Ok, I have a working prototype, now what? (massive scale building)

Thread Starter

Edgar Gil

Joined May 10, 2018
13
Hi friends,
I don't know if there is a previous post about this, I did a little research and did not find something related.

I have a project and I achieved a working prototype, so I want to know some advisory or experience in case to build in massive scale.

Thanks in advance.
 

dl324

Joined Mar 30, 2015
16,839
What does massive mean in this context? 100 thousand units? A million?

Do you want this completely assembled by your manufacturing partner? Do you want them to handle stocking inventory and shipping?

Better information will get you better suggestions.
 

MisterBill2

Joined Jan 23, 2018
18,167
Once you have a working prototype the design still needs to be optimized for production and the best production yield. That means reducing the chances of assembly errors, and part tolerances causing incorrect operation. It includes assuring that all BOM items will be available for the production run. Finally, it includes determining the most efficient sequence of assembly steps. That is the process of making a collection of parts into the finished product.

So there you have a general list of the next steps for whatever it is that the product may be. You are a fair distance from production at this point.
 

Papabravo

Joined Feb 24, 2006
21,158
Hi friends,
I don't know if there is a previous post about this, I did a little research and did not find something related.

I have a project and I achieved a working prototype, so I want to know some advisory or experience in case to build in massive scale.

Thanks in advance.
Raise massive amounts of money, to pay professionals to setup and run a factory with three shifts. $50 million for starters over the next 12 months. At least it is cheaper than a semiconductor fab. Those are probably up to $10 billion now.
 

MisterBill2

Joined Jan 23, 2018
18,167
PPB, it could be that the TS is a new hire at a company that already has the capabilities. And the size of a "large" production run varies a whole lot with the product. At one place I was employed, a production run of six engine test stands was huge, while at one factory I visited a run of a thousand motorcycles was considered a "custom build" for a specific dealer group.

The steps that I described are rather basic, but very important. They apply just as much to a big factory as to a hobby-type creator wanting to go into production of a new widget. Huge stacks of cash are not always required, unless one is starting from bare earth.
 

MrSalts

Joined Apr 2, 2020
2,767
Raise massive amounts of money, to pay professionals to setup and run a factory with three shifts. $50 million for starters over the next 12 months. At least it is cheaper than a semiconductor fab. Those are probably up to $10 billion now.
not even Apple is so dumb to set up a new factory for a new device. The key phrase is "outsource" or 3rd party manufacturing. Call Flex (formerly FLEXtonics) or Foxconn if it is 100k+ parts per run. Let us know the number. Also, you'll need an industrial designer if you want to make a sleek looking consumer product (and even helpful for industrial products. Then mechanical design for the cad files to turn the designer's renderings to CAD files that can be turned into molded, machined, stamped, (insert other fabrication methods here).

@Edgar Gil send me a PM if you need names of designers, manufacturing method consultant, mechanical engineering or part producers.
 

Papabravo

Joined Feb 24, 2006
21,158
PPB, it could be that the TS is a new hire at a company that already has the capabilities. And the size of a "large" production run varies a whole lot with the product. At one place I was employed, a production run of six engine test stands was huge, while at one factory I visited a run of a thousand motorcycles was considered a "custom build" for a specific dealer group.

The steps that I described are rather basic, but very important. They apply just as much to a big factory as to a hobby-type creator wanting to go into production of a new widget. Huge stacks of cash are not always required, unless one is starting from bare earth.
The strategy that we employed with our enterprise could be described as:

"Build a few, sell a few."
This had the enormous advantage of controlling the risk involved. Our biggest problem was suppressing the urge to overbuild so we could quote improved delivery times on the next order. We almost succumbed to the pressure. Had we done so we would have had to eat about $80,000 in finished goods that would never have been delivered.

Given the limited amount of information in the original post, I felt it was appropriate to take some liberties with the answer. If the TS wants a more serious answer let him refine the question. The TS is making the classic mistake definitively identified by Lancaster; he is thinking like an inventor, instead of like a businessman. If he keeps it up, I can guarantee with 100% certainty, that he will be fleeced, like a heard of sheep.
 
Last edited:

Papabravo

Joined Feb 24, 2006
21,158
not even Apple is so dumb to set up a new factory for a new device. The key phrase is "outsource" or 3rd party manufacturing. Call Flex (formerly FLEXtonics) or Foxconn if it is 100k+ parts per run. Let us know the number. Also, you'll need an industrial designer if you want to make a sleek looking consumer product (and even helpful for industrial products. Then mechanical design for the cad files to turn the designer's renderings to CAD files that can be turned into molded, machined, stamped, (insert other fabrication methods here).

