complete circuit board lab & pov business card - food grade mylar rolls
Although there are many instructions on how to make a circuit board, this is different.
This is a description of what is needed to make the board, especially a gorgeous business card toy.
Over the past six months, I have built a fairly complete PCB manufacturing lab in my apartment, low cost, safe and reliable, and I intend to cover all aspects of the whole process from start to finish
You may have seen some before, but here it's all in one place with references.
All in all, I have to design and build a gas tight etching tank with laser cutting acrylic, SMD vacuum pick-up tool, reflux furnace and temperature controller to improve the toner transfer process using an improved laminating machine, build custom programming fixtures, and of course design, program and build every aspect of everything I do from scratch: my business card.
Finally, it's really worth having the time to have the board in my hands one night.
Business cards use the classic persistence of visual illusion, flashing your name and number in mid-air when you wave your business card.
Based on PIC12F508, an 8-Pin 6 I/O ultra-
Low cost micro controller, it is completely surface mounted, very thin
It uses PCB Laminate as thin as a standard business card.
The price of the parts is about $2 per piece, depending on how good you are in purchasing the parts, they are cheap enough (
To the right person).
But why do these troubles just to do something for the people I just met?
Why not print out Kinko's for less than $10 in an afternoon?
Why, because I want a card that won't be thrown away.
A card that reflects what I do, rather than awkwardly trying to generalize it with a clever position name.
A card that can give me the location.
This scene can imagine the following if you like: you just met someone you need to know.
In fact, they need to know you.
After exchanging the introduction, "the man", wearing a dark suit, wearing a tie, casually handed you his cloning company --
"This is my card," he said with a smile, knowing that you will be impressed with the fact that he has the card, or at least have the power to the person who owns the machine.
He might want to see you scribble down your number on a torn piece of paper.
But when you reach into your pocket and take out your card, he certainly doesn't want to see it. . .
Your name shines in the air, floating in front of his eyes!
When you hand him your card, you say, "This is called the persistence of vision . ".
"I made these in the basement. From scratch.
"You don't even need to say another word;
More just gloating.
You can see from his eyes.
He already sold it.
Statement: I should note here that this statement involves a lot of things that could be dangerous if done without a small and informed plan, including the risk of fire, the risk of electric shock in a variety of terrible ways, the treatment of toxic chemicals and toxic smoke.
Guys, this one really has everything. So be safe!
We will start with hard cans: build leak-proof, acid-proof and closed Etching tanks without using metal parts.
Others have done the same before, but I have to make myself difficult.
I used to build one with a plastic ammo tank, but it's too big.
I want to do things well this time.
I design from scratch in illustrator and then have a friend cut out parts from acrylic with a laser that meet my specifications (
One for laser cutting machine! ).
For those new here, you need an etching tank to hold the acid you use to eat the copper on the board that is not covered by some kind of etching-resistant etch.
Depending on the type of acid you use, it can be something that is very annoying, if you plan to do this on a regular basis and in the house where you live and breathe, you need to control it safely.
Or near the kitchen sink
This is doubly important if you plan to store it for any time, although make sure it's not a safe permanent storage container.
The design should be sealed, with a safe lid, an air inlet with a bubble for stirring and filling the solution with oxygen, and an outlet leading to the House to ventilate from the window.
I also added a check valve to the air inlet of the tank so that no exhaust gas goes into the air pump.
It should keep the plate upright with a minimum amount of acid and be opaque to the light, so that the solution does not degrade due to exposure to the sun.
However, I keep one side very clear so I can look at the board and see if it has been etched and also because it looks cool.
I designed a total volume of about 2 liters for my tank and plan to fill only 1 liter.
The 6L leaves room for bubbles and hoses, the board itself, and the top safety lip. The 1.
The 6L value is determined with the graph.
The area of the board reported here is 016 liters per square centimeter.
Assuming 50% covers a, 8x10 double sided PCB of 36 μm ("1oz")copper.
For reasons I will elaborate later on, I chose to use copper chloride (CuCl2)
As my board exercise Song
CuCl 2 is corrosive to most metals and some plastics.
When studying the materials you plan to use, there are a large number of data tables under the search term "xyz chemical resistance.
"Understanding the performance of the material is critical in engineering, so plan ahead.
It's not fun to find a hose that melts and pollutes the solution in your acid bath and then have to take out every small piece.
Few things can survive in highly concentrated acids forever.
As far as these things are concerned, we do not use anything highly centralized, but the effect is the same; it's acid-
Resistant to acid.
The following materials are generally safe for tank construction: The following are absolutely unsafe and should be avoided from being exposed directly anywhere in the solution: the most difficult part of building this thing is to find the parts to be found.
My whole life seems to be just an endless pursuit of what I need.
The hardest thing to find is a pair of partition fittings that are small enough.
Small accessories are not common locally, because everything is pipe-sized, and on the internet, small accessories at lab level can only be ordered in bulk.
Some of these parts are stolen, most of them are special parts that I have to order, most of them after several rounds of searching and receiving parts that don't fit together.
