actually there was some comment on the Ormerod and RRP's design decisions e.g. [forums.reprap.org] , in particular that it seemed to be less self replicating as it used less printed parts and more laser cut parts etc and how it was probably a sign they were moving to more of a classical mass production.

There are also general crowd printing initiatives like [printathing.com] .

Personally while I think the crowd printing including distributed manufacturing is a good thing and hope it continues to flourish, I'm not convinced it's going to overtake classical mass production anytime soon, it will just supplement it and I don't see anything wrong with that. For RRP in particular, while perhaps they should consider having crowd printed parts as an option, I suspect as they continue to grow, moving even further to classical mass production is likely the way forward at least for their mass appeal products.

There will of course always be some tension between the various aspects like 'self-replicating', 'DIY', 'fixable', 'tinkerable', 'mass production', 'cheaply or easily producable in quantity', 'easy to assemble and calibrate', 'sturdy', 'reliable', 'usefulness', 'quality of printed parts'. Some clearly go together better than others. These sort of arguments have been going on since the relative early days of the RepRaps, before Makerbot, RRP etc even existed. Even the switch from through hole microcontrollers and components to surface mount was or more accurately is fairly controversial.

Each company or designer is going to choose whatever they think is best so I expect there will be great diversity in the RepRap community which will be a good thing. It may be that RRP will have one model that is more self-replicating and one that is less so (to some extent they may have this with Mendel vs Ormerod but it's not so clear since the Mendel is also more of an old design).

On the other hand, as much as possible I think RRP should ensure their products are capable of self replicating even if it's sort of a RepStrap rather than RepRap but otherwise I don't care how produce their kits. Or in other words, if they choose to sell kits with completely mass produce parts or even assembled printers, I'm fine with that, provided then ensure as many of these parts of RepRappable as possible. Similar to what Ian said about laser printed parts in the discussion I linked to early. (It goes without saying they should stick with an open hardware design as much as possible. Some stuff like the CPU I don't know when this will be achieved.)

Ideally, they should continue to sell printed parts, whether crowd sourced or manufactured by themselves, as an option for those that prefer it, which will also help to ensure that these parts are printable. In fact some parts which which change a lot may not be mass produced, ever. (And of course, printed parts is likely to be how they do stuff in the 'beta' days, probably with a more defined beta period than they did with the Ormerod.) Ensuring that this works due to the different properties etc may not be easy but IMO should be doable, after all if a printer can either use ABS or PLA printed parts, or even laser cut ones, and many including the Ormerod do there's no reason both a classically mass produced and a printer part should be fine for the same purpose.

Actually I've always felt that this was a good way forward and have never had as much an aversion to mass production of RepRap, call it a RepStrap or RepMasRap if you want. I think this may be particularly important in bringing costs down. I've always felt it's all very well to say RepRaps should be everywhere including in poor communities. And of course there's no point distributing printers to such communities if when it breaks, they can't fix it etc. So you always have to consider replicability. But there's no harm if what you initially supply is partly mass produced to reduce costs (this can't be the only way you bring down costs) since as has been said, costs are always going to be a big issue.

On a slightly different tack, I don't think it's a good idea to always maximise usage of printed parts when it comes at too great an expense of sturdiness, ease of assembly or whatever. Or in other words, I'm fine with increasing vitamin count if there's a good reason for it. Somewhat similar to the way many including RRP have adopted SMT. Of course similar to the SMT/through hole issue, it's great that some will continue down the low vitamin route for those that prefer it despite the limitations it may impose. (There's always the Gada prize to keep people interested.)

I'm actually a bit suprised we're still here with so few RepRaps using clasically mass produced parts. Perhaps I'm just not following things closely enough but I thought there would be much more now, if not by the various RepRap designers, at least Chinese cloners producing RepRap kits. Yet while I know there are a few Chinese cloners (I'm not sure if their kits contain mass produced or printer parts), they don't seem to be that significant and even most major kit manufacturers like RRP still seem to be mostly using printed parts.

BTW, about the larger market and existing companies, well it's an interesting question. It's clearly hotting up and people have been talking about HP and similar for ages (and HP did actually have a deal with Strasys before) and for several years now there's been plenty of people saying 'this' is going to be the year 3D printing takes off. I don't think the patent minefield helps but still every person and their dog seems to want to Kickstart a 3D printer many of these have been fairly successful. And plenty of people have aimed for the cheapest (or at least a very cheap) 3D printer. While not all of these get that much attention in the RepRap forums as many of them don't involve RepRap much (even some of the open source ones which are openly based on the RepRap tend to concentrate on their own forums), and not all of them are even open design, from what I've read both here and elsewhere, many of these don't work very well often with poor support etc.

(Excluding a heated bed does help with costs but does mostly rule out ABS and even PLA is far from ideal, but is probably worth considering for cases when you need very low costs. And considering also the power issue, I expect is going to be the norm for 3D printers distributed to poor and isolated communities. Of course it may seem funny now but IIRC when I first started paying attention to the RepRap community, heated beds were not even the norm.)