@Edgar Gil send me a PM if you need names of designers, manufacturing method consultant, mechanical engineering or part producers.
Right. Apple is not run by inventors, it is run by businessmen who don't need to post on a hobby electronics forum for advice on monetizing an invention.
 

MrSalts

Joined Apr 2, 2020
2,767
I've come across a small-time inventor and mentored them to show them how parts can be made. Not everyone with an idea knows how to get it manufactured. In fact, very few do know how to do it. Luckily for the inventors who I mentored had something quite easy to make once I showed them the right manufacturing method - allowed them to save about $125k by avoiding a two-shot injection molding tool.
 

MisterBill2

Joined Jan 23, 2018
18,167
Certainly this is true! After the functionality of something is verified then the hard part is creating a good producible design. That is where so very many products go wrong, and has been the down-falling of so many products over the years. AND, there are some products that even though they do work shuold never be produced.
 

Marc Sugrue

Joined Jan 19, 2018
222
What do you mean by working prototype? I have had many first stage prototype that "work" but were a long way from being ready for formal testing and certifications & approval. So the next thing you do will depend on how 'working' the prototpye is. As this is an electronics site i assume your product is electronic so you need to consider what have you verified and passed so far? Consider things like safety, voltage range, current, interrupts, power factor, isolation, insulation resisance, Conducted Emissions, Radiated Emissions etc.. to name a few. If by working you have tested and passed all these to meet your relevant certifcation requirements then your are perhaps looking more towards the productionisation.
 

geekoftheweek

Joined Oct 6, 2013
1,201
Before even considering manufacturing you need to check out patents, copyrights, and whatever other sort of intellectual property laws are applicable in your area.
 

Wolframore

Joined Jan 21, 2019
2,609
I’m ok talking generally as there are considerations that should apply to any invention. In order to justify mass production a market test for feasibility and marketability would be in order. Plus during these tests attention should be paid to find faults and solutions. Worse scenario would be to go direct to mass production and find there are fatal flaws and no demand with capital locked up in useless inventory and R&D.
 

Audioguru again

Joined Oct 21, 2019
6,672
What about detailed tests of the finished products?
I worked for a Pro-Audio company who developed a new little speaker. But the speaker sounded awful unless it was equalized.
So I designed a 4-stage equalizer circuit that made the new speaker sound great. I also designed a test circuit with pass or fail indicators that tested the equalizer frequency response, distortion, noise level and supply current. Tens of thousands of equalizers were built by a local manufacturer and only two of the products failed. One had a defective electrolytic capacitor and the other had its TL074 quad audio opamp IC mounted backwards. Of the entire production that was sold, none were returned.
 

Thread Starter

Edgar Gil

Joined May 10, 2018
13
Hi guys, thank you for your answers,

What does massive mean in this context? 100 thousand units? A million?

Do you want this completely assembled by your manufacturing partner? Do you want them to handle stocking inventory and shipping?

Better information will get you better suggestions.
I'm thinking about 5k-10k units, I want it totally assembled, I want to handle stocking inventory and shipping.


Once you have a working prototype the design still needs to be optimized for production and the best production yield. That means reducing the chances of assembly errors, and part tolerances causing incorrect operation. It includes assuring that all BOM items will be available for the production run. Finally, it includes determining the most efficient sequence of assembly steps. That is the process of making a collection of parts into the finished product.

So there you have a general list of the next steps for whatever it is that the product may be. You are a fair distance from production at this point.
Thanks, the need to minimize the chances of making mistakes in mass production, very good advice.


not even Apple is so dumb to set up a new factory for a new device. The key phrase is "outsource" or 3rd party manufacturing. Call Flex (formerly FLEXtonics) or Foxconn if it is 100k+ parts per run. Let us know the number. Also, you'll need an industrial designer if you want to make a sleek looking consumer product (and even helpful for industrial products. Then mechanical design for the cad files to turn the designer's renderings to CAD files that can be turned into molded, machined, stamped, (insert other fabrication methods here).

@Edgar Gil send me a PM if you need names of designers, manufacturing method consultant, mechanical engineering or part producers.
Thanks for the offer, surely will do if I need specific advisor or consultant, thank you again.

