Gate Monitor Printed Circuit Board Complete

It took all of my spare time this weekend, but I got the PCB printed and populated. It looks pretty good:


It’s getting late so I won’t test it until tomorrow. Hopefully there are no really bone-headed mistakes.

My first attempt at the board failed miserably. It etched OK, but when I started trying to solder components, I was having various problems such as pads literally melting off the board, and traces breaking. I think after I had 4 messed up traces I gave up, very unhappily so.

I started thinking about it and I had the same problem with pads separating from the board due to heat before and I had been using .02″ traces. CRUD! I had built this board with .02″ traces for reasons I couldn’t say because my notes to myself say USE .03″ traces! Dang it.

So today I started over. I redesigned the board with wider traces and clearances. I then etched two of them JUST IN CASE. That is now part of my notes as well. If I have the materials available, just build a spare. It takes about 3 hours to go through the etching process.

This time I had no more trouble soldering the board than usual (my technique is just horrible – I really need to take a class or something).  I also remembered to use a flux pen on the pads this time and that seems to have helped a bit as well.

On the plus side, my toner transfer process is working fantastically. I have had absolutely no problems transferring toner from paper to copper and etching. I even figured out a little more about balancing oxygen and acid during the process this time and updated my notes accordingly.

Here are some pics of the board with .02″ traces, then the two with .03″. You can barely see the difference, but when it comes time to solder components on, it makes a big difference.

P1020734 copy


This entry was posted in c-arduino, Gate Sensor. Bookmark the permalink.

2 Responses to Gate Monitor Printed Circuit Board Complete

  1. Dan Theman says:

    Checked the solder points this morning with an ohmmeter and found 2 that were bad. Fixed them, wired all of the switches/sensors in, powered it up and it worked on the first try!

  2. Sean Straw says:

    The reference to oxygen implies you’re probably using a Cupric Chloride etchant (Muriatic Acid + Hydrogen Peroxide). Am I right? Are you refreshing using Peroxide, or a bubbler stone? If using a bubbler stone, are you just running air through it, or are you driving oxygen (say from a Bernz-0-Matic cylinder)?

    3 hours seems a bit excessive though. A slight amount of heat (of the solution), plus some agitation will speed up greatly. By using the filled zones as you have, there’s very little to etch away, so the oxygen in the etchant shouldn’t be depleted that rapidly. Cutting the PCB to just the masked area – not having a large surround that needs to etch away – might help.

    Are you using a packaged tinning solution or rolling your own? I recently decided I needed to up my homebrew PCB game and start tinning all my projects, so I ordered the individual chemicals for a tinning solution, and now measure off and package so many “doses” of them into old plastic film canisters. When I need a fresh tinning solution, I heat up 100mL of distilled water (a quick hit in the microwave), dump a canister of pre-mixed powder into it and stir it up, and I’m ready to go – the heat helps to rapidly dissolve the solids, and it also accelerates the plating. Because a dose is 0.5g of one of the components (2g and 3g each of the other two), I use a jeweler’s scale to measure them out. One dose will tin several small projects, but once introduced to water, it has a limited shelf life. It is a lot cheaper than “Liquid Tin”, and I’ve been quite happy with the results.

    I also found that a desktop beverage heating plate (a teflon-or-similar coated surface a bit larger diameter than a coffee mug) works great for keeping the solutions warm (which increases their effectiveness, and thus reduces the time to etch and plate). Something like this:

    I keep the lead clippings from through-hole components in a (very) small plastic bin at my workbench – they’re useful for a variety of things, not the least of which is beefing up a crappy trace – tin the piece scrap lead, cut it to length, set it down on the trace, and run the soldering iron over it enough to flow the solder between it and the trace. All good. Not just for crap etchings, but also for beefing up (within reason) the current capacity of a trace.

    I realize you’re done with this project, but on the next project involving an AVR processor, you could replace the Arduino Nano with an ATTiny (or an ATMega, if you really need it) and a 6-pin ICSP header (2×3 berg header), which would permit in-circuit reprogramming of the uC. For one, you’d have a lot less drilling to do. You can drive the ICSP from an Arduino, or using an inexpensive device like a USBASP ( a USB stick incorporating an AVR uC loaded with a programmer sketch and sporting a 6-pin cable to plug into the header on the target). Another not insignificant plus: some models of ATTiny processors can be had for < US$1 apiece in small quantity (< 10, not even 100's). The ATTiny84A with 8K of flash, and 512 bytes of SRAM and EEPROM, and 12 I/Os is US$0.83 at qty 25. Even a full-blown ATMega328 (if you need the added I/O or flash space) can be had for ~ US$3.25 at qty 1 ( and circa US$2 apiece at qty 25 – check Mouser). Project cost id driven way down when you're not shelling out $25-35 for the brains. The uCs are so cheap, one doesn't need to hem and haw over whether or not to craft another project that uses a uC, or making a backup, or one to give to a friend or neighbour.

    The only real downside to using an ATTiny in the project instead of an Uno/Nano, is that serial debugging doesn't "come for free" – you have to design that into your project if you need it. However, if you use a DIP for the uC, you could pull the uC and plug in an adapter from the appropriate I/O pins of an UNO (or Nano, etc) soldered to a suitable wire-wrap socket that you could plug into the vacant ATTiny socket on the PCB, allowing you to load a bigger, badder sketch onto the Uno/Nano with serial debugging enabled. Not 100% transparent, but close.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s