oPossum 1,083 Posted August 17, 2012 Share Posted August 17, 2012 The LogicBoost is a 6 channel 16 MHz 256 ksample logic anlyzer booster for the MSP430 launchpad. It works with 3 to 5 volt logic levels. Many common communication protocols can be observed, such as: - Async serial ("UART", "RS-232") - SPI - IIC - Dallas/Maxim one wire - Low speed USB (1.5 Mbps) - DHT11/22 sensors - Consumer infrared and IRDA It is also useful for PWM and other applications. SPI RAM chips are used for the acquisition buffer. This greatly simplifies the design relative to using parallel SRAM chips. The SPI RAM have and internal address counter that eliminates the need for an external address bus and counter. The MSP430 sets up an address and write mode, and then data is clocked into the RAM on every SPI clock. The PCB is 50 x 50 mm for low cost and uses thru-hole components so anyone can build it. An optional FTDI board can be used for faster host communication. It can be stacked with the Launchpad Mini - also 50 x 50 mm. An FTDI board is required for host communication and power in this configuration. bluehash, dkedr, yyrkoon and 5 others 8 Quote Link to post Share on other sites
oPossum 1,083 Posted August 17, 2012 Author Share Posted August 17, 2012 The SUMP protocol is used for host communication. Recommended software is Open Logic Sniffer. The application UART of the launchpad is usable, but slow. An FTDI board is recommended for faster and more reliable communication. The prototype was initially tested with a 74HC4040 counter clocked at 10 MHz. Aliasing of the 5 MHz output is noticeable. As a general rule, 10x sampling is desirable, and 4x is a minimum. So signals up to 1.6 MHz are fine, and up to 4 MHz may work with limited timing resolution. The C2000 Launchpad just became available and the the Stelaris Launchpad has been announced. I will be investigating compatibility with both as soon as I can get them. There may be a PCB revision to allow support of all three launchpads. Firmware and PCB files: logicboost.zip Quote Link to post Share on other sites
bluehash 1,581 Posted August 17, 2012 Share Posted August 17, 2012 The C2000 Launchpad just became available and the the Stelaris Launchpad has been announced. I will be investigating compatibility with both as soon as I can get them. There may be a PCB revision to allow support of all three launchpads. looking forward to this. Quote Link to post Share on other sites
V0JT4 34 Posted August 18, 2012 Share Posted August 18, 2012 Is it possible to expand the solution to 8 channels? The latch is already 8 channel and there are free pins available on MCU. Is there some software limitation? From layout point of view I would be willing to go for SOIC and no crystal on LaunchPad in exchange of 8 channel solution. Quote Link to post Share on other sites
oPossum 1,083 Posted August 21, 2012 Author Share Posted August 21, 2012 Sure - P2.6 and P2.7 can be used as GPIO instead of for a xtal. Some minor firmware changes would be required to properly setup port 2 and use the additional 2 bits. Search the source code for 0x3F to find where most of the changes need to be made. V0JT4 1 Quote Link to post Share on other sites
GeekDoc 226 Posted August 22, 2012 Share Posted August 22, 2012 The C2000 Launchpad just became available and the the Stelaris Launchpad has been announced. I will be investigating compatibility with both as soon as I can get them. There may be a PCB revision