Almost all of this comes from the American plastics company;
As far as I am concerned, they are like Digi keys for tubes and lab accessories.
List the sources available.
As far as the tank is concerned, the first step is to assemble it.
The standard for waterproof acrylic bonding is Alpha chlorine, which is a very annoying thing, and cancer, as well as outside.
It is definitely not enough to open a window.
So, given the recent regulations, it's hard to find, so you may have to order the AFC-based glue such as IPS Weld-online-
On the 4 th, the water is also very thin.
I got a small bottle from a professional friend who made acrylic.
To be sure, you may be able to get through with the right epoxy, but ideally these parts should be chemically combined with each other rather than mysterious foreign reagents.
If you want to be safe, use Weld-
Finish it on the 16 th.
It is a more viscous solvent that can be treated as glue and will close any gaps.
Due to the lack of a replacement, I had to use the Asian armor blue.
When bonding the parts with a sub-Alpha chlorine, you will fit as closely as possible.
The edges of laser cutting need special attention as they are melted and rounded.
It took me about an hour to clean up the parts with files and sandpaper to keep them in good health. .
Even for the parts that fit perfectly together from the beginning, especially the uncut surface of acrylic resin, it is important to polish the adjacent surface with sandpaper, otherwise there will be no grab.
Once the parts are assembled, fix the parts together and run the solvent-filled syringe or needle of the improved extrusion bottle along the crack;
It is attracted by capillary action.
Parts will be safe enough to leave alone in about 60 seconds.
It takes only about 30 minutes to set it up, sometimes much less.
I put it overnight to be sure, but after about an hour it was actually waterproof.
After that, you need to make a lid.
I cut a self-adhesive gasket for the lid and lip of the tank to fill the gap, to seal it, I installed a pair of wings on both sides of the tank and pulled it down with a magnet, and keep it in place.
What happens is that the lid tends to align because the polarity on both sides is the opposite, so it is impossible to place the lid back.
I still think it's smart.
The only thing to note is that in order to prevent corrosion, the magnet needs to be sealed in the heat shrink tube.
The partition fitting I want is only straight, not 90 degrees, so I need to cut one side and weld it on the reclined fitting.
It works very well with polypropylene and the melting point of polypropylene is low enough that you can melt the parts with just a heat gun.
Please note that the equipment you modified is now a potential fire and electrical hazard!
So you probably shouldn't leave it unattended.
Make sure it's unplugged before you do anything.
The toner transfer process is considered to be the most direct process in the manufacture of an active etching anti-corrosion mask
As we all know, the forward method is the toner transfer method, and its components are easy to obtain and low cost.
The basic idea is that you print your design onto a smooth sheet of paper and output the print to your board with heat and pressure.
The paper is removed, leaving only Toner as an etch-resistant etching, thus protecting the copper below from the acid bath.
There are a lot of variables involved, so it's easy to understand why many people give it up.
However, the process is likely to stop if there is enough time.
If you have tried this before and give up, I suggest you not give up!
The return of time and money is worth it.
The process can also be used to add a black component label or a text mask to the board as the last touch after cleaning up the copper trace.
For the POV card, I use this method to add my name to the back of the card in case the led is too hard to read.
Modifying the laminating machine is the most important part of the whole process and is the balance between heat and pressure, which is consistent across the board.
It is easy to reach the proper pressure with standard bag laminating machine.
All you need to do is make sure to find a roller with a heating roller or you won't get the same result.
GBC has produced two commonly used models of GBC creativity and GBC individuals;
Apart from the fact that individuals have additional switches, they are functionally the same.
Now we just need to adjust the temperature.
Since the two layers press I mentioned are all controlled by a simple thermal switch, all we have to do is replace it with a higher temperature rating.
This is very simple.
When the roller is too hot, the switch contact is turned on, the heater is turned off, and when it is too cold, it is turned on again.
This is a very simple lag phenomenon and the heater is never the right temperature because it is almost always slightly higher or lower than the right temperature to keep /-
About 7 degrees.
This is very different from the Watlow PID Temperature Controller I used later in the reflux furnace, which understands the system and dynamically compensates.
You want to try to find a variable hot switch instead of a fixed value switch.
Like a Heat switch with a fixed value, it uses a metal strip that changes as the temperature changes.
The two metals have two different thermal expansion coefficients, and the whole block is bent forward at the appropriate temperature to close the contact.
However, unlike the fixed value switch, it has a screw that can be adjusted to change the tension on the metal strip, so change the distance it needs to turn off the contacts and turn on the heater.
This is very smart, simple and perfect for our laminating machine;
We can calibrate it according to our wishes, and even match it to the melting point of the specific model of toner for the printer we use. Huzzah!
Can't find the adjustable switch, don't worry; a 160-
The 190 ° c fixed value switch works fine.
We will have to replace the low temperature thermal fuse for safety (
Instead of bypassing it completely.
Definitely not recommended! )
With a higher rating, so if overheating, the whole unit will lose power and hope it won't catch fire.
By the way, you risk doing this mod yourself, so don't leave it unattended!
