DIY PCB STARTUP

[coppercone2]

what do you do if the JLCPCB order button is greyed out?

This is just a question of their organizational philosophy. Not everyone agrees that just because currently there is a advanced overseas manufacturing facility providing some service that the skill is not worth learning or that it might not help develop a mind. Not everyone just had the assumption that its 'bad' to learn because of availability.

I mean they can easily get the money from university to setup a assemble line to make a custom complex bus bar, or wind a inductor or even get fancy and actually make your own professional quality capacitor.

It seems like a perfectly fine exercise that has the prospect of actually offering a service, complete with local expertise (i.e. friends) that can be used for team building between departments and all sorts of educational things.

I swear, if he said that he was gonna get the equipment together to make a old time 74 type chip everyone would just think its awesome for some reason, even if its basically the same thing but smaller and possibly more useless then being able to make a PCB, because no one wants DIY 74 series chips.

A university can hire basically any service to do any one of their electronics design projects in half the time and possibly cost, but the goal of educating people would not be met... i.e. run it like a private research contract. In private the boss does not care if you learned something or if your a clever manager that hired someone to do something, because unless there is some other business objective (involving building expertise) , it gets done faster and cheaper.

[PlainName]

If it took all day it would still be a winner if you don't want to wait or throw money away.

How is it sensible to spend an entire day making a simple one-sided PCB?

Because you want it now and can do it for now. I really don't know why it's so hard to grasp such a simple concept!

[PlainName]

I don’t think you understood what I am saying.

No, clearly you haven't thought about what I wrote. The board layout is already done because you have the process already in place. For someone with the DIY board making capability the process is already in place. You are moving the goalposts and on the one hand saying this works because the process is in place, and that doesn't because the process isn't in place. Yes, that's true, but it's bloody apples and oranges. Play with a level playing field, eh.

So, even with your 'process already in place' thing, at one time it wasn't. So that's the same situation as someone without the board making process. Apples vs apples, see. And, as I said, you gotta start at the beginning right at the beginning, and the ONLY reason you can whip off a board design fast is because the process was started at the beginning and now you are a long way down the road. Yes, tinkering to get the process nailed down included.

All I was saying is that if you have your home PCB production process already perfected (that’s what I mean by “nailed in”) — you have already perfected your print settings and material selection, exposure times (for photosensitized boards) or transfer temperature (for toner transfer), and have perfected the etching time and temperature for your etchant — THEN you can quickly and reliably produce a PCB from your prepared PCB layout.

Yes, but like everything else, you gotta start before you get there and, like everything else, you can just keep putting that off until it never happens. Or, you can figure it's worth a go and do it, after which you're in the situation where you have a nailed-down process and then argue oranges vs oranges with someone on the same playing field.

As long as you don’t have dozens and dozens of holes to drill and don’t have complex board shapes to cut.

Wrong. Oh, that might be a problem for the way you would do it (and explain why you hate it so much you want to put everyone else off too). But for me it's no real difference - it's just gcode and how fast the mill can churn through it in the end. Hell, I can even do mid-board cutouts and slots with no more effort. Castellated holes should be just as easy too, I reckon (haven't actually tried that because I've not needed them).

If you haven’t been doing enough DIY PCB manufacturing to have a repeatable, reliable process, then you may end up spending a lot more time, thanks to failed attempts.

Yes, and as I pointed out, the same applies for designing the board, cooking, driving, understanding posts on forums, soldering... Oh noes, not long until the only way to make boards according to EEVBlog gurus is to have them assembled by JLC too. After all, only a few bob and a couple of days extra.

[coppercone2]

if you drill it all by hand you really get a stream lined design because your gonna question every single hole. no unnecessary swiss cheese

I think it limits how many 'rules of thumb' get included in a design for no reason. just a observation 

Sometimes with the fancy boards I wonder if I am seeing a shotgun approach taken, on some crazy multi cnc driller

I am sure its out there, relay dry contact with 0.05mm via stitching connected to a screw terminal with a 10kohm resistor stuck in there, connected to something electromechanical. Because a custom DRC routine that does "the best boards ever" got activated by a mechanical engineer. And then the board snaps in half when its screwed into the enclosure

[PlainName]

I think it limits how many 'rules of thumb' get included in a design for no reason. just a observation 

Sometimes with the fancy boards I wonder if I am seeing a shotgun approach taken, on some crazy multi cnc driller

Not holes specifically, but I wonder if the ease and cheapness of having decent boards done affects things. If it's going to cost serious money and take several weeks then you really don't want to have to redo it, so you'll be a lot more careful than if you can have it in your hands tomorrow for pennies. Perhaps a bit like programming, where once upon a time it took a while to compile and burn your EPROM(s) so you did things differently to how they might be done now where you can go from writing stuff to using JTAG to single step it in (seemingly) moments.

