[quote author="kmmankad"]Hey, no its possible to run it at 3.3V logic.I had gone thru the schematics I found someplace online(cant locate em now) and ended deciding to short 'R7' (yes,usually all have the same R7,tried this hack with a few different LCDs) and you can have it going at 3.3V
I find that your method is very subjective to part manufacturing and performance of the components on an LCD module particularly the screen itself. 3.3V logic is possible for most LCDs but not for the power supply.
After analyzing your picture and the datasheet for HD44870, the resistor from the array are actually to divide the voltage supply to pins V1-V5 for proper drive waveform. In the datasheet, there are 2 types of bias which are 1/4 and 1/5 by which the former has V2 and V3 shorted while the later has them separated by another resistor. Each of them has a pot after the series of resistors which is the trimmer that we usually used for contrast adjustment. Also notice that later in the datasheet there is also requirement of more than 1.5V from VCC to V5, which means possible for 3.3V operation.
But one thing to note is that, although your LCD has 5 resistors in the array, that does not mean your LCD is made to the 1/5 bias configuration. I have a "relic" 40x2 LCD here using the original FP-80B package with 1/4 bias configuration and it has 5 resistors in the array. The first 4 resistors are wired for the 1/4 bias configuration from V1 to V5 while the last resistor is of 0 ohm and connect from V5 to contrast pin. That means if the contrast pin were to directly connected to ground, this last resistor in the series could be used for fixed contrast by applying the correct fixed value.
Here's my thought. You could be the lucky one to have LCD modules that has screens that could work at lower voltage. But the LCD modules has to be compliant with the standard LCDs out there. With 1/4 bias configuration the 5th resistor could be used to compensate for the lower voltage acceptance. From the picture of your LCD, it is obvious that there is a jumper pad for shorting contrast to the ground and from there, links to R7. I couldn't find any trace after R7 that connects into the chip under the plastic blob. So I really think that your LCD has 1/4 bias configuration with low voltage screen.
Alright, did some more readings and I found that RS485 and RS422 could actually be combined into a single adapter.
On the board design, I'm thinking of using an IDC socket underneath the board to piggy-back on the Bus Pirate. This is a problem as I notice that BPv4 hardware has different pin number and arrangement than BPv3. Should I drop this idea and make the connection using normal female wires instead?
[quote author="arakis"]TIp, you didn't have to break out all the pins that are connected to GND and VDD, VCAP, since you wont be using them[/quote]
That's a nice tip! I didn't thought of that when drawing.
My initial thought was to make a breakout for this dsPIC that is comparable to the microcontroller plug in modules on Microchip Explorer 16 Demo Board. Microchip's plug in module uses 1.27mm pitch sockets, which is not as user friendly as the usual 2.54mm pitch, but has one to one mapping of pins onto the microcontroller.
As the reply to "TQFP 100 break out for Microchip devices" on the blog here's the breakout board that I mentioned.
I have to admit that the placement of ICSP and JTAG are very tight in this design. The board fits exactly the 5cm*5cm board order at SeeedStudio. Another part that I'm not really satisfied is the placement of the crystal. It is placed close to other pins, although on the bottom side, which is susceptible to noise. Unless I go for 4 layer board design this is the best that I can do.
Yes, I'm targeting specifically on the UART mode of the Bus Pirate although the adapter could also be used with Bus Pirate in logic analyzer mode. The adapter only have to connect to TX and RX pins of the Bus Pirate for translation of voltage.
Both RS422 and RS485 are just like RS232, which are standards defining the electrical characteristics of transceiver. RS232 is defined for single-ended signaling with split TX and RX and only allows one device for each connection. RS485 has differential signalling for better noise immunity over long distance and support multiple devices in a form of network. The only downside for RS485 is half-duplex instead of full with TX and RX combined into a differential pair. RS422 is also differential signaling but with both TX and RX split out into 2 differential pairs and is full-duplex.
