Crowd-sourced PCBs

[firebug24k]

Greetings all,

I came up with a crazy idea a couple of weeks ago about crowd-sourcing the cost for PCB prototypes.  Basically, I had the idea to subsidize the cost of PCB manufacture by giving away free PCBs to the author, and then selling off a few copies of the design to cover giving away more PCBs.  Basically you just order a 5x5cm or 10x10cm PCB from the site, email off the Eagle files, and in a day or so after confirming the PCB as valid, it gets sent off to get made, you get 2-3 free copies, and the extra copies go up on the site for other people to buy (which covers your free boards).  I've got a deal with one of the board houses in China to make them, as I buy quite a few boards already for another one of my side projects. 

There's a flat $5.50 shipping fee to cover materials, but other than that it's no-kidding-free.  Right now US addresses only to keep shipping costs reasonable, but if there's a lot of outside-the-USA demand I can change that.

I figured I'd give it a shot for a month or so and see if it can become self-sustaining - if sales of boards equal out free boards mailed, then I'll keep the site up - obviously it's a no-go if there aren't any sales to cover the costs of manufacture!  But I know I like building random projects and browsing a site that had a bunch of circuits boards that are already ready-to-go (along with schematics and BOM) would interest me, so I figured I'd take a stab at putting this together.

Anyway, the site isn't live quite yet, but wanted to get feedback from the community on whether or not this has any chance of success.  Thoughts?

[c4757p]

Basically you just order a 5x5cm or 10x10cm PCB from the site, email off the Eagle files

Please support something in addition to Eagle. Gerber, KiCad, gEDA at least.

[firebug24k]

That's a good point.  I can support anything I can generate PCBs from, actually, or at least, programs that don't cost a crazy amount of money.

But what I want to encourage is making the schematics available too, not people just giving me raw gerber files.  I hate it when someone posts a cool project, puts up a schematic JPEG and some Gerber files, and then I want to change one or two tiny things and can't without redoing the entire layout from scratch.

So short answer... good idea!   I'll update the page to accept more formats, with the stipulation that the editable schematic has to go with it.

[c4757p]

I think the idea of using Eagle for something like this is quite contrary to the spirit of making full schematics and layout files available. It's not quite like using Altium for it (*cough*Dave*cough*), but still... The paid version of Eagle is a usable EDA program (I really can't stand it, but that's just personal preference), but the free version is a turd. The limitations imposed are just stupid.

I hate it when someone posts a cool project, puts up a schematic JPEG and some Gerber files, and then I want to change one or two tiny things and can't without redoing the entire layout from scratch.

Agreed.

Do you think you'll be able to make the cost worth it? I can get some damn cheap 50x50/100x100mm boards from iTead.

[firebug24k]

As to making it self-sustainable on cost... I'm not sure   I'm willing to put a few hundred bucks into it to see if I can get it rolling.  There just isn't an open hardware repository type of place that I can find on the web - some place you can go browse designs and get a board or two to build on the cheap.

Definitely not going to make money on the free ones, but hoping that enough people will visit it as a "let's see what new cool stuff is here this week I could build" kind of store.

I think the advantage of buying from a site like this (for a PCB you didn't design yourself) is that there would be a small inventory in stock, which could ship out in 24 hours, rather than 2-3 week lead time waiting on a batch back from China.

[mikeselectricstuff]

Don't seeed or itead already do something like this ? Not free but discounted for OSHW designs or something.

The problem I see is how many other people would want to buy your prototype PCB - at the prototype stage it probably has bugs and minimal documentation so probably not very useful to anyone else. 

[FrankBuss]

Don't seeed or itead already do something like this ? Not free but discounted for OSHW designs or something.

Itead has something similar: you can "open source" your PCB and you'll get 2 random additional PCBs, from people who open sourced their PCBs, too. Not useful, but I like it.

The problem I see is how many other people would want to buy your prototype PCB - at the prototype stage it probably has bugs and minimal documentation so probably not very useful to anyone else.

