This board has a Beckhoff (c) ET1200 ASIC which handles the EtherCAT protocol. This board translates EtherCAT to RS232 and RS485 with the help of the LPC1114 from NXP. EtherCAT - Ethernet for Control Automation Technology - is an open high performance Ethernet-based fieldbus system. The development goal of EtherCAT was to apply Ethernet to automation applications which require short data update times (also called cycle times) with low communication jitter (for synchronization purposes) and low hardware costs.
Just finished up a 4 layer printed circuit board I have been working on. This board has split analog/digital gnd and power planes (1.8V and 3.3V). Below are two photos, where the left is a photo-mode printout of the top (component) layer and the right is a screenshot of the program I used to build it (PCB of the gEDA project). Here are the photos in larger resolution —> HERE1 HERE2
This board will allow me to control a 0-400 MHz Analog Devices DDS chip from my host computer via the FT2232HQ Mini Module which interfaces to my board via female headers. Just to give you an idea of why I would do something like this, other arbitrary waveform generators (AWG) from companies like HP or Agilent are very expensive (but very nice as well). For something up to 400 MHz you could even be looking up into the $20,000 range. This board, once its all done will cost approximately $150.
All the parts have been ordered except the for the Analog Devices chip which is on backorder till December!!! In the mean time I got the PCB fabbed using BatchPCB (see slideshow). I have virtually finished up coding the GUI that will control everything from the host. Here is a a look at one of the many tabbed windows.
This is a DDS breakout board (Analog Devices AD9958 DDS chip) that I designed. I haven’t soldered the components to the board as I accidentally broke 2 pins off of the AD9958 chip as I was desoldering it from another board to put it on this board. Will need to get another chip from digikey or maybe a sample from Analog Devices. Here are the schematics and pcb files along with some of the pictures of the project. The schematics and pcb files (gEDA) are included here: ad9958-bo1.sch ad9958-bo2.sch ad9958-bo.pcb
Screen shots of the schematics: ad9958-bo1.sch ad9958-bo2.sch
Here are some of the etched boards that I made of this breakout board using the toner transfer method with CuCl etchant.
Here are some pics of my work bench where I do most of my electronic design. More toys to be added in the future.
Tesla Coils. You have to love them. These high voltage circuits are a blast from the not so distant past. They have a large hobbiest following and its no wonder why because producing artificial lightning is just so much fun! I built my Tesla Coil back in 2003/2004 as an undergraduate for one of my Physics classes. We had a project that we could do that would give us some extra credit depending on how good it was. Well, when I came in to demonstrate it, lets just say I knocked people’s socks off and got a roaring applause. Ahhhh the good old days.
A Tesla Coil is a type of air-core resonant transformer invented by Nikola Tesla around 1891. It is used to produce high voltage, low current, high frequency alternating current electricity. Today their main use is entertainment and educational displays. The mailing list that I used while building mine is http://www.pupman.com/ which I used to make contact with other enthusiasts whenever I had questions. Below I am going to discuss some of the main parts of the design:
Here is the schematic for the Tesla coil. Its operation is actually fairly straight forward. Power from the wall goes to a high voltage transformer such as Neon sign transformer (that is what I use in my design). This Neon sign transformer takes the 120 Volt output from the wall socket and converts it to 15,000 Volts! This high voltage then energizes and fills up the large capacitor bank. The rectifier in the schematic is just showing that the energy is not allowed back into the high voltage transformer and in my design I actually have a spark gap that will arc if the voltage gets to high as to help protect my transformer and improve its longevity. However, right before the capacitor bank gets maxed out there is a spark gap that arcs and allows all the energy in the capacitor tank to drain its energy into the large primary winding or inductor. This inductor creates a magnetic field and by Faraday’s Law induces a voltage and current in the secondary winding. Any energy that wasn’t radiated out of the primary winding then goes back into the large capacitor tank as it is energized again by the Neon sign transformer.
The large capacitor bank and primary winding make up an LC circuit which resonates a specific frequency depending on the values of L and C as well as some other parameters specific to you Tesla coil’s dimensions which are beyond the scope of this discussion. So every cycle of the primary LC circuit gives energy to the secondary LC circuit. The capacitor for the secondary LC circuit is a small C and is not large like the big capacitor bank and this is exactly why you see large sparks coming from the toroids or other objects that make up the capacitor for the secondary LC circuit. This small capacitor can only store a small amount of energy compared to what is being feed into it and it has no other place to go except arcing out into the air. The voltage is literally so high that it ionizes the air and creates a plasma for the electricity of the circuit to travel along. Beautiful stuff if you ask me.