The Ormerod is far from perfect but still it seems that it's likely to be one of the decent ones as it develops. There are of course a number of other designs in the sub $1000 range which are quite good and a number of makers other than RRP who have a good name although it's true most of these are more targetted at the hobbyist and tinkerer.

But there's also the $1000-$4000 or so range (these are obvious arbitary ranges) both open design and proprietary, with stuff like large build volumes, enclosed platform, coming assembled etc but also a fair degree of existing calibration and more targetting of the commercial market etc. While these aren't really what I'm interested in there seems to be some quite decent stuff there.

There are of course sub $1000 printers that do some of the earlier things e.g. Solidoodle comes assembled and I think an enclosed platform is also an option. But there's generally a reason why they're at that price. I'm not saying they're bad but I don't think you can say the Ormerod is terribly overpriced coming as a kit and even with the RS markup, the same for other similar good RepRap kits e.g. Prusa, Mendel90, MendelMax. Probably it's because the calibration is the biggest thing you can't easily get from the sub $1000 range.

E.g. buying an assembled printer may seem easy, but it may make things harder when something goes wrong and may mean you're less ready to calibrate it. So you have to consider is spending the extra $200 or whatever the assembly costs worth it or will it be better to get a printer with more features at a similar price?

Similarly unless you use a totally transparent enclosed build platform which likely reduces effectiveness while increasing cost, it makes it harder to see what goes on.

Of course, when it comes to an enclosed platform, the Stratasys patent may be of concern for anyone in or selling to the US. Although with the Alfina lawsuit it seems if you are worried about that, just having a heated bed may be enough. (Then there's also z-axis leveling concerns with 6,629,011, seam concealment although thats software and others I'm not aware.) From what I've seen, patents haven't been that big a concern to hobbyist targetting printers yet (although I seem to recall some have avoided enclosed platforms because of it hence why I brought it up), but as companies producing these expand, they will likely become bigger concerns.

Edited 11 time(s). Last edit at 01/29/2014 01:39PM by Nil Einne.

Hi, Nil Einne - very interesting to read your take on this. It makes me wish I'd gotten into 3D printing from the beginning. I'm usually in on new technologies from the start but in this area I have so much catching-up to do. I think when I first saw Ormorod, it looked like something I could see sitting on the bench without being a massive curiosity like most of the other systems I'd seen up until then. I now realise that the size and look of a printer is going to be another issue to be overcome if the 2D printing companies want to muscle in.

---

RS Components Reprap Ormerod No. 481

Quote
Radian
In another topic:

Everyone who buys a printer is given the opportunity to sell back printed parts to RRP ...kind of like getting a deposit back. It strikes me that a significant part of the purchase cost could be discounted in a scheme such as this.

Anyway, probably unworkable but I'm just curious to know what the issues are in mass-producing 3D printers the way RRP do it.

I have an idea perhaps everyone who buys a printer and helps debug it gets shares in the company, like a cooperative, then it will be in our interest to send in a set of good quality parts (made to company guidelines of course).

This would fit perfectly with Adrian Bowyers (3D Supremo) of RRP's mutualist symbiosis model, in that by helping the evolution of RRP by making parts and improving its products we would be in a sense replicating money in the form of share value, because the shares would increase in value the more successful the company, which we would in turn be directly contributing to, oooh I must sit down and have a cup of tea, oooh but would Adrian hold true to this model now he is in the commercial world.

As far as the way RRP do it (mass production that is) - I have seen a picture of RRP's printer room, looks like they have 10 or so printers, and I think I read they have suppliers, also Adrian describes in a recent article in reprapmagazine having to nip out to feed the printers when friends come round. It is pretty much a given that to supply any kind of volume RRP will require the application of not only robust modern mass production techniques but also adequate funding of growth. It will be interesting to see how this fledgling company fares over the next 3 years. (3 years being the make or break point for new businesses). I think they could survive because the market is growing at about 100% per year at present, that gives a lot of wiggle room to get things wrong and still survive. For a first product I think they have done a reasonable job, and using RS as a way to market was a stroke of genius. Someone made the point that so far they have an understanding clientele willing to work through the flaws with them. Have they unwittingly, or intentionally managed to get the community to debug their product for them? This is a model that software companies use often, they just release early and let the early users report on the bugs. This is a great way to cut development costs and get product to market, I've used it myself. Now that RRP have generated some cash it will make it easier for them to attract funding and grow, I expect great things from them.

Cheers Radian

---

# 500

Quote
rm2014

Have they unwittingly, or intentionally managed to get the community to debug their product for them? This is a model that software companies use often, they just release early and let the early users report on the bugs. This is a great way to cut development costs and get product to market, I've used it myself. Now that RRP have generated some cash it will make it easier for them to attract funding and grow, I expect great things from them.

Cheers Radian

[cough] Clive Sinclair [/cough]

Dave
(#106)