And... since it's easier to get into the game there will be a lot of new users who don't have the experience (aka previous cockups) or senior engineers to know how to do stuff 'properly' so it's a bit of a cargo cult approach.

[ebastler]

what do you do if the JLCPCB order button is greyed out? [...]

Huh?   Sorry, I have no idea what you are talking about. The picture you included did not help.

[coppercone2]

a argument was had about the curriculum of a academy in a famous movie. I think its a situation that students and teachers often run into involving using simpler methods that have been automated.

[PlainName]

Gotta be honest: if DIY PCBs involve big knives and hole in hands, I'm going JLCPCB in future.

[coppercone2]

they can involve both, if you have students use a shear and a manual drill press without supervision 

I think Denmark can handle it, I think their smart

because even if you get them made, you need to make sure they don't throw them at each other like cards

[dietert1]

I will never order from them. I don't like their animated Ads taking away useful window space. And i don't like to send my designs to people i don't know. Who orders from them supports a destructive effort financed by the Chinese government with the aim to destroy western business. It's kind of a war.

Regards, Dieter

[tooki]

I don’t think you understood what I am saying.

No, clearly you haven't thought about what I wrote. The board layout is already done because you have the process already in place. For someone with the DIY board making capability the process is already in place. You are moving the goalposts and on the one hand saying this works because the process is in place, and that doesn't because the process isn't in place. Yes, that's true, but it's bloody apples and oranges. Play with a level playing field, eh.

So, even with your 'process already in place' thing, at one time it wasn't. So that's the same situation as someone without the board making process. Apples vs apples, see. And, as I said, you gotta start at the beginning right at the beginning, and the ONLY reason you can whip off a board design fast is because the process was started at the beginning and now you are a long way down the road. Yes, tinkering to get the process nailed down included.

All I was saying is that if you have your home PCB production process already perfected (that’s what I mean by “nailed in”) — you have already perfected your print settings and material selection, exposure times (for photosensitized boards) or transfer temperature (for toner transfer), and have perfected the etching time and temperature for your etchant — THEN you can quickly and reliably produce a PCB from your prepared PCB layout.

Yes, but like everything else, you gotta start before you get there and, like everything else, you can just keep putting that off until it never happens. Or, you can figure it's worth a go and do it, after which you're in the situation where you have a nailed-down process and then argue oranges vs oranges with someone on the same playing field.

As long as you don’t have dozens and dozens of holes to drill and don’t have complex board shapes to cut.

Wrong. Oh, that might be a problem for the way you would do it (and explain why you hate it so much you want to put everyone else off too). But for me it's no real difference - it's just gcode and how fast the mill can churn through it in the end. Hell, I can even do mid-board cutouts and slots with no more effort. Castellated holes should be just as easy too, I reckon (haven't actually tried that because I've not needed them).

If you haven’t been doing enough DIY PCB manufacturing to have a repeatable, reliable process, then you may end up spending a lot more time, thanks to failed attempts.

Yes, and as I pointed out, the same applies for designing the board, cooking, driving, understanding posts on forums, soldering... Oh noes, not long until the only way to make boards according to EEVBlog gurus is to have them assembled by JLC too. After all, only a few bob and a couple of days extra.

Dude. The ONLY “process” I have referred to in this entire discussion is the process of manufacturing the board, not for designing it (doing layout), nor the process of developing the manufacturing process. Laying out the board is NOT part of the process of manufacturing the board! To be doubly clear, by “manufacturing the board”, I mean the work that companies like JLCPCB will do: physically making the board from the design you give them. They do NOT design the board for you. That’s not me moving the goalpost, it’s stopping YOU from moving it by attempting to move board design into the manufacturing process as you did in your earlier reply, or expanding the manufacturing process to include developing/perfecting the manufacturing process itself as you did in this reply!!