Someone could create another PCB wall with it

[8086]

What you will end up doing is paying for a few customers' free PCBs with your few hundred bucks, and then realise you need to close because you're not making money back.

I'm afraid that's the reality. 

[free_electron]

- email off the Eagle files
- you get 2-3 free copies, and the extra copies go up on the site for other people to buy (which covers your free boards). 
- projects and browsing a site that had a bunch of circuits boards that are already ready-to-go (along with schematics and BOM) would interest me,

<harsh semi-sarcistic semi-joke mode on>

1) EAGLe only? RACIST ! why only the trailer-trash of cad programs ?  Gerber RS274X is what you need to handle. Besides , i dont want to give you my source cad file.
2) who says anyone would buy my boards or could be interested in what they are for ? i may be making a on-of-a kind adapter to probe in a harddisk or to debug a chip. totally useless for anyone else. but hey, if you want to give me free boards.. i'l play along
3) schematics ? not in your lifetime. design your own stuff. i'm not giving anything away for free. i just want free boards so I can sell them for lots of money ...

</>

<serious mode>
see what is wrong with your approach ? it's set up for failure.

1) the world does not revolve around 1 cad tool.  the lowest common denominator is Gerber data and ncdrill. And there's your problem. every tool has its quircks . layer misalignment , drill to track misalignment , plot sizes , origins , filenames ... and you'll have to clean it up unless you have some really good gerber tools to panelize (frontline genesis). it'll end up being a drag... you'll either have to do it or be swamped in emails from beginners asking 'how do i create gerber ? what is ncdrill ? my tools does not have gerber , its a free one from pcbexpress... can i send in BMP or JPG ? dealing with source files is also problematic. you'll have to be proficient and have all the cad tools out there...

2) is your biggest problem. if all you are after is some arduino shields it will work. The other stuff ? Besides , you are hoping the guy will release the schematics and documentation. The guy ordering the boards is NOT interested and has no time to deal with other people that want to build the same thing. Even if i were to release the schematic to one of my designs  there may be firmware involved .. and then the crap begins. What to release ? Binary ? I'll have to front all the questions 'i want this modification , can i have the source , why did you use compiler xyz that costs money .. i only have the free great-crap-compiler. why did you write it in PL1 and not in C ? why do you use AVr, i want PIC ? Why is it full of 0402 parts and tqfp 's. Why does the cpu need to be in BGA ? i can't solder this.. yadda yadda yadda waa whaa whaa.

me as board originiator am not prepared to deal with all that stuff. It's my design i dictate the rules. not the guy producing the boards (you) nor the guy buying the pcb from the guy producing. It's not a democracy. it's a dictatorship. If am so good i may tell you : Here is schematic, here is binary, buy board there.

But it doesnt stop there. Then it's going to be: can i substitute the LT1006 opamp that cost 12 dollar with an lm741 with 2 missing pins ? Do you really need 0.1% resistors ? Why does the board only fit this enclosure that i can't buy in my country ... 

You are opening a potential can of worms that the guy ordering the board is NOT prepared nor willing to deal with.  If the designer is unresponsive guess who will get his forum overloaded with angry 'buyers' posting stuff like 'this site is crap , they sell boards but there is no support' yada yadda ( you know how people are...)

And the last problem. i built it to the letter( i think) but it doesnt work .please help ... deafening silence....

3) schematics? not in your lifetime ! this is a prototype board for a device i am hoping to throw on kickstarter when done so it can become the next big thing and i can become a bajillionaire. thanks for the free boards..
And if i were to release schematics and pcb layout i am still not willing to deal with number 2 above ...

So you are essentially setting up a pooling service for a very small niche market.. the eagle user who want to do open source open hardware and is willing to support other people who want to buy his design...

and now i'm going to be really harsh.. the quality of work that comes out of that pool of 'designers' is the stuff pigs like to roll around in...
So you still will have another problem : fielding all the questions . why can't i do this  or that? why did my trace burn up ? this is my first board please help. what are file extensions ? i used a crcked version of eagle 1.0 will it work ? and the entire shitstorm of other noob questions.

