Bread board etc.

afaiu once it has been written off just keeping it in storage it is the sam e as crushing it far tax goes it has no value but eventually you run out of s pace, if you sell it you probably have to jump through all kind of hoops to figure out how much was written off and how much you made selling it

If they can't classify it as some sort of in-kind contribution then just give it away. There is more to it that taxes, for example the recognition of a corporation in the community. A nicely equipped high-school or maker club lab with a placard stating who donated all that stuff goes a long way.

There should be a database of all capital equipment, including its book value. If the equipment is sold for more than its book value, sure, it's write-off has to be recaptured for tax purposes. If for less, the difference is written off. If it's crushed or donated it can be written off immediately, too. I don't see the paperwork being all that much different. It's what computers do. The difference is that it takes work to find someone who wants it. The dumpster is easy.

Nah, he could fit 3x as much stuff in there

NT

Last time it took me about 10 seconds. I simply asked a physics teacher. The times before that it was never more than a few minutes.

This goes for tech stuff though. For other things such as our Hammond organ that will be very different. Having spent a lot of time and money restoring it I hate to see it go but due to longterm consequences from a wrist fracture my wife can't play it anymore.

People should always ask themselves if the easy route is the proper way for society.

What's expensive is the engineering and layout time to spin the board, and another set of documentation, stencils, and p+p programming. Three to six weeks, usually, before we see another built rev.

Why not spend 15 person-hours of design review to eliminate a full PCB spin?

You have a choice between cultures: review and check and get it done right the first time. Or be sloppy and find your mistakes through testing.

One problem with the second style is that testing is a terrible way to find all the bugs. The same sloppy psychology prevails throughput the entire process. If you habitually shy away from looking at your own design mistakes, you will subconsciously avoid seeing the bugs too.

My manufacturing does what we tell them to do. Changes and kluges are documented in ECOs. Before we buy another batch of PCBs, we review all ECOs and the public NEXT file to see if we want to spin the rev. Sometimes we spin to improve manufacturability or testability, too.

It's a cultural thing.

Most of that is dirt cheap. Layout takes a lot of time the first spin but after that it's pretty quick. Complete documentation doesn't happen for anything that doesn't get released. Stencils are a couple of hundred bucks. P&P programming doesn't change a lot, either, and it's not like people are sitting around for six weeks doing nothing.

Because it doesn't.

False choice. Changes *are* going to be made anyway. Take advantage of the changes and get the first article to the programmers as soon as possible.

You're only seeing what you want to see.

That's fine when you're making a few units. Tell me that manufacturing does anything you want them to do when your production is 50K/month and it has to fit along with another few dozen similar products.

No, it's a different business sort of thing.

We do a full formal doc release for rev A, the first board that we etch. That board is built by manufacturing from the released drawings and BOMs, just like any other production order.

Usually, that is the one we sell. The FPGA and firmware are released after the first articles are tested.

Stencils are a couple

It does for us. Almost every time.

I see other companies assuming that there will be a breadboard, a prototype, a pre-production unit, and a pilot unit before release for production.

They do all that, too.

Assume that you will screw up multiple times, and you'll screw up multiple times.

"Make it perfect and beautiful the first time" is actually fun. It's our team sport.

It is a question of attitude. The assumption that it will not work the first time creates a self fulfilling prophesy.

Because you intend to sell Rev-A. The production groups use fully released BOMs, and such, because they use our manufacturing lines for their prototypes.

The first article is *never* the one we sell, even if it's perfect. There will be changes long before it goes to production. The plan is for the third of fourth spin to be the "production sample". Even that likely won't have all testing and certifications done. Testing an documentation costs as much as the rest of the development process so it's not done on early hardware.

It doesn't/can't be, in our process. Needs are different.

It's not a matter of screw ups, though that certainly happens. It's a matter of priorities. The number of spins isn't a huge priority. Schedule is, however, and unplanned spins (i.e. delays) are frowned upon.

If it's got a "blue wire", it's not beautiful *or* perfect. It wouldn't be manufacturable. Priorities and needs are different.

I love your boards John, but that one seems a little much to light an LED. :-) Mikek

I got to thinking, and looked back, I was joking about lighting an LED, But it looks like that is what you did! The first LED flashlight I bought at the Orlando Hamfest 15 or 20 years ago, it had more ICs and parts than that, but they fit on the end of a 9V battery. A point made was that it always glowed just a little so you could see it in the dark. Well wouldn't you know I let the kids play with it and then had a flat tire at night on the way home from Orlando. Shut the light off in the van and found it glowing under the rear seat. I used it to see to change the tire. Mikek

This is exactly my way. When I started experimenting with electronics some 40 years ago I bought a solderless breadboard but I soon experienced contact problems. Thicker wires plugged in bend up the contact and if the pin happens to be used with a thinner wire in a later circuit it may have contact or may not. Unfortunately, neither Spice nor a computer was available then. So I soon changed to soldered breadboards and verowire like this one:

I'm happy hot having trouble with stuff like this :-)

Especially if the management has 'ideas' when they see the first prototype laying on their table...

Cheers

Robert