Anyway, you’re acting as though I don’t understand the concept of someone who makes PCBs themselves regularly. Of course I do, and the fact that my original comment started with “unless you have your process nailed in…” makes it obvious that a) I know that people who do make lots of PCBs themselves can exist, and b) that such people aren’t the ones I’m talking about! Students — the initial target group discussed in this thread — are almost guaranteed to NOT fall into the group of people who make lots of PCBs themselves. If they did, they would not need a novel device or service provider.

Now you’re suddenly talking about milling your PCBs? You didn’t mention that earlier. I’ve been working on the assumption that by doing it yourself, you meant the traditional methods of cutting a board by hand, etching, then drilling the holes by hand on a drill press. You never mentioned a CNC mill. But that means you are already discussing something radically different from the DIY everyone else is discussing.


I don’t know why you’re so angry and condescending with me. I wasn’t rude to you. I can only respond to what you write, not what you think but don’t actually say. You accuse me of not knowing how to read a forum post, yet you’re repeatedly ignoring or misunderstanding what I am saying even after I re-explained it to try and clarify for you. (I know how to read forums; I’ve been doing it for over 25 years.) But even if we are totally misunderstanding each other, that’s no reason for you to be rude.

And it’s certainly not fair for you to put words in my mouth: I never said that there is never any justification for DIYing a board.

[tooki]

what do you do if the JLCPCB order button is greyed out?

This is just a question of their organizational philosophy. Not everyone agrees that just because currently there is a advanced overseas manufacturing facility providing some service that the skill is not worth learning or that it might not help develop a mind. Not everyone just had the assumption that its 'bad' to learn because of availability.

I mean they can easily get the money from university to setup a assemble line to make a custom complex bus bar, or wind a inductor or even get fancy and actually make your own professional quality capacitor.

It seems like a perfectly fine exercise that has the prospect of actually offering a service, complete with local expertise (i.e. friends) that can be used for team building between departments and all sorts of educational things.

I swear, if he said that he was gonna get the equipment together to make a old time 74 type chip everyone would just think its awesome for some reason, even if its basically the same thing but smaller and possibly more useless then being able to make a PCB, because no one wants DIY 74 series chips.

A university can hire basically any service to do any one of their electronics design projects in half the time and possibly cost, but the goal of educating people would not be met... i.e. run it like a private research contract. In private the boss does not care if you learned something or if your a clever manager that hired someone to do something, because unless there is some other business objective (involving building expertise) , it gets done faster and cheaper.

I work at a university in the workshop. Not precisely the same thing as you describe, since we are a service provider for the department (which is not an electrical engineering department), and students rarely do any work in the workshop themselves (though they can if they ask to and do the training).

Regardless, what I hear over and over again, and witness myself in various ways, is that higher education is moving away from doing things in-house. In the electronics workshop (my little domain within the department workshop), I have drawers full of things like inductor cores, bare copper-clad PCBs, etc. that used to be done in-house. But as the workshops have been cut back, you just don’t do those yourself anymore. All the PCB equipment that used to be around (etching tanks, UV exposure box, etc) were disposed of years ago; only the (homemade!) PCB table saw remains. I know it used to exist because of the remnant materials and older in-house equipment that has homemade boards in it, as well as the binders full of old design documentation, including the PCB layouts — later ones laser printed onto transparencies, earlier ones made by hand with that black tape and rub-on pad shapes!

The university I work at still has several machine shops in various departments, a few electronics workshops, as well as a shared glassblowing workshop. Each of these staffed with skilled tradesmen. And the researchers are absolutely delighted to have these resources (and the ability to have us collaborate on their designs, since the researchers aren’t engineers). What I am told over and over by researchers who have worked at other universities around the world is that most of them have gotten rid of most or all of their workshops, requiring most or all jobs to be outsourced. And that is clearly the direction this university would like to go in the long run, too. The executives that run universities today just see in-house shops and services as cost centers and don’t realize the value they add. </rant>

[PlainName]

You never mentioned a CNC mill. But that means you are already discussing something radically different from the DIY everyone else is discussing.

It's DIY, and many people do it. Nothing strange or professional about it. And, indeed, it might be a useful thing for the OP to follow up since it won't involve nasty chemicals.

[tooki]

Not holes specifically, but I wonder if the ease and cheapness of having decent boards done affects things. If it's going to cost serious money and take several weeks then you really don't want to have to redo it, so you'll be a lot more careful than if you can have it in your hands tomorrow for pennies. Perhaps a bit like programming, where once upon a time it took a while to compile and burn your EPROM(s) so you did things differently to how they might be done now where you can go from writing stuff to using JTAG to single step it in (seemingly) moments.