That's all fine if you are really into that but you have to be a friggin masochist to take that on.

Itead and Seed have done it and it is successfull for them. Why ? becasue labor over there costs nothing. if you are to set this up over here : you will be swamped in labor. the little money profit you will tack on to the pcb's + the cost of the 2 or three you give away will blow the price waay above itead or seeed.. and poof goes the idea.. you'll do all the work and sit on apile of boards thay may sell one or two a month.

here's another thing you haven't covered  : you give me 2 or 3.. how many extra will you make ? 5 ? 10 ? 100 ? before you know it you will have a warehouse full of batches of 97 boards that nobody wants... ( you ordered 100 , gve away 3 and the other 97 sit there ).
by the time anyone buys them they are so corroded that they are unsolderable...

oh, and what about all the other stuff.
i wantyellow soldermask with black silkscreen.. and enig finish.. someone else want green silkscreen on purple soldermask... you'll have to collect a massive amount of orders to be able to pool successfully... and itead and seed already have the ball rolling.

in other words : it's going to be very hard... and the return is going to be zilch.

itead is successful becasue the board originator pays for the entire board cost. they retain 2 and give you 2 random from their open pool. it costs them nothing.
you willfail becasue you will front the entire board coast, give the two free upfront and hope someone else will buy the remainder...

[c4757p]

1) EAGLe ... the trailer-trash of cad programs

I must remember this one.

[Corporate666]

- email off the Eagle files
- you get 2-3 free copies, and the extra copies go up on the site for other people to buy (which covers your free boards). 
- projects and browsing a site that had a bunch of circuits boards that are already ready-to-go (along with schematics and BOM) would interest me,

<harsh semi-sarcistic semi-joke mode on>

in other words : it's going to be very hard... and the return is going to be zilch.

itead is successful becasue the board originator pays for the entire board cost. they retain 2 and give you 2 random from their open pool. it costs them nothing.
you willfail becasue you will front the entire board coast, give the two free upfront and hope someone else will buy the remainder...

What a *fantastic* post, Free Electron!

The first thing I thought when I read about this idea was "10 boards from iTead costs $25... so to save $25, you want me to give you my proprietary designs, schematics, source files, BOM and possibly firmware - that would have taken me many hours to create".  So to save $25, I get paid maybe $0.25 per hour.... and I also sign away the rights to my design to you and allow you to sell it for whatever price you want?


There is nothing in it for me as a designer. 

There is nothing in it for customers (buying non-working PCB's or obscure products that are meaningless to them, or have no documentation)

There is nothing in it for the site owner (spend money on PCB's,then give them free to the only person who it's actually worth something to).


Sorry to sound harsh, but this is a bad idea all around.  It will never work as a business and if tried, will be a giant money sink because nobody will use it other than people with designs so useless to anyone that they don't care about giving away the source, and people so cheap that they can't or won't spend $25 on their design.

[benemorius]

The comments about refusing to provide CAD source files or schematics are fairly amusing, but I don't understand where they're coming from. It should have been immediately obvious that the service in question would be only for use by those who aren't prevented for some reason from putting their work in the public domain. If you can't release the source, then you can't use the service.

The main problem that I see with this has already come up, and it looks like a big problem. Most of the designs that are ordered from such a service seem likely to be of little or no value at all to anyone but the designer. This is because, although the cost of getting a batch of 10 boards from china is now rather cheap per board, the cost of getting just two or three prototypes made is still comparatively high. This means I would probably use this service to order rev. 1 of my new design for prototyping, and that board design may well prove useless. However, I will most probably not use this service when I've finished rev. 2 and am ready to order a larger batch. This means that you'll only ever see my broken rev. 1 design.

Granted, not all cases will follow that pattern, but I'd be concerned that too many might.

