Wednesday, November 9, 2011

No Room to Work

We have moved back to the boat, but not really moved in! I 'm not unpacked. This is my desk/ work area/ computer/radio station.
The arduino/ATmel stuff mostly in the boxes to the left. Soldering iron, etc under everything to the right!

Everything that doesn't have an immediate home seems to wind up here. I keep hoping it'll only be a few more days till I can start on this pile. I really need to get the new instruments installed and I'd like to get the wind and solar monitors finished and installed.

Sunday, October 2, 2011

Friday, September 30, 2011

A Parts Order Arrives

I've been really busy lately. We moved out of the apartment and back on/in to the boat. But an order did come in the morning of the move. I only had a chance to verify that it was a fun box, before I had to close it up and get to work! It mostly was more stuff for building boards and SMD part sets plus a new Arduino clone from Evil Mad Scientists, the Daivolino. I had bought a datalogger shield back when I first started on the AmpHour project, but have not gotten around to testing it because of having to figure out how to connect it to my plain boards. So the $10 daivolino looked just too good. And it is. 20 minutes to solder up in the new panavice Jr.
A few of the parts, Diavolino, FTDI friend, 4x20 blue display, Panavice Jr, jumper wire set and 15x lighted loupe with headband.

I had to hook it to something so there was the new blue/white 4x20 LCD display! A url on the package had the Arduino LCD test suite. It passed all tests and  is simple to use. It does consume too many pins though.

The package of jumper wires are really nice to use with the breadboard!
4x20 ran all the examples
Then on to the datalogger. It has a real time clock chip besides the SD card holder, so I ran all the arduino IDE tests for each part. It too worked well and easy. One note, the CS chip_select in the IDE tests is set to pin 4 but the Adafruit datalogger uses pin 10. it only took about 15 minutes of digging to find that.
Adafruit's datalogger shield attached
So I have lots to play with and no time to play! The ACS714 current sensor below is still packed in a box, as are the 758's 100 and 150 amp sensors.

I also got a package of powdered Tinit mix to try on the next board. Also a 0805 resistor/capacitor kit, hand PC board drill, and lots of little stuff. The resistor kit I picked came as lots of unmarked paper strips in a big ziploc bag. After I ordered it I found a kit in cool little individual boxes. 20x20 hindsight sucks!

Wednesday, August 31, 2011

Current Sensor Functions

I temporarily connected the current sensor to one of the TuxGraphics 3 digit meters and turned it on. I modified one of the included sample programs for the calibration and got what I expected for a change!

8 AA batteries to a set of resistors didn't exercise the whole range, but I did get something both plus and minus by switching the battery leads.

Thinking back, I don't believe I got the right numbers, 4 ohm and 10 volts=2.5A and 2 ohms should have been 5A but I remember 2.3. Maybe the batteries died faster than I got the measurements?

But at least the magic smoke stayed inside the little boxes this time!

Friday, August 26, 2011

Beginning the Current Sensor

I received another care-package thanks to my dear sister. This hobby stuff from the cruising world is slow and painful (if you are as impatient as me). The key additions this time are below. There seem to be several versions of the "helping hand", this one is not so much in the helping department. The little screws are very difficult to turn to tighten the rods. Then it slips with the pressure of the soldering iron. I couldn't tell from the adds which were good and not. They were all about $6.99.

The LED lighted loupe is fine. As well as the flux and tip-tinner.

I found this PIC ammeter at Cool Circuits.com
http://coolcircuit.com/project/digital_amp_meter/digital_amp_meter_project.gif (copied here without permission yet).

I had bought several of the tuxgraphics 3 digit mini-voltmeters which are similar. An AVR instead of a PIC. However it used a sense resistor and voltage divider for the current mode. My setup had so much noise that I gave up for a while. So I ordered an assortment of Hall Effect devices.