A little dangerous.
At the new temperature, it may initially have a little smoke when worn out, but more importantly, it melts and you need to lower the temperature. :)
My laminating machine is set at about 180 °c (356º F). It varies +/-
About 10 degrees, test directly with convenient thermocouple on the roller, be careful not to let the probe be pulled into the laminating machine, of course.
For good safety margin, the fuse is 240 ° c.
You may have noticed the trend of automation, which I do not deny.
The whole project is an attempt at small scale mass production.
So let's start: Don't weld every board by hand, hey, guess what-
Can I punch in the tin paste, apply on the real good stuff, and put the whole thing in the toaster for a minute or two?
Yes, that's the general idea.
The part where surface mounting technology makes this possible is the choice of surface mounting components instead of the traditional pass-Hole Assembly.
They are much smaller in size, lower in cost, no drilling, and the size of the plate is reduced.
Any of their shortcomings are side effects of their merits;
They are easily too small to operate without the right tools.
Typical SMT manufacturing works like this: solder paste spreads to the pad of the circuit board through the template.
The assembly is then glued to the paste, and depending on the welding and heat resistance of the Assembly, the entire assembly is raised to several set temperature points at a set speed.
In our case, each component can be baked except for two;
The battery and switch need to be welded manually later.
In any case, the long-term advantage exceeds the higher entry cost, so it's time to build a temperature-controlled re-flow furnace so we don't need to have the wrist tube weld hundreds of small sand.
The price of the controller must be several thousand dollars. The cheapest commercial reflow oven is definitely thousands of dollars.
On the other hand, the cheapest independent controller is in the range of $100.
I had to build it myself because the expensive ones were too expensive and the cheap ones were too expensive.
All the controller has to do is monitor the temperature of the oven and change it when it needs to be changed.
The ideal way to achieve this is to use a ramp/soak temperature controller for factory automation.
I was lucky enough to get a Watlow SD6R-LCJE-AARG ramp.
/Less than $50 soak controller, I built the whole thing around it.
You should be able to find something similar within the same price range without being too much trouble.
Just make sure it can raise the temperature to multiple set points instead of simply maintaining a stable temperature.
Walow is quite tolerant;
It will use 24 AC/DC as a power input, can display the temperature of Fahrenheit or Celsius, and can accept almost any type of thermocouple.
More importantly, it has a low voltage DC as the control output and we will use it to drive the solid state relay, which will turn on and off the heater of the standard toaster oven.
The toaster oven will work, although the smaller the better, because the smaller the air volume means it can be heated faster.
Try to find a new oven with a quartz heater, but the ceramic heater is also suitable.
You'll be fine if your oven has a continuous running setting like baking.
Otherwise, you need to turn it on and wire the heater so it is always on
This is a dangerous thing I don't recommend.
You also need a thermocouple with heat.
Since our oven will reach a temperature of 260 °c, it is resistant to insulation.
This brings another fire hazard, so don't leave it unattended, don't leave it away from the wall, and don't set anything on it (
Unless there are heat-resistant plastic feet! ).
The last thing is to simply drill a hole large enough on the too large back to get the thermocouple through.
There is no need for our oven to worry too much about heat loss, but the holes should not be much larger than the thermocouple.
On the other hand, too small holes and fiberglass insulation on the electrical line will start to wear from repositioning.
The holes should be placed slightly above the center rack of the oven so that the thermocouple can touch the surface of the circuit board.
Reference parts stools and suppliers now we need some way to get the part to the PCB quickly and efficiently, it must be a vacuum pen when the tweezers do not cut the part.
Dissatisfied with the old springin-a-syringe trick?
The modified aquarium pump can do the job well.
I found that the best aquarium pump for our purpose is Tetra Whisper.
With a few exceptions, only the diaphragm pumps five or ten years ago are easily reversible, and they all require some kind of permanent modification of the valve system.
The most modern pump (
Like Pier 100)are single-
Part design that has been improved to a point where it cannot be modified in the same way-
In addition to the whisper series!
The way it is designed, just pull one out and it can be reversed
Check valve assembly, turn it and put it back in place. It is a tool-
By performing the mod in reverse, you can resume the free modification of the normal operation.
If you want to take a cheap route, it's still a safe bet on the old pump.
Any thrift store (
Goodwill, depositors, etc. )
There's less than $5.
We also need to make a vacuum pen.
Any cheap pen with plastic tube body will do a good job.
Pay attention to the standard Luer lock connector for syringe needles.
They came in handy for a lot of things.
In this case, it will make it easier to swap tips.
It is difficult to find them;
I have a quantity of remaining shops in my local area and it turns out to be ordered by a doctor (groan).
I even found a Luer lock valve for adjusting the airflow, although it turned out to be completely unnecessary as we wanted as many vacuums as possible.
I also managed to piece together a micro Lazy Susan with some weird parts.
Considering being able to turn the board in the direction of the part you picked up, I find this an indispensable tool.
Not even the best clips --
This is really the most time-consuming step;
It is to design and plan the whole equipment.