[Corporate666]

The comments about refusing to provide CAD source files or schematics are fairly amusing, but I don't understand where they're coming from. It should have been immediately obvious that the service in question would be only for use by those who aren't prevented for some reason from putting their work in the public domain. If you can't release the source, then you can't use the service.
 

It's not hard to understand.

There are not simply 2 kinds of project, commercial where you are not 'allowed' to disclose your source and personal where you are.  Some people have a design they are "allowed" to disclose, but do not want others buying it for a low price - as in FE's example of a Kickstarter project. Also, just because someone is doing something for themselves and not commercially does not mean they are OK with someone else selling their design and work for a profit. 

I'm also surprised at people talking about "the expense of a PCB".  Are people really so cheap?  It does not matter at all what the cost per PCB is, unless it is a commercial design where you have a unit price to worry about.  All that matters is the total cost to get the 1 or 2 boards you need.  $9.90 from iTead for a small board, or $25 for a larger one is comparatively expensive?  Compared to what?  The cheapest shipping for a single resistor from one of the distributors is going to be something like $6 or $8.  And it is very hard to do any real work in electronics without at least a couple of thousand dollars worth of equipment, hardware and software.  I can't imagine there is anyone who has a design they want to implement and doesn't mind paying the single unit prices for chips, passives, switches, etc...but who then says "my god!  $10 for a PCB?  They must be crazy!  I can't afford that!" 

[benemorius]

It's not hard to understand.

There are not simply 2 kinds of project, commercial where you are not 'allowed' to disclose your source and personal where you are.  Some people have a design they are "allowed" to disclose, but do not want others buying it for a low price - as in FE's example of a Kickstarter project. Also, just because someone is doing something for themselves and not commercially does not mean they are OK with someone else selling their design and work for a profit. 

It may be that I am underestimating the disparity which exists between the hobbyist world and the professional world. We appear to be facing this from different perspectives with several significant differences.

You're certainly right - there are still a number of personal designs that need to remain private for one reason or another which are excluded from a service like this. The commercial designs you mention, which are also clearly excluded, don't even enter the picture to begin with. What may not be as obvious as I'd have thought is that the number of designs remaining after these exclusions is both considerable and growing. Just because from where you sit you can see a lot of designs which don't apply doesn't mean that there aren't also a lot of designs which do. Most of my designs have no need for privacy, and I would surely welcome the opportunity to pay for their manufacture with the source files rather than money. For everything else, well, there are obviously a lot of other places to get boards made.

I'm also surprised at people talking about "the expense of a PCB".  Are people really so cheap?  It does not matter at all what the cost per PCB is, unless it is a commercial design where you have a unit price to worry about.  All that matters is the total cost to get the 1 or 2 boards you need.  $9.90 from iTead for a small board, or $25 for a larger one is comparatively expensive?  Compared to what?  The cheapest shipping for a single resistor from one of the distributors is going to be something like $6 or $8.  And it is very hard to do any real work in electronics without at least a couple of thousand dollars worth of equipment, hardware and software.  I can't imagine there is anyone who has a design they want to implement and doesn't mind paying the single unit prices for chips, passives, switches, etc...but who then says "my god!  $10 for a PCB?  They must be crazy!  I can't afford that!" 

The disparity between worlds really stands out here. From my perspective, it does not matter at all what the cost per PCB is unless it isn't a commercial design where the cost is covered by a customer. If I'm not selling the boards then I'm paying for them myself, so you bet I care about the cost!

Your figures of $10 and $25, although not unrealistic, are based on small boards and are not necessarily typical. Further, you're only talking about a single purchase of a single design. Obviously the numbers will be all over the place since designs vary in size so much, but a good estimate for me would be more like $60* for a minimum quantity order. So that's $60 on average every time I want a board made, which could happen several times in one month. So, it isn't "my god! $10 for a PCB!" but rather "my god! $200+ per month just for PCBs!"

Since indeed I cannot afford that, I'm certainly glad to see people looking for new and interesting ways to solve the problem. As long as the monetary cost of having a PCB made remains a limiting factor on how many PCBs I design, there is room for improvement.

