This week we discovered the total cost of the ledworks become to high if we make 25 so we scaled it down to 12. this was we also didn’t need to buy extra mini arduinos and a lot of extra accelerometers. We ordered the ring magnets from germany. We ordered Lithium batteries and sparkfun charging chips for and some extra accelerometers. the rest of the accelerometers we already got. The mini arduino’s from the minor stock run on 5 volt so we needed to us a dcdc converter that makes 5 volt of the 3,3 of the batterie. For this we found out that the lilypad AA packs can also do they job. They can only suply 100 mA but it makes a big difference in the budget. so the leds will run with a transistor directly form the charging chip.  We ordered RGB LEDs for the led works and strips for the skin. The network of ledworks needs to be powered on 5 volt but it will use more ampere as the standard 5 volt adapters so we made a 5 volt supply out of a old computer supply we found in the EWI building. It woks really nice and costed only 1,20 euro for the resistor we needed. We finished the PCB design and are now waiting for the final quote.

Last week we had a workshop from Jerome Decock he tipped us to use PCB’s for the connection. on the PCB’s will be different rings that make the connection for power and communication. If two PCB’s are between the magnet, the force is to weak to hold 2 ledworks.  If we use use ring magnets the conection can be in the middle. last week we also had a meeting whit rob luxen the electronics hero of studiolap. We discovered several things:

• Charching lithium batteries isn’t easy.
• The MAX1555 chip can charge Lithium but is SMD so you need a custom PCB and it is a lot of work so solder.
• We can use DCDC converters to change voltage.
• We need FET transistors because they are are voltage controlled.
• Whit 12 volt LED strips you lose a lot of energy in the resistors.
• 3,3 volt of a lithium batterie can be to come to less for bleu leds if it is not full.
• we can use a separate LED controller to: get better dimming of the leds, get a constrant current and use less digital outputs.
• Multiple serial inputs is possible with a multiplexer chip.

We discovered ther were more problems to solve as se thought. we started with ordering the multiplexer so we could try the communication. and we searched the internet about the components we needed. For the magnets we mailed 15 different magnet suppliers in china. They all were fast with communication, they were ex students, tought we were great and prommised a good price. but in the end the had nothing in stock or there storagehouse manager was sick so they needed to fabricate the magnets specially for us. for that we did not had time and the only actual price we got was to high.

As some of the major issues concerning the software and the magnet-data connection were resolved we are diving with our noses into which components we want to make the LEDwork of. Ofcourse a lot of it has been around for months already, but what LEDs, what cups, and what finishing do we want?

We tried sandblowing the cups today and it gives a really nice effect. Because the LED’s are so bright, spots can still be seen through the sandblown surface, so we still might need some material inside the cups to deflect it. Also we shaved of the cups in angles so an almost perfect surface appears when pushing the cups together. We think this could be done easily when making a hard foam mold.

The LED’s we got from HEC electronics were really nice. We will try to find it’s Chinese counterpart and compare it to the LED’s proposed by Thomasz. We hope materials will start dropping in next week so we can finish our prototype and start testing the behaviours.