The goal is to make a 5-bit LED micro-controller-
The number of parts with small enough to fit a circuit-based 2 "by 3.
5 "standard business card.
To reduce complexity, it must be a single panel and must be built entirely with surface mounting components to save drilling time and keep the profile thin as much as possible, and prevent the end user from being poked by the lead that sticks out at the bottom.
Most importantly, it needs to be cheap and fast.
Again, this means reducing the number of parts and the installation of components on the surface.
I ended up designing the whole thing around the PIC12F508 chip, which is the smallest and cheapest micro controller I can find, with the right number of pins.
It's not surprising that the PIC12F508 is not interrupted or PWM, but it certainly does the job.
At the time of writing this report, they ran about 46 cents per person.
A similar chip can be bought from ATMEL, which is located under the banner and is more fully functional, but costs about twice as much as it does.
This is important in terms of the quantity I want.
The PICF508 operates nominally at 4 MHz on its internal oscillator.
It has only 750 bytes of program memory and 25 bytes of RAM, but we don't need that much anyway.
It also has no interruption or PWM, but, again, we can pass without these.
As far as I/O is concerned, it is enough.
There are 8 pins, 2 for power supply and 6 I/O pins, one of which is for input only.
We already know what that means: it will have five LEDs and one sensor, that's all.
The most important feature is that it is programmable. circuit (ICSP)
, Which means you don't have to take it off the blackboard in order to re-program;
An important point when you are dealing with welding on SMD parts.
When running the program, the pins 2, 3, 5, 6, and 7 of the PIC are configured as 3 v LED outputs, pins 4 are configured as inputs, and 1 and 8 are 3 v DC power inputs.
In ICSP mode, PIC is programmed through five pins: pin function 1 Vdd 2 (unused)
4 Vpp 5 (unused)
ICSP clock 7 ICSP data 8 makes the circuit simple to a certain extent.
In addition to the five LEDs and their current limiting resistors, a tilt switch that pulls pin 4 (GP3)
When the board is tilted, there is a bypass capacitor on pins 1 and 8 (. 1 uF C1)
To prevent unstable CPU behavior and an instant power switch that prevents the battery from running out, it makes no sense.
It's simple, as promised.
With the right program, it can do anything!
The only limit is yourself!
It took me a bit of time to work on other components;
I ordered about 20 different tactile switches from DigiKey to find the thinnest, best looking, softest tactile switches possible.
The battery will overheat (
And potential explosions)
I play with conductive epoxy and conductive tape if welded, bad luck.
Finally, I bit the bullet and paid 26 cents for a beautiful, sturdy SMD battery fixing clip.
They eventually got into the design very well.
Now all we have to do is focus on the artistry of the PCB layout and we have done it.
Ironically, I put more time into scale and layout than anything else on this project, and got three sentences of description.
Well, after all, things are really beautiful.
The final design does leave room for improvement.
On the one hand, it needs brighter LEDs so that it can be seen better in bright light.
The tilt switch also makes it look a bit cheap. g.
"What's that thing calling around? Is it broken?
"Other flaws, like having to wave to see anything, are inherent in the design and were accepted even before I started.
By the way, if you don't explain how it works, it will end up being an intellectual test, but there is an obvious advantage to that;
Only someone who can figure it out will call me. :)
To get the ICSP to work, there can't be any interference components on the pins it uses, and I 've had a lot of trouble with that.
After modifying the original copper trace design, rearrange the ICSP pad (
But still exactly matches the schematic! )
I found that I can't program anymore.
The simple solution is to remove the capacitors that connect pins 1 and 8 and now work fine.
In any case, components are not indispensable to the design.
Now we need to use it.
This is the first step in actual PCB production;
The rest is just foreplay.
Find the Right Transfer paper and the first step is to print the transfer to the right paper.
If you have enough money, you can make commercial products with Pulsar Pro, about 10 cents each, or if you feel frugal, you can take the DIY route, which is free of charge
I don't know why I'm questioning that though. Sorry.
Finding the right free newspaper can be tricky.
These two key features are that it is able to maintain Toner when printing, and it is able to release Toner when heating.
One predictor of this is how smooth the paper is;
A smoother, better release, less gloss means a better hold.
We need a medium of happiness.
People on the Internet will recommend a variety of strange store brand glossy photographic paper that has been discontinued for a long time or must be imported from overseas.
Some people have successfully used the wax backing on the shipping label and it is recommended that you buy them and peel off all the labels (
Or use all tags and save the backup).
Also, most of the paper has to do with the need to rub, peel and soak the board, just to get the paper off. Forget that.
If I'm going to compromise, it has to be actually free, not just cheaper than what's real.
I tried a lot of things.
It turned out to be a spam trick;
Menard's coupon email gave me the best results.
In the end, I don't get spam anymore, and there are only a few odd-sized pages per mail, so I have to find a more consistent sheet of paper available.
But I know where to start.
The best simulation so far is the shiny apartment they rent at the grocery store.
Look at the free literature rack for something similar.
It works very well;
It has no loss of toner when printing, and it will be released immediately when you run under hot water. Presto!