As a side note, any notion of needing expensive equipment to get any "real work" done is also completely absent in the hobby world. I don't know what "real work" means exactly, but all I'm interested in is whether I can call a design "useful". There are a lot of products that don't exist, not because it is required that someone to apply "real work" in order to produce them, but simply because nobody happens to have produced them yet. It is a fallacy to think that "real work" must be applied in order to produce something which is useful. In a large number of cases, all that is required is to put some puzzle pieces together.


EDIT: *actually, $60 doesn't even include the $24 charge for DHL delivery. I can amortize that all the way down into the noise for commercial projects, but for personal stuff that's a huge amount of money to blow per order/board!

[pickle9000]

I see a few problems here first is you have two groups the professionals and non-professionals. Next you have the fact that people in general are cheap and want a great deal for the dollar. I don't really see any real bang for the buck or at least I want to have that kicking around the shop for next time.

- Perhaps a set of boards that people could use, general purpose op amp boards to plug into a breadboard for example or a signal conditioning board. Perhaps a whole set of them and you just snap off the one you need when you need it. Basically straight off the datasheet.
- For the non professionals or educators building blocks rather than complete systems, a whole set as well and just snap off what you want. Simple pulsers, and so on.

How many boards could you fit on to one page, or on a 2x10 inch strip? Is that worth 20 bucks? I don't know.

[free_electron]

So, if we impose rules : must be open source , open hardware or no desl.
then It all pivots around the usefullness of the designs...
Do you really want to stock up on 5 flavors of blinkylights, 10 flavors of a steppermotorshield ?
Maybe i'm no longer classified as a hobbyist , but in 99% of the cases i don't find exactly hwat i want so ir oll my own board. I hate contraptions made from 5 miles of wire and 15 different pcb's held together by spit and ducttape and stuffed in an old cigarbox. In the time it takes me to make suchan assembly i have captured the schematic, done a real nice layout, tweaked it exactly the way i want it and it will fit in a nice housing a look like a real product. Even if its a one off.

If you are going to build.a tube amp you also will want a nice mahogany or gold plated case with some mirrors to show the shiny glowing tubes. Nobody makes an amp as a hairball and then puts it in the living room. My hobby projects need to look like they are a commercial product. That is the fun of electronics. Building something that is 'finished' in all aspects.

So yeah , i will roll my own board. But yeah , maybe i'm not your average hobbyist so maybe what i'm saying is irellevant.. But i cant help wondering, when i visit the local electronics gathering, why all the 'arduino crowd' projects look like the stuff pigs roll in. Invariably a stack of boards and wire spaghetti. Why don't you make a nice board ?  That seems to be a step too far.

So there is the problem. The arduino crowd doesnt make their own boards. They dont know how. They just buy premade and slap wires between them.
The crowd that does make boards wants nicer boards... And those are so customized the arduino crowd isnt interested in em.

Tough call...

[marshallh]

So, if we impose rules : must be open source , open hardware or no desl.
then It all pivots around the usefullness of the designs...
Do you really want to stock up on 5 flavors of blinkylights, 10 flavors of a steppermotorshield ?
Maybe i'm no longer classified as a hobbyist , but in 99% of the cases i don't find exactly hwat i want so ir oll my own board. I hate contraptions made from 5 miles of wire and 15 different pcb's held together by spit and ducttape and stuffed in an old cigarbox. In the time it takes me to make suchan assembly i have captured the schematic, done a real nice layout, tweaked it exactly the way i want it and it will fit in a nice housing a look like a real product. Even if its a one off.

If you are going to build.a tube amp you also will want a nice mahogany or gold plated case with some mirrors to show the shiny glowing tubes. Nobody makes an amp as a hairball and then puts it in the living room. My hobby projects need to look like they are a commercial product. That is the fun of electronics. Building something that is 'finished' in all aspects.