Here is the board layout and r8c/1B datasheet: tssop20-dip.pcb, R8C/1B-datasheet
In this project I created a 20-pin SSOP to DIP adapter for the R8C/1B MCU. This pcb was created with the open source PCB program. To program this MCU we are using the FT232R usb-to-serial chip that handles the entire usb protocol. This chip also has some extra GPIO pins that we can use to monitor different actions if we chose to do so.
I would like to thank DJ Delorie who is regular on the gEDA IRC channel. He was the one that actually etched the boards at his home where he has his own pcb fabrication lab. The r8c/1b chip has many peripherals to play with such as I2C, UART/USART, and SSU. Take a look at the datasheet for more details. If you are interested in doing designs with Renesas’s chips using open source tools take a look at http://people.redhat.com/~dj/m32c/
Today I took some time for myself to try and get a better grasp on Paul Dirac’s 1931 paper: Quantised Singularities in the Electromagnetic Field. At first blush this read was a bit heavy for me as my math skills have grown a bit rusty. However, today I really dived in to get a better understanding of how Dirac came about in theorising that magnetic monopoles could exist and would fit into the current theory of Quantum Mechanics. Actually, nature practically begs for them to exist. The way that Dirac derives the quantisation of the electron and arrives at magnetic monopoles seemed a bit abstract to me but the journey to this answer is quite beautiful. Due to the non-integrability of the phase of the wavefunction of the electron we ultimately end up seeing that: hc/eu = 2 (where e = electron charge, u = magnetic charge, h = h bar, c = speed of light).
The question remains for me though: Is it possible to have magnetic monopoles without electrons? What I mean by this is, in the derivation of finding that magnetic charges may exist there was a clear dependence on the wavefunction of the electron. Without that wavefunction, there would be no theory on magnetic monopoles. Some may say that magnetic monopoles could exist by themselves (that is they could exist in space all by themselves without an electron being present) but I am willing to go out on a limb and say that magnetic monopoles are directly linked to electrons. That is, no electrons present (or protons for that matter as the derivation does not exclude them) then no magnetic monopoles present. Maybe I am missing something here but the magnetic flux through the surface of the closed curve is only going to be present when the wavefunction of the electron is traversing the closed curve.
As Dirac’s derivation shows, this topic is subtle and anything overlooked is stuff that could lead to great discoveries.
Well its been a while since I have added a post to my blog so its time for a little catchup on what has been going on. The RFID research is continuing but my main task now (which I am excited about) is to help in the creation of an underwater acoustics modem that is being created for some NASA funded research. We are using an ITX Intel Atom motherboard with an Avnet PCI FPGA development board that we have plugged into the motherboard’s PCI slot. On the FPGA board we have an ADC/DAC board for the acoustics frontend that connects to the expansion slots. My job is to get samples into user space and have the board fully programmable. That is, its my job to program the FPGA and write a PCI device driver for the FPGA board. Getting this to work is going to be a lot of fun and I am really excited to have the opportunity to “slap it into shape”. Hopefully I have my way with it more than it has its way with me.
The Intel Atom motherboard is pretty sweet and is really nice for the price tag. Here are some shots of the little bugger. As you can see, the motherboard is on the bottom of the case with the FPGA board coming out of the expansion slot with its daughterboard connected to it. Alright! Enough writting….about it….time to take action.
For my Research Assitant position in the Fundamentals of Networking Laboratory (FunLab) at the University of Washington I have been setting up a software defined RFID reader who’s code base was created by Michael Buettner who is currently a CS PhD student at UW. The RFID Reader uses GNU Radio and the USRP. Now it is my job to take this setup and do further research on the MAC/PHY layers by writting custom code.
For this setup we have two RFX900 boards with two Alien RFID Reader antennas (915 MHz circular antennas) connected via coax. One antenna is dedicated for Tx while the other is for Rx. For more information on this please checkout some of the papers on the initial research that Michael conducted. Stay tuned for more research on its way.
Well I have been busy lately hacking away on the gEDA Manager so that I can try and get a stable release sometime soon. I fixed the segfault problems that I was having. The reason I was having issues is that I have two threads running in my pygtk application and I was calling “gtk” GUI code from the non-main loop thread, which is a no-no. That is all remedied now. If anyone is having issues on a similar project, I followed the first example here. Initially I had made it to work using the 2nd example in that link but since a Windows port may be something to do in the future I went with the first example since the second one is known to not work on Windows.
In any case, feel free to download and run the application. You can find download instructions here.