RS422 transceiver RS485 transceiver
There is no need for firmware modification to use RS232 ans RS422 adapters except for RS485. As shown in the picture above, the transmit and receive pair for RS485 are coupled together and each have their own enable pin. This, I suppose could use the clock and chip select pins of the Bus Pirate to manipulate. I have yet to try it out but I'm expecting that if both transmit and receive were to be enabled, this forms a loop back where we receive what we've sent and is just a matter of filtering.
An example out there that uses UART over RS485 communication is the RX-28 servo motor by Dynamixel. In the datasheet of the servo motor, the specified communication is Asynchronous Serial Communication. On the open source side, there's the electronics of RepRap 3D printer which uses UART over RS485 for communication from the mainboard to their extruder controller.
Being a silent reader of DP most of the time, here I would like to contribute a little.
The idea came when I was designing board for the half priced FT232RQ chips that bought from Dontronics (thanks to DP). From the data sheet of the chip, there's application examples of using this FT232RQ with RS232, RS422 and RS485 transceivers for serial communication.
The Bus Pirate has UART function that drives at 0-5v level which is comfortable for microcontrollers but not for the above mentioned standards. These three standards are most likely to communicate in UART. So I thought why not design these transceiver adapters for Bus Pirate. This will extend Bus Pirate's reach to sniff out and possibly manipulate devices out there that use these three standards. Of course we could choose to crack open the target device to trace out the actual 0-5v UART signal. But I imagine these adapters simplify the job to only requiring a custom cable that match the target.
Before I continue with the design, I would like to know what's the likelihood and if it's feasible to have these adapters for Bus Pirate.
This is my second entry for the 7400 logic competition. It’s a discrete logic amplifier made of mostly discrete logic chips. This amplifier is a continuation of my previous discrete logic preamp. The main objective here is to amplify the input signal to a suitable level for driving a buzzer or other voltage driven application.
The amplifier works by converting the analog signals into digital form. The converted digital data is then feed through a logic level converter to bring the digital data to a higher voltage level. The digital data then converts back to analog signal but with larger amplitude.
The circuit consists of a delta-encoded ADC with a feedback R2R DAC, logic level converter using two sets of buffers, a crystal oscillator and another R2R DAC for final output. There are two 74LS169N four bits binary counters cascaded as 8 bits with a LM361 fast comparator that forms the ADC part. The fast comparator compares the input with the feedback and outputs logic high or low that causes both counters to count up or down. The 8 bits output of the counters goes through the first set of R2R DAC to form back as analog feedback. Because of the simple ADC design, the output theoretically will always oscillate by one bit assuming a stable input.
The crystal oscillator was made of 74LS04N hex inverter and clocked at 40MHz to allow headroom for higher frequency analog input sampling. Operating at such frequency, the counters will take 6.4us to count from 0 to 256 or about 156kHz ramp when probed at the DAC with no feedback and counters' direction fixed. When Nyquist rate was considered, this allows us to sample input frequency up to about 78kHz so that the counters could always catch up and won't cause distortion at output.
The 8 bits output of the counters also goes through the logic level converter which consists of 8 TTL to CMOS logic converter (2 x 74LS07N) and 8 CMOS buffers (2 x 4050N). The key here is that CMOS logic chips are capable of higher operating voltage (up to 15V compare with TTL 5-7V) and this is where the actual amplification takes place. Volume adjustment could be done by varying the supply voltage to the CMOS buffers but was not implemented in this entry. The amplified 8 bits then goes through the second set of R2R DAC to convert back to analog signal but with larger amplitude.
Here’s my first entry for the 7400 competition. This is a discrete logic preamp using SN74LS86AN from Texas Instruments. I was focusing on the misuse of 7400 logic chips of this competition and thought of using TTL to do analog signals.