So yeah , i will roll my own board. But yeah , maybe i'm not your average hobbyist so maybe what i'm saying is irellevant.. But i cant help wondering, when i visit the local electronics gathering, why all the 'arduino crowd' projects look like the stuff pigs roll in. Invariably a stack of boards and wire spaghetti. Why don't you make a nice board ?  That seems to be a step too far.

So there is the problem. The arduino crowd doesnt make their own boards. They dont know how. They just buy premade and slap wires between them.
The crowd that does make boards wants nicer boards... And those are so customized the arduino crowd isnt interested in em.

Tough call...

summed up my thoughts exactly.. if it is worth doing it is worth making a pcb for, though tbh there is no way to NOT use a pcb for what I do

[pickle9000]

In the time it takes me to make suchan assembly i have captured the schematic, done a real nice layout, tweaked it exactly the way i want it and it will fit in a nice housing a look like a real product. Even if its a one off.


The crowd that does make boards wants nicer boards... And those are so customized the arduino crowd isnt interested in em.

Tough call...

Nice if you have the time, now I am an old retired bastard and I hate to admit it it but I have sent some one-off's out the door that made me cringe. They worked, lasted, had documentation but looked like crap on the inside. The customers where happy so should it have bothered me, yes I'm a professional it should bother me. It is nice to do it the right way but prototyping is often messy and time constrained. As for a hobby, I say do it however you want if it smokes do it better next time.

[Corporate666]

It's not hard to understand.

There are not simply 2 kinds of project, commercial where you are not 'allowed' to disclose your source and personal where you are.  Some people have a design they are "allowed" to disclose, but do not want others buying it for a low price - as in FE's example of a Kickstarter project. Also, just because someone is doing something for themselves and not commercially does not mean they are OK with someone else selling their design and work for a profit. 

It may be that I am underestimating the disparity which exists between the hobbyist world and the professional world. We appear to be facing this from different perspectives with several significant differences.

You're certainly right - there are still a number of personal designs that need to remain private for one reason or another which are excluded from a service like this. The commercial designs you mention, which are also clearly excluded, don't even enter the picture to begin with. What may not be as obvious as I'd have thought is that the number of designs remaining after these exclusions is both considerable and growing. Just because from where you sit you can see a lot of designs which don't apply doesn't mean that there aren't also a lot of designs which do. Most of my designs have no need for privacy, and I would surely welcome the opportunity to pay for their manufacture with the source files rather than money. For everything else, well, there are obviously a lot of other places to get boards made.

I'm also surprised at people talking about "the expense of a PCB".  Are people really so cheap?  It does not matter at all what the cost per PCB is, unless it is a commercial design where you have a unit price to worry about.  All that matters is the total cost to get the 1 or 2 boards you need.  $9.90 from iTead for a small board, or $25 for a larger one is comparatively expensive?  Compared to what?  The cheapest shipping for a single resistor from one of the distributors is going to be something like $6 or $8.  And it is very hard to do any real work in electronics without at least a couple of thousand dollars worth of equipment, hardware and software.  I can't imagine there is anyone who has a design they want to implement and doesn't mind paying the single unit prices for chips, passives, switches, etc...but who then says "my god!  $10 for a PCB?  They must be crazy!  I can't afford that!" 

The disparity between worlds really stands out here. From my perspective, it does not matter at all what the cost per PCB is unless it isn't a commercial design where the cost is covered by a customer. If I'm not selling the boards then I'm paying for them myself, so you bet I care about the cost!

Your figures of $10 and $25, although not unrealistic, are based on small boards and are not necessarily typical. Further, you're only talking about a single purchase of a single design. Obviously the numbers will be all over the place since designs vary in size so much, but a good estimate for me would be more like $60* for a minimum quantity order. So that's $60 on average every time I want a board made, which could happen several times in one month. So, it isn't "my god! $10 for a PCB!" but rather "my god! $200+ per month just for PCBs!"

Since indeed I cannot afford that, I'm certainly glad to see people looking for new and interesting ways to solve the problem. As long as the monetary cost of having a PCB made remains a limiting factor on how many PCBs I design, there is room for improvement.