The paper is too thin to go through the laser printer alone, so you'll need to make the carrier with plain white paper folded and wrinkled to the top of one side.
The laser printer at this point, I should mention the printer itself.
In most cases, the resolution of the printer limits your final resolution, Motherboard quality, and minimum functional dimensions.
In all other ways, any black and white laser printer can work like any other printer.
Toner density should be set to maximum level, hope you have a printer with a single Toner
The page slot in it so you can enter the page manually.
If you can figure out how to load your special paper, you can even use regular inkjet as a copy source;
Any method based on toner is OK.
I found it too late and one thing to note: It is said that the standard brother brand toner melts at a higher temperature than other types of printers, so, I had to buy an off brand refills cartridge using more common toner to get consistent results.
But it's still good so far.
This is a cheap solution because I have to buy itbrand stuff.
Transfer the pattern to the PCB and you will find that you need very clean toner to stick to the copper.
Even fingerprints cause problems with the final adhesion, so wear cotton gloves.
Acetone, even just soap and detergent.
When it is clean, wipe it with a lint-free cloth and let it dry.
As far as FR4 is concerned, my existing PCB is very thin. 020 in (20 mil)
This is exactly what I want.
The thinner PCB not only goes through the laminating machine, but once the transfer is printed, it can be a translucent standard PCB, you need to align the pre-cut PCB with the alignment mark on the transfer paper, the toner side of the copper.
It's really helpful to have a light table here so you can see it through the transfer paper.
Fold on the edge of one side of the transfer, crease it, make sure it does not move, and put the end of the crease into a preheated layer Press.
Even though I use a very thin PCB, I didn't notice any difference in adhesion based on transfer paper on the bottom or top rollers.
For thicker pcb, you may notice the difference.
Technology of double-
The side PCB is similar but the alignment is tricky.
Instead of leaving the margin to fold on your design, you print two layers of pattern side by side, and the gap is the thickness of the PCB.
Then, when clipped on the PCB, the two sides should align exactly the opposite.
The board passes about 1/4 "every second ".
I usually run three times and don't have time for the board to cool down between two passes.
Of course, the number of passes will vary greatly due to the thickness of the plate, the type of Toner, the type of paper and the setting of the laminating machine.
Cotton gloves also help to remove the hot plate from the laminating machine.
Now go to a sink where hot water is ready and waiting.
The paper should be peeled off immediately after about 10 seconds, although finer details require a quick clean-up of the remaining pulp.
It's ready for etching!
Tools and suppliers are the most dangerous step.
This can be very dangerous without proper precautions. Be careful!
You take the risk to do it yourself, so put on the right protective equipment (
Goggles, masks, rubber gloves)
Never do this near the kitchen.
It's time for magic to happen!
This is the part where we're going to break that bad etching tank we made before, side by side out hundreds of circuit boards.
The key factor in this approach is some kind of etching agent.
As mentioned earlier, I decided to make a decision in cul2 (Copper chloride).
It is reusable and very cheap and can be made with ready-made chemicals.
This is a good alternative to iron chloride, which is a common hobby that is not easy to regenerate, and is very expensive for products that are used at once.
Surprisingly, it is quite commonly used in the PCB industry, even more often than iron chloride.
Unfortunately, there is not enough space here to look at any major details of chemistry, which is described in more detail in both links, so I'll go through most of them with ease.
Just say that the article quoted at the bottom of this step should be your primary resource is enough.
In short, given the right amount of oxygen, HCl (
Copper and water, we will have a lot of etchant.
The reaction we would like to have is: Cu 2 HCl · h2o--
> Cul2 2H2O copper originally came from the board itself.
Hydrochloric acid is a common pool cleaner that is sold in the form of Muriatic acid. Mine can as 31. 45% (
20 degrees, or 10 degrees. 01 Molar).
High concentrations of hydrogen peroxide can be found in health food stores.
You want 35% of the food grade (11. 6 Molar);
The 3% kinds you get in the first aid department won't cut.
You also need a pot of distilled water because the tap water is too impure.
I only need 1 for my 2l tank.
The monthly rise of rot.
The ratio in the formula above, assuming that my etching agent should be 3 mol of HCl in the final volume, I came up with it. 48L HCl, . 92L H2O, and .
My reagent is hydrogen peroxide of 20L.
Add water, then peroxide, and then HCl.
Always add acid to the water, not the other way around (
As the mnemonic says, "What you should do, add acid to watah [water]").
Use pyrex plates outside.
At first, it produces quite intense heat and smoke, but when you start using the solution etching stereotype, the smoke decreases when it becomes copper chloride.
Note: Do not mix your etching agent in a homemade etching tank.
Never etching too many plates (side by side)
Or, like me, mix raw copper into your etching agent, especially in a closed sealed tank.
This reaction can emit intense heat and produce sticky smoke.
It will melt, break or open your etching tank from pressure and heat, which is far beyond the pressure and heat produced by the commonly used controlled stable etching agent.
With enough air bubbles, you don't need to mix in copper at first anyway.
