So I was very disappointed when I realized that the cable to the soldering iron RT 80 is made from PVC. It’s not heat-resistant, quite heavy and most of all its flexibility is not much better than that of a steel cable … How can Ersa risk their good reputation by installing a cable as bad as this one? Especially when the former products had very good silicone cables …

Searching the web, I found many people complaining about that cable but no report of someone who had successfully replaced it.

After a few weeks of disappointing soldering experiences I decided to take the risk and try to change the cable. But compared to their previous devices Ersa made it a lot harder to upgrade this one: First of all there doesn’t seem to be a way to open the iron without destroying it. So I had to cut and attach the cable at the outside. Second, the diode plug at the station itself doesn’t seem to be an off-the-shelf one. So I had to recycle the existing one. And third, the plug has four pins instead of three. As it seems to be impossible to get a decent-sized silicone cable with four wires, I decided to go with the three wire cable from my old station and sacrifice the potential equalization. But in the end I didn’t need to, because after cutting the cable at the socket I happily discovered that the original PVC cable had three wires only.

Changing the cable at the diode plug is very simple. All I needed were some heat shrink tubes.

The risky part of the job is changing the cable at the soldering iron. The old wires have to be shortened as much as possible in order to get rid of as much of the PVC cable as possible. Fortunately the bend protection can be removed from the iron.

So finally I succeeded replacing the cable. The only evidence for my mod is a little piece of heat shrink tube. I am now very happy with my new soldering station.

Disclaimer: Do not make this kind of upgrade yourself if you do not have enough experience in such things or do not want to risk losing your soldering iron. I will not take any kind of responsibility if you fail. Also: Before cutting the cable, make sure to have a second, working soldering station available in order to be able to connect the new cable.

@Ersa: Please do not do this again! You’ll get lots of disappointed customers!

Update (20th June 2011):

I thought the plug was proprietary, since my local dealer did not have it available. However some people on hackaday pointed out that the plug is a standard DIN one.

Meanwhile I got one from a different store and upgraded the cable once more. Now it looks much better. Many thanks to all the people posting useful hackaday comments!

The original driver circuits consisted of a Zilog processor, an EEPROM, an optocouppler for driving the 42V synchronous motor, transistors, two reflective IR-sensors for segment position and a few passive components. Since I do not have any information about the original interface and protocol, I decided to implement my own driver circuit and directly connect to the sensors and motor.

So each segment has its own driver circuit, built around a PIC12F683. Cheap optocouplers MOC3023/MOC3063/IL410 are used for switching the motor. The controller has to switch the IR sensor LEDs individually, because there is only one signal output for the two sensors. There are also two status LEDs and a button for debug features and for configuring the segment address. The driver boards are connected to a simple one wire bus, where they listen for commands and write status response to.

One of the greatest problem was to connect all the devices to only six I/O-pins of the PIC. I got it done by only using one pin for the two status LEDs, using the high impedance state and multiplexing for driving both of them via one port pin. Also the motor output signal to the optocoupler is formed by the two sensor LED outputs or-ed together using two diodes. Since the sensors are only needed when the motor is on, this saved me another port pin.

The clock application is quite funny, but far too loud to make a good project. So I am looking for a better way to use the displays. Maybe some kind of build status display for our continuous integration server at work.

What is a Reverse GeoCache?

Like any other GeoCache it is a box that is linked to a certain coordinate on earth. The challenge is to find this coordinate, in order to access the contents of the box.

Other than with ordinary GeoCaches you do not need to find the box at the target coordinate: You will already have it when the puzzle starts. But the box is closed and it will only open when you are near the target coordinate. In order to find the coordinate, the box has a button: When you press the button, the internal GPS receiver will detect the current location and the distance to the target will be displayed. Of course the number of trials is limited. If you press the button too often, the box will be sealed forever.

Starting the Project:

One of my main goals for the new Reverse GeoCache was to make it much smaller in size. Of course I also wanted to add some new features. I have been working on this project for about four months. After finishing two fully functional prototype versions I now consider it “production/stable”

Technical data:

• Size: 95mm x 73mm x 53mm
• Weight including batteries: 199g
• Maximum payload size: 57mm x 53mm x 9mm
• Power supply: 2 AAA cells
• LCD: 2×16 characters

Features:

• Piezo buzzer for some sound feedback



• Each try is logged to the internal EEPROM. By using a secret key combination the logs can be shown on the display. So if you know how to do it, you can see how clever the puzzle was solved (or not). There is also a RS232 output to send the data to a computer.
• Emergency open by software: You can reset the box if you know the key combination (The piezo buzzer was a very important precondition for implementing this feature).



• Emergency open by hardware: You just have to remove four screws to be able to slide off the lid (I assume that none of the recipients will spoil all the fun by opening the box this way)
• The power consumption in sleep mode is about 310uA. 800mAh rechargeable batteries should last for at least  about 80 days, assuming many tries with bad GPS reception. Power consumption will drop to about 60..70uA with the next generation of PIC controller, which should result in more than a year of operation.

Some details about circuit and software:

• A highly efficient MAX1674 along with some SMD parts is the step-up converter I am using to increase battery voltage to 5V. The LCD and the servo require this voltage – otherwise I would have preferred 3V3.
• The GPS receiver is taken from a Haicom HI-204 SiRFStar III PS/2. It is the device that draws the greatest current. So I use a BS250 MOSFET to cut the power supply when not needed.