As a side note, any notion of needing expensive equipment to get any "real work" done is also completely absent in the hobby world. I don't know what "real work" means exactly, but all I'm interested in is whether I can call a design "useful". There are a lot of products that don't exist, not because it is required that someone to apply "real work" in order to produce them, but simply because nobody happens to have produced them yet. It is a fallacy to think that "real work" must be applied in order to produce something which is useful. In a large number of cases, all that is required is to put some puzzle pieces together.


EDIT: *actually, $60 doesn't even include the $24 charge for DHL delivery. I can amortize that all the way down into the noise for commercial projects, but for personal stuff that's a huge amount of money to blow per order/board!

I think all of this ends up being academic because we both agree at least that the proposed business idea will not work for the secondary reason that the boards submitted will be of little to no use to anyone else.

There are businesses that do not exist because nobody has yet thought of doing business in that way - and while developing a new niche usually costs metrics craploads of money, some companies find a way to make it work.  Then there are businesses that do not exist because they just won't work.  I think this one falls into the latter category.

[edavid]

I agree that the original idea wouldn't work unless the OP could reject unmarketable designs.  How about this, though: subsidize multiproject panels by filling in the wasted space between boards with small standard board designs.  I'm thinking in particular of SMD to DIP/breadboard adapters, which seem to be in demand, but are overpriced even from China.  Are there other small boards that could be made?

[marshallh]

I agree that the original idea wouldn't work unless the OP could reject unmarketable designs.  How about this, though: subsidize multiproject panels by filling in the wasted space between boards with arduinos

Fixed

[Corporate666]

I agree that the original idea wouldn't work unless the OP could reject unmarketable designs.  How about this, though: subsidize multiproject panels by filling in the wasted space between boards with small standard board designs.  I'm thinking in particular of SMD to DIP/breadboard adapters, which seem to be in demand, but are overpriced even from China.  Are there other small boards that could be made?

If the business idea is to use the designs from customers as a loss leader so that the free space could be used for items that would turn a profit - like standard board designs (whatever that means), then as a business guy, my first question to someone proposing this idea would be "why not just make the whole panel these profit generating standard board designs, and stop killing profitability of the panels with loss-creating boards".

[pickle9000]

If the business idea is to use the designs from customers as a loss leader so that the free space could be used for items that would turn a profit - like standard board designs (whatever that means), then as a business guy, my first question to someone proposing this idea would be "why not just make the whole panel these profit generating standard board designs, and stop killing profitability of the panels with loss-creating boards".

What? Make useful stuff that's in demand? Does that work? Will it sell?

[johnboxall]

Anyway, the site isn't live quite yet, but wanted to get feedback from the community on whether or not this has any chance of success.  Thoughts?

The success of your business (i.e. ability to make a profit) will be dependent on whether the boards you have will be wanted by others. If your first fifty orders are for obscure personal projects that nobody else is interested in - you'll be bleeding money for a long time. To be honest I think you'd be better off keeping your money and directing people to OSHPark or similar.

[westfw]

So the idea is basically to allow an "open source HW designer" to avoid the $25 cost of having ten PCBs made at Seeed/Itead/etc by selling, say, six of those boards for $8 each to other people who might be interested.
$48 - $25 = $23 profit for the service, and the original developer gets four boards free in return for their design efforts.  Is that approximately correct?

I have to agree with other people that it won't work.

1) the cost of having PCBs made isn't high enough to be a barrier to the original developer.
2) The original developer could do this pretty easily themselves.  With much less effort expended on the shipping/etc than needs to be expended to make the difference between a "PCB for my own use" and a "PCB usable by the general public."
3) In essence, you'd be a shipping service, and I don't think there is enough "value add" to splitting up PCB shipments for it to actually work as a business.

Several of the PCB services seem to offer mechanisms for allowing your board to be bought by random strangers, if you want.  I'm not acquainted with the details, but a first step would be to check on how those work, and how well they work...