It happens that I have a SN74LS86AN XOR chip on my table and I began the testing by hooking up a pot to one of the inputs while tying the other high. Through the scope, I find that the XOR gate does show analog property. However, the analog property is only limited to a region where the input ranges from about 0.8v to 1.2v with the output swings from about 1v to 4v. Since this chip had been sitting on my desk for quite some time, I thought that it might have had electrostatic damage. So I bought another same chip for test and to my amaze, the new chip also shows the same property. Below is a footage showing my test of both old (left) and new (right) chips. Apologize for the blurriness as the video was captured using my phone.
I went on to order a few other 7400 TTL (not CMOS) logic chips with different logic gates and none of them demonstrate the analog property. Perhaps the manufacturers had optimized the chips to only switch the output to either high or low only when reaching the thresholds. I also tested another 74F86 and the analog property was observed.
Based on the result and observation above, I came up with this discrete logic preamp. Below are the schematic and board design for the preamp. All four XOR gates in the chip were brought out so totaling four preamps in a board. Each of the inputs was tied to a trimmer for offsetting the input voltage to about 0.95v and about 2.5v at the output. Another important thing to note is that the circuit has to be exactly in the arrangement shown to reproduce the result. The inputs and trimmers have to be on pins 2, 5, 9 and 12 of the chip. The other input of each gate also has to be tied high.
I would say that even if you move the ICSP to the right side of the board, you still couldn't have the PICKit directly plug in because the stacking of the shields, if there is, would also prevent it. My opinion is that the board is already quite crowded and moving the ICSP to somewhere else would also make it inaccessible. So in the end we still have to use the 6 pins cable like the one provided with the PICKit clone by Sure Electronics.
As for the FTDI chip, the main concern is the compatibility with the Arduino environment. Looking back to the original Arduino, the FTDI chip only serve for the USB to serial communication. So the FTDI chip on this board should be only for that purpose too. If they were to use a PIC18F2550, yes, it is possible too by an external switch to change between the serial communication mode or PICKit mode. But the main problem would be to rewrite the firmware for it which then involves the licensing of the FTDI and PICKit.
Today I received my order of chipKIT Max32 from MicrochipDirect. I placed my order after knowing of the free shipment from DP blog article a few weeks ago. Below are the pictures of my unboxing.
close up attempt on the PIC32
I have to say the assembly of this chipKIT is really good compare to Arduino. Probably Digilent was not in a rush as Arduino. They also have better layout with reset button on the top left corner, power select jumper and LED right beside the power connector and the ICSP in between power and usb connectors. Everything looks clean and their soldering are solid. The single row pin header sockets are nicely aligned and none of them slanted. The edges of the board were cut by V-grooving but they're still nice to the touch. The board also packaged in the box with an anti-static foam underneath it.
If there is one thing I don't like about this board is the locations of the mounting holes. They're too close to the edges and might not be strong enough. And also the red marking on the PIC32.
Good suggestion, thanks. But currently the two breadboards that I have do not match with each other on the latches. Probably would also need to remove the horizontal rails normally used for power to get the board in correctly.
Had also programmed it with the LED toggle demo using BusPirate with XSVF player. Everything works fine and soon going to download the ISEWebpack to try on.
The PCB layout is nice but a downside is that it can't be used on my breadboard. It is possible to fit the board in with the pin headers but there will be only either one of the sides available for wire connection. It could be better by shrinking the width of the board and have the distance between the two rows of headers at the sides down by 2.54mm.
Guess that I could only use the dual female jumper wires and breadboard jumper wires from Seeedstudio to wire the board to the breadboard.
[quote author="ian"] We want 'out of the box' compatibility with the widely-supported olimex JTAG programmer, but the voltage difference was a real problem. The 5volt chip can be used with the AVC logic family to interface 5.0volts to 2.0volt targets, but when run at 3.3volts the maximum becomes 3.3volts max. [/quote]
Could you please elaborate further on this 3.3volts max problem?
I still couldn't find the source code for this USB RGB color changer. All I can find is the compiled hex file in the archive at DIYLife.com. Anyone knows where can I obtain the source code?