The solution will be implemented effectively.
After mixing the etching agent, it's time to etching!
If you build etchtank carefully, all you need to do is slide the pcb into the clip on the lid and put it into the tank.
The etching progress can be observed through the window on the side.
The hangers I am currently using need some improvement as the circuit board needs to be turned off and replaced in the middle of the way to etching the parts covered by the clip. -
A initially, in less than A minute, the circuit board will be rapidly etched and the etching time will increase to five minutes or more as the solution begins to saturate.
Once the time becomes unreasonable, you just need to add the right amount of acid and/or run the bubbles for a while to regenerate the solution and reduce the etching time.
Again, consult the above documents on how to do this correctly;
This is a good reference.
When you see that the plate is completely etched, remove the cover and pull out the plate with a plastic clamp.
Let the excess etchant drip into the tank and rinse the board in the plastic tub.
It will then be placed on the Scottish Brite pad and some acetone to remove the toner layer and the traces of copper should become bright.
When the tank is not in use, etchant should be stored in closed glass containers that are not used for food, lack of light and properly marked.
Etchant should not be poured into the sewer in any case because it is toxic (
High copper content)
Corrosive to metal
If you do have a leak or need to be treated, the solution can be neutralised with lime and mixed with concrete.
Be sure to handle it properly according to your local hazardous waste policy!
There is a facility where I live to receive such garbage for free. It is O. K.
If some flushing water eventually appears in the sink, let's say you don't have a copper tube and you'll let the water flow out of the system for a long enough time.
I even covered and re-used my flush water to treat everything in contact with acid as a hazardous waste.
So don't really pour anything into the sink.
If you are so eager, it is easy to make a thin layer of tin on the surface of the copper trace.
In addition to welding masks, most commercial PCBs are equipped with at least tin plated pads to prevent short circuit of traces (
Although other colors are used sometimes, they are usually unique Green).
However, in our case, tin-plated plates do a good job to prevent corrosion and improve the solderability of the joints.
Chemical plating solution is a very good result I got with Tinnit, a powder "just-add-
Water "plating solution.
It is quite cheap, including shipping costs, about $10, with a shelf life of 6 months, and can lay a lot of boards before it needs to be replaced (
475 ml pint for up to 600 square feet.
Inches of copper).
It works by chemically depositing a very thin layer of tin on the bare copper of the circuit board.
This is the same result as the plating method and is simpler and cheaper.
The traces of copper must be clean oxidation or discoloration as well as etching agents before you can plating.
If you just etching it, it should be clean from removing Toner.
If not, first you need to clean it with a Scottish cream pad and then rinse it quickly.
You also need some way to heat the solution to 50 °c and a container to heat it.
I came up with a stirred hot plate which definitely gave me a good start.
This is a type of motor that installs the heater and magnet below so that the solution can be stirred by a special glass coating
Magnet at the same time
If you can't find a good baking tray, the kitchen stove is also available.
The plating container was found, and the tricky part was to find a non-
Metal container for electroplating.
A real lab might
The shock-resistant glass tray is also called Pyrex.
You thought you could use your casserole, but it was a very bad idea.
What most people don't know is that "Pyrex" in the kitchen is not the same thing;
The name of Pyrex is actually licensed by Corning Glass to a company called the World Kitchen.
They use a completely different glass called lime glass.
It is well known that when exposed to the temperature difference, it can be broken catastrophic, such as on an uneven stove.
So don't do that.
However, there is also a cheap option.
In my research, I came across an early Corning product called pyroceram that was sold in the form of "Corningware.
It's not glass, it's actually a glass.
Ceramic, very sensitive to heatshock. Real hi-
Tech stuff, Corning actually made casserole with it!
You can buy them at a depositor or goodwill for less than $5.
00, complete with cover.
Looking for unique white and "Corningware-
There is the "Rangetop use security" tab at the bottom.
I also need a small harness to hang the board on a rotating stirring magnet at the bottom of my plate.
Safety Note: Tinnit contains acidic tin salt, so wear gloves and goggles and don't reuse any food to eat.
Follow the warnings on the package. Please!
Tinnit is a powder that needs to be dissolved in hot water first.
After the following instructions, the heat resistance is 475 ml (1 pint)
Distilled water at 55 °c (120º F).
Once it is completely dissolved, it will be a milky white
Some solid yellow is floating around.
In the solution, give it 10-
30 minutes, stirring every 5 minutes.
If everything goes as planned, you should see copper starting to turn into a dull gray almost immediately.
The longer you put it in, the thicker the plate will be.
Just don't heat it over 140 degrees F or it will start releasing strange ammonia and sulfur dioxide smoke.
Once you are satisfied, pull the board out and rinse it thoroughly.
The instructions say that it should be cleaned up with ammonia to remove corrosive residues that will make the circuit board dim.
Use ammonia as a window cleaner at a critical moment.
Then illuminate it with a scratch-free Scottish Brite pad and it will be welded!
Tools and suppliers use metal welding templates from pcb manufacturers, etched in stainless steel or aluminum, for high-volume circuit boards, but you can order them from thick mai la online.