• The LCD and the servo are connected to a second MOSFET, so they can also be powered down.
• The controller is a PIC18F2520. I have started with a much more efficient PIC18F14K22, but it turned out that I needed more program memory to implement all of the cool features. Now I am impatiently waiting for the PIC18F26K22: With its nanoWatt XLP Technology I expect an enormous decrease of power consumption (especially in sleep mode). Also it has some cool new hardware features built in.
• The software is written in C. It used 10kWords of 16kWords available in the PIC.
• The custom PCB has been designed using KiCad. I have used the toner transfer technique to print the board.

What I might implement in future revisions:

• Adding a low battery warning by using the PICs ADC (the LBI feature of the MAX1674 turned out to draw too much current!).
• Allowing reprogramming the target location by key input or via connection to a PC instead of having it compiled in the program code.
• Using the power down feature of the MAX1674 in sleep mode in order to save even more power (requires new controller that works with lower voltages).
• And finally one of the most interesting features IMO: Adding a tiny GSM module to post a SMS each time the device has detected the location.

[Update] Project files:

The project files come without warrenty of any kind. They are free for non-commercial use.

• PCB as PDF for printing
• Complete schematics and board layout files as KiCad project
• The C18 source code. Note that the target location is hard coded in configuration.h

But now my problem is solved: I built a ring of 20 sunny-white SMD LEDs. It can easily be attached to my multi-tool and – as you can see on the photos – it always assures perfect light for drilling.
The first version last week was built with 10 LEDs only. I was not happy with the result then: the cast shadows of the drill were still distracting.

The LEDs are driven with constant-current of 20mA. 2 x 10 LEDs are connected in series. I built the power-supply some time ago for a different project. I plan to report on this 70-LEDs-Macro-Ringlight project (along with its power supply) in this blog soon, so I won’t go into more details now.

You can download the KiCad project files and the board layout as PDF. The layout should fit various LED package sizes, but works best with 3020. The board has a diameter of 46mm, which I consider a good compromise of good illumination and small size.

These days some of the coolest ICs are only available in SMD packages, e.g.

• the fast SSP FRAM FM25256B-G and
• the famous USB to RS232 converter chip FT-232.

For me as a hobbyist that’s a problem, because I don’t have the equipment to etch PCBs, and I want to develop my circuits on breadboards.

Fortunately there are some SMD adapter boards available. So yesterday I soldered two SSOP-28 chips on adapter boards. Soldering with a pin distance of 0.65mm is a challenging task – and for me it’s definitely the limit. Using good and thin solder (0.5mm) and a soldering rod with a very thin tip are two of the most important premises. Another one is a tremble-free hand.

What I do not like about the SSOP-28 adapters though, is that there are two rows of pins on each side of the chip. This is not good for breadboards. But hey: it’s still lots better than SMD.

For SOIC-8 however, I am very happy that I found better adapter boards. When assembled, the SMD chip can be used like an ordinary DIL-8. The adapter boards are available in Germany from W2micro. They are cheap, and they come complete with double sided SIL sockets and even with SMD capacitors. Thumbs up for this solution. The guys from W2micro really saved my day.

Do  you know about other great SMD solutions? Let me know in the comments.

I am still not sure what project to use my rgb 7-segment display for. There were some very interesting ideas posted to the Hackaday forum. I like matthiasrs idea most: A 2-digit clock, using the red LEDs for hours, the green ones for minutes and the blue ones for seconds.

In order to get a better idea, rather then implementing it in hardware I wrote a little simulation as Java applet. Have a look and see how hard or easy it is to identify the time

RGB 7-segment display

7-segment LEDs are available in red, green, yellow and blue (maybe even in white?). There don’t seem to be any in RGB though, so if you want to dynamically use different colors in your project you either have to use multiple devices or use a different technology.

So this seemed like an opportunity for a nice DIY project: Why not take an existing 7-segment display, remove the original LEDs and add some RGB ones?
I started with an old SL-1119 display and ordered some SMD-RGB-LEDs from a seller on ebay. The LEDs came with magnet wires soldered, which came in handy, as the LEDs are as small as 1.6mm x 1.2mm.

Step 1: Remove what’s in the way

Using my Dremel tool it was a matter of a few minutes to completly remove the back of the display, including the original LEDs and contacts.  Afterwards I sanded the display in order to have a flat basis for the new LEDs.

When light shines through the display body, one still can see what’s left from the original LEDs.

Step 2: Attach RGB-LEDs

For attaching  the new LEDs, I took transparent two-component adhesive with low assembly time. It is important to place the LEDs accurately, so I was not able to attach more than two or three of them at once.

Step 3: Attach socket pins

32 wires  (8 LEDs with one common anode and three cathode wires each) are not easy to handle, so I decided to add some kind of socket.

The result

I am pretty happy with the result:

One thing I could have done better though, is the optical isolation of the segments. As you can see from the behind, the light from each segment spreads to its neighbors. So when a segment is not lit, you can still see a bit of the light from the other segments shining through.

What’s next?

Now that the display is finished, what is left to do is to connect it to some kind of driver circuit. I am not yet sure how to handle the 24 lines. Implementing a I2C driver circuit would be nice, so I could control the display from a PIC microcontroller for example.