100% agree.

I mean, both the “work quickly and iterate a bunch of times, since Chinese PCBs are cheap” and the “design it carefully in the hopes of not needing many revisions” approaches have their merits.

But we certainly have seen the downside of easy software and firmware distribution and updating: lousy, rushed software/firmware with tons of bugs, since “you can fix bugs later”. Not so much back in the days when firmware resided in EPROMs or even mask ROMs, and software had to be updated by mailing floppy disks to customers!

[tooki]

You never mentioned a CNC mill. But that means you are already discussing something radically different from the DIY everyone else is discussing.

It's DIY, and many people do it. Nothing strange or professional about it. And, indeed, it might be a useful thing for the OP to follow up since it won't involve nasty chemicals.

CNC milling PCB traces has proven to be one of the least attractive methods, one that a lot of people who try it quickly abandon.

I disagree about it being “professional”. It’s not quite good enough for real professional use (in that you can’t use it to prototype the multilayer boards that are now commonplace in professional settings, and that its resolution limits become real problems with many modern components. That problem is only going to get worse.). But it’s too complex and fiddly for most hobbyists to bother. (And most hobbyists don’t have a CNC mill.) So the niche where milling the traces makes sense is quite small.

DIY PCB manufacturing has traditionally meant etching after using either photosensitized boards and UV exposure, and to a lesser degree using toner transfer.

[coppercone2]

You never mentioned a CNC mill. But that means you are already discussing something radically different from the DIY everyone else is discussing.

It's DIY, and many people do it. Nothing strange or professional about it. And, indeed, it might be a useful thing for the OP to follow up since it won't involve nasty chemicals.

I am sick of this nasty chemicals business, particularly when people take a dremel to a piece of fiberglass in front of me and act like some how its better then a tiny bit of corrosive gas. 

Or when someone gets the sanding wheel on a painted object, thinking its some how better for them then the new formula of paint stripper. I find almost everyone I know does not take dust hazard seriously.

I really think if you use chemicals responsibility in a shop, it ends up being cleaner and also healthier. I mean corrosive stuff like millscale remover or even alumiprep. Easy to contain mild fumes vs high velocity jets of particles, usually involving healthy alumina or even silicates!

The one that is amazing to me is paint stripper. If you paint it on a moderate object, you will see JUST HOW MUCH paint comes off that sucker. You can scoop it up into a cup to measure it. Scary that someone might try to take all that paint with a wire wheel lol

On youtube you see some restoration guys standing in a paint dust cloud for like a day in high speed stripping paint off a refrigerator or car hood with a sander instead of just taking it off with a bit of chemicals.

That is why I think people should at least make an attempt to try a chemical PCB process, I like how most of the atoms get contained inside of a liquid and the fiberglass is unaffected. If you hate it you hate it.. but you might not


And, save for a few exceptions (glass working), more chemicals will enter your body through your permeable lungs then they will through skin contact.

[PlainName]

Whatever you use - chemicals, cutters, lasers - you have to take appropriate precautions. Personally, I got fed up with the sink turning yellow even though I didn't use it! And now, 20 years later, there is still a huge ferric chloride tank out in the garage because I have no idea how to dispose of the stuff.

[coppercone2]

mix it with baking soda and I bet in a month you can decant all the metals into a small amount of sludge

i would try to neutralize 1/4 cup and let it stand for a long time to see what happens

I think its pretty non soluable and non corrosive at this point, so you can throw it in the trash

and if you find it does something nasty like stain a stainless steel sink, the best solution is citric acid. If you can afford it, citri-surf gel is kind of like navel jelly for stainless steel.

not sure if it will clean porcelain too much, but I am pretty sure in the past I got ferric chloride stains out of porcelain by using toilet cleaner (HCl based)

I used my stainless sink as a container for a PCB etch, and the bubbles got a bit out of hand, and despite rinsing it, I did get a few spots of corrosion, but I sprayed it down with high concentration citric acid solution (you can add a drop of soap too) and it got rid of all the problems.. now its just kind of standard procedure to spray it down with citric acid to dissolve all the iron (it leaves chromium and nickel alone). I think a tiny bit of soap maybe makes it evaporate slower.

[PlainName]

I'll try that, thanks.

Or get run over by a bus, and then it's someone elses problem