If you have a laser cutter (
Laser cutting machine No. 2
Or vinyl cutter, you can cut your own thing with acetate or transparent film.
In a pinch, you can even print out your pattern on transparent film and use X-
Acto knife cutting pads. Whatever works.
In my case, I had a friend cut me a template with his vinyl cutter with thick acetate.
Some adjustments need to be made in order to set it up correctly, as it is an unusually thick and brittle material;
Thicker than most wines.
In the end, we determined that it would take at least four passes for the tool to be cut all the way.
For the actual screening, I made a fixture with a thick silicone rubber.
I cut out the exact size of a piece of PCB, and then I piled up the extra PCB under the frame, resulting in a cut that was accurate to the depth of the board.
This provides a surface for the template so that it can lay flay all the time while the scraper is dragged and keep the plate in place.
Arrange your template and tape it down at the top with packing tape.
Be sure to wear rubber gloves because in my case solder has a lot of horrible things, lead.
Dig out a drop of solder paste, place it on the scraper, or place it on the top of the template, above the cut.
Tin paste should be sticky but not too sticky to spread. It's a non-
Newton material, which means that it flows more easily under pressure and will solidify when the pressure is released. Cheap decent-
Premium solderpaste and flux are available from DealExtreme but you should be ready to slow (but free)shipping.
If the solder paste is old or just too sticky, you may need to dilute it with a flux.
Hold the bottom of the steel mesh tightly and pass the scraper through the steel mesh, be careful not to lift it from the surface.
Repeat until all areas are filled and peel off the template carefully.
This part definitely needs to be practiced and you may need to do a few attempts before you get it right.
Now just put the board on the Lazy Susan and pick and place the right parts with tweezers or vacuum pen.
Then go to the reflow oven, where you wish not to hit the board, not to spill the parts you carefully placed on the floor, because it sticks to the paste that is still glued to it, so you have to go from the parts
First of all: it is important to configure the reflux furnace correctly, otherwise this will not go smoothly.
This is all about setting point parameters.
The maximum temperature and time to maintain should not exceed the time that the most sensitive part can handle, suggesting that it is actually 5 degrees lower than the value.
The most sensitive component in this case is the surprising LEDs.
Based on its volume, the micro-controller can withstand temperatures up to 260 ° c, and the tolerances for resistors and capacitors are higher than this.
The recommended maximum temperature of the Led does not exceed 225 °.
The manufacturer kindly provided the return slope so I just used this slope.
If your oven is heating slowly, you may have problems near the peak of the return ramp.
The idea is to slowly raise the temperature and raise it to "re-
Flow "welding and then drip it to room temperature quickly but not immediately.
So if reaching the peak is an issue, you may have to step in and open the door of the oven to speed up cooling from the peak temperature.
Once you program the oven well, place the plate in the center of the shelf and make sure that the thermocouple is close to the PCB surface so that it can get a reading close to the actual situation of the connector.
Close the door and start the program.
Of course, there are fire extinguishers nearby.
Wait until the temperature drops to room temperature.
Take the board out because if the solder joint denies its cooling cycle, it will surround the solder joint.
I should point out here that if you don't need to make a lot of the same pcb, you really don't need to make a reflow oven or filter solder;
It is very possible to weld parts by hand.
However, you need a relatively fine soldering iron.
The trick is to tin on the mat on each assembly and fix it in place with tweezers (
Or vacuum tools)
Stream the soldering tin with the iron on that joint.
It is then safe and perfectly aligned with the other side.
The micro-controller can be programmed in many different ways.
The traditional way to do this is zero insertion (ZIF)
Socket, the chip must be removed from the circuit board for programming.
Some chips can be in-
Circuit serial programming, as you collect from the name, it can actually be programmed when the chip is still on the motherboard.
This is very important for our small business card as it will be a tiny surface mount chip that cannot be placed on any normal socket.
Also, we will have the ability to re-program at will without having to do it every time.
With PIC, the programmer puts the chip in high-voltage programming mode, and it works surprisingly, assuming that the rest of the circuit is relatively isolated from the programming pins.
In order for the ICSP to work, there needs to be some way to connect the programmer to the correct pin on the PIC.
Most permeable PCB designs use connectors welded to the board, but in our case we need cheap flat solutions that don't require connectors.
This is the source of the programming fixture.
I had to make several programming fixtures during the project, depending on which stage the prototype is in.
The first simply uses a scrap prototype board with an SMD pad for IC and wires issued from the appropriate pins.
Then I just position the chip and hold it down with my finger while programming, be careful not to short-circuit the lead.
I also try to place magnets under the board to hold the chip in place.
The final solution relies on a smart little mechanical connector called the "pogo pin.
"It's essentially just a spring --
Load pins that form good temporary contact with the board.
They are a little vulnerable but work well.
With this in mind, this will help if the board is planned like this, as you need to connect to the contact pad on the appropriate pins.
For ICSP, it is important to isolate the programming pin from the rest of the circuit and connect directly to the PIC as it may interrupt the programming signal.