The strategy that we employed with our enterprise could be described as:

"Build a few, sell a few."

This had the enormous advantage of controlling the risk involved. Our biggest problem was suppressing the urge to overbuild so we could quote improved delivery times on the next order. We almost succumbed to the pressure. Had we done so we would have had to eat about $80,000 in finished goods that would never have been delivered.

Given the limited amount of information in the original post, I felt it was appropriate to take some liberties with the answer. If the TS wants a more serious answer let him refine the question. The TS is making the classic mistake definitively identified by Lancaster; he is thinking like an inventor, instead of like a businessman. If he keeps it up, I can guarantee with 100% certainty, that he will be fleeced, like a heard of sheep.
"Build a few, sell a few" Good advice for controlling risk as you said, I have some sponsorship and advisoring from some bussinessmen, just want to read some experiences and advices in case that is required to make a massive build of the developed device.

The struggle is real, we all have so many great ideas and it takes so much work to get anywhere with them.
It would be really nice if this helps others makers in the development of their ideas.

This article is nearly 30 years old, and I have posted a link to it maybe a dozen times. Everybody with an invention needs to read and understand it.

https://www.tinaja.com/glib/casagpat.pdf
I'm sorry to tell you that I disagree with the article, I know that a patent is difficult and expensive to get, maintain and defend, I think the way when you get a patent is to sell it to someone that handle patents, and in some way get involved in the business, kind of combination of the advisor, thanks for sharing,these types of articles about someone else's experience are totally worth reading, it helps to visualize the situation from other perspectives, thanks again.

What do you mean by working prototype? I have had many first stage prototype that "work" but were a long way from being ready for formal testing and certifications & approval. So the next thing you do will depend on how 'working' the prototpye is. As this is an electronics site i assume your product is electronic so you need to consider what have you verified and passed so far? Consider things like safety, voltage range, current, interrupts, power factor, isolation, insulation resisance, Conducted Emissions, Radiated Emissions etc.. to name a few. If by working you have tested and passed all these to meet your relevant certifcation requirements then your are perhaps looking more towards the productionisation.
This is the kind of advisoring that I'm looking for, thank you so much, will dig more into them, if you know any paper or recomendation about this will be really helpfull for me.

Before even considering manufacturing you need to check out patents, copyrights, and whatever other sort of intellectual property laws are applicable in your area.
That's the actual step where I am, waiting to receive intellectual property advisoring, from the team that is sponsoring me.

Thus far we have no clue as to what the prototype might be. So at this time there is no more need to comment.
I'm sorry, but I can't give too much details (by sponsor recomendation), what can I tell you is that is a measurement device, based on ADC's and sensors, pretty simple.

What about detailed tests of the finished products?
I worked for a Pro-Audio company who developed a new little speaker. But the speaker sounded awful unless it was equalized.
So I designed a 4-stage equalizer circuit that made the new speaker sound great. I also designed a test circuit with pass or fail indicators that tested the equalizer frequency response, distortion, noise level and supply current. Tens of thousands of equalizers were built by a local manufacturer and only two of the products failed. One had a defective electrolytic capacitor and the other had its TL074 quad audio opamp IC mounted backwards. Of the entire production that was sold, none were returned.
I thouhgt about this, to make a bench test, the device is modular, so i thought to make another devices to test in line all the modules, even for calibration purposes it work, thank you for your good advice and sharing your experience.


In general lines my friends, what I want to know is how to build electronic devices, board and components, and the best way to build the case, I've worked at this point with a general purpose molded plastic case, (even I've used food recipients for prototyping lol), but I want to know about your experience (if you have been involved in getting devices built somewhere), about your recomendations in that particular field, according your experience.

I want to thank you all your comments, honestly didn't think I was a have such a great receptivity from your part, great community.

I'm sorry if I'm a little bit rude about sharing about this, but i dont want to mess this opportunity talking more that I have to, I would love to share more about this, even know your recomendations about what Im doing (that im surely you are more qualified an have more experience than me), but sadly at this moment can't share more details.

Thanks in advance my friends.
 

dl324

Joined Mar 30, 2015
16,839
I'm thinking about 5k-10k units, I want it totally assembled, I want to handle stocking inventory and shipping.
Do you already have all of the design documents? With volume that low, they'll probably want to source the components so they can increase profits by marking them up.
 
Top