The ACS174 is a -30A to +30A sensor out puting a voltage (66mv/A) so it looked perfect. Then while searching for the DIP I discovered it only comes in SOIC. I always wanted an excuse to play with SMT devices. I am hoping it will be noise immune enough to put a couple meters of cable between the sensor and the micro. The sensor needs to be in the boat's engine room and the micro, a more humane environment. Not so much for it, but for me, to read the output.

No tooling here, so here we go again with the markers and scrapers. The chip only needs GND, +5, decoupling cap and filter cap, then signal out. So a simple board.

The same as the first time. I found "Muriatic 29%". Thought I had it figured, but no! I started with 2 cap fulls hydrogen peroxide then one cap muriatic. Nothing after 10 minutes. So I added a little splash of the "Duck" again. Not much more action. I let it sit. Just over an hour and I have holes and low places where the marker was thin but the etch is finally finished.

A few minutes to solder my first SOIC. The web paegs that say "lots of flux and solder, followed by wick" seem to be right on. I just pretended the caps were SMT and put them on top too.

The cable was left over from a failed instrument system. With boat gear, every factory assumes you only install their "1 year warrented system" on new boats, only selling the system with all the wires over each time. This is a 3 wire, shield with drain. Hope it will keep the noise out.

I'm still unsure of the final micro board. This is a evilmadscience.com mini development board and a USBtinyISP and a display.


I must be on the right track, as after I got the board finished I was surfing to relax and found this from China, only $9. If they could just fix the international world-wide shipping and customs import in some countries, this all might work.

I hope to get some testing in soon. All this web posting uses up the time I could have tested and programmed!

Sitting here wondering where I might find 25 amp of current here in the temporary apartment to test with?

Friday, August 12, 2011

My First Etched PCB for SMD


Ok, so I really wanted to try some SMD or surface mounted devices. I have ordered some parts for an amp-hour meter that are only available as SOIC. That's a little hard here cause I am renting an apartment in Carmen, Cebu Island, Philippines while my live aboard sailboat is being repaired in Republic Dry Dock, Danao, Cebu Is, PI.

Cebu is 45KM away and the second largest city in the Philippines apparently has no SMD component outlet. I see young people on the sidewalk in front of grocery stores with hot air reflow stations repairing cell phone circuit boards, but no parts. Could be I just haven't asked the right question. I did find a guy selling the hot air stations and standard DIP parts, but no small ones or unusual parts. He says everyone orders their own. I got blank board from him and the parts below.

The circuit I chose to copy for my first board was: LED Tester from the Robot Room

Since I couldn't find SMD parts, I opted for surface mounting some standard parts. No drill handy. My dremel is stored with most of the stuff from the boat.

The top of the board. I laid out the parts on a small piece of perf board then copied the paths with a Sharpie. Then I tried to touch up the lines with a really fine point one.The more I messed the worse it got. The dried ink seemed to make the Sharpie point act dryer than it was and rub off what I had prior.

The bottom, just to save the etch. All I had was a double sided blank.

Ready to roll! I've got the peroxide, the HCL and a baggie just bigger than the board.
I pour in a couple cap-fulls of peroxide, cause it says Always Add Acid. That means put other stuff before the acid. Then add a cap-full of Muriatic. Notice it says "PURE" on the label.

Then slip in the board.

That's not working very fast. What could be wrong. Maybe the acid isn't very strong. So I see the "Duck" in the bathroom has only one ingredient on the label, HCL. Maybe it is a little stronger.

On a second trip to the local Gaisano (grocery store here in the Philippines) I discover that Muriatic Acid comes in 29% or 19% or mild. I now have a bottle of the 29% for the next board.
So I add an equivalent cap-full of DUCK. Much better action and turns green like the others say it should.

Looks like 5 minutes or so and I have a reasonable facsimile of a DIY PCB.



Cleaned up and I see that my lettering is not so good, but the 1/16 traces(if you can call them that ) are ok. There is a "V" and an "A" at least I thought I drew them.

The finished board. I(t doesn't look so bad with components and at a distance.


My "Workstation"! Ha! I have to share the small plastic table with my wife and her hobbies, currently sewing a quilt(but that's another story).