There are simpler ways to do this, and of course there are more lazy ways, but there is nothing better than a dedicated fixture.
The one I built uses a hinge-type clear cover and the pogo pin is mounted in acrylic.
The lid is locked with two rows of nd magnets and the pogo pin is tightly pressed on the programming pad.
It also presses the instant power button as the PCB layout changes after the programmer builds;
To connect pin 1 to the Programming Board, a switch needs to be pressed.
The base of the unit uses a piece of black acrylic scrap, and the lid is a piece of transparent part that helps align the connection when inserting the plate.
Where it needs to bend.
The PCB is firmly fixed in the right position, two PCB tracks are cut to the right length, there is a gasket insert, and when the PCB is pushed in, it raises the board to the right height, and stop it in the right position.
Using the hole saw, I cut a half circle from the Walker and the black frame so that the card can be easily caught to take out.
In order for the lid to be aligned parallel to the rest of the unit, it needs to be bent at a slight angle.
It's easy to do this with a spray light.
The lid is attached to the black frame on the hinge, passing the torch back and forth from about 2 inch away.
Be careful not to heat too long or too close as it will burn so practice on a piece of scrap first.
When it becomes plastic, bend it on the card guide until it is flush with the top of the Card Guide.
Keep it in place until it does not move and let it cool.
Be sure to do so outside or near the window as there is a chance of smoke when heated.
The holes used to install the pogo pin are drilled into the unit cover using a drilling machine.
I made a template for the hole and printed it out on the transparent film.
Slide them one by one and make sure it is aligned with the right pad on the board.
After closing the lid, press the pin to the depth of pushing the spring-
Load poker a little more in the middle.
Then mark the depth with sharpie and add a drop of solder paste on both sides of the pin.
It should not move up and down, adding a small amount of glue to prevent it from swinging in any direction.
Connect the ICSP wire to the pogo pins, and the pogo Holder starts counting from the marked red line and connects to these pins on PIC: icsp pic pin 1. . . . . . . . . . . . . . 4 2 . . . . . . . . . . . . . . 1 3 . . . . . . . . . . . . . . 8 4 . . . . . . . . . . . . . . 7 5 . . . . . . . . . . . . . .
Once done, connect the programmer, slide on the filled board and remove the lid.
Ready to program!
Be sure to remove the coin battery from the stand, otherwise the programmer will be finished;
There is no protection on its output pin, so any external current can damage it.
A simple solution would be to put a bumper on the cover of the programmer, which would prevent the lid from closing if it hit the battery.
The card is not turned off until the switch is held down in an instant.
When powered on, the program simulates a small ball with a binary counter that goes in and out with the card's tilt.
After tilting around four times before and after, it switches to the persistence of the visual mode and loops through a series of programming modes.
At the moment, it's talking about my name, then a neat square wave pattern as a separator, and then my phone number.
When the tilt switch is off, the program recognizes it as the beginning of the wave and plays the current pattern at once, and then waits for another wave to start.
As long as the waves don't double too fast and the pattern won't be too long, there won't be overlap because the pattern will be blank before resetting.
Each pattern is an array of Bytes, each of which is a column in a sequence.
Time per column (frame)
Remain open in the pattern, the time between frames determines the speed of the display, and the interval between frames when the card is waved.
Basically, it changes the width of the character and the entire pattern.
In addition, the speed at which you wave the card also plays a role in the spacing, but not enough to make it illegible;
Since it changes the frame at a constant rate, this change is not important, but enough to make it look vibrant. Neat, huh?
To create the schema, I modified a script written by Andrew Mason.
My version is custom to allow alternate byte formats and number of LEDs, but the position of the LED is fixed for the time being.
As long as your pattern is small enough to fit into the picture memory, you can make any pattern you want.
The rate of mode playback can also be changed in the code.
This program is very simple.
Without PWM or interrupts, I can't do a lot of tricky things even if I want.
Still, space saving and optimization make it more interesting as the code needs to fit into the tiny memory in the picture.
To play the pattern, PIC just check if the tilt switch is activated and then cycle through the pattern.
It sets the GPIO output to the current byte in the array representing the frame, waits for the set period, and then increments the array.
After the pattern is played all the time, it will empty the display and wait for the tilt switch to be activated.
This shows that another wave has started and it goes through this pattern again.
The frame code looks like this without quotation marks: "0bym0000 ".
0b depicts a byte, and the comma separates each byte from the next.
This is all placed in a constant array, as it will not change during playback, which will save RAM.
The five bits in the array represent the LED output, while one is not used and the other is read-
Bits for input pins only.
The bits used are: 0bXX12X543, where the numbers represent the pins.
The code includes all the code is written in Notepad, this is a good free moreLanguage Editing.
You also need mplab ide with HI-copyTechnical C compiler.
Both products have a free version.
You need a copy of the chip PICKit 2 programming software.
After connecting the chip and inserting iCP01, set the device family to "baseline" and select PIC12F508 from the device drop-down list.
Then load the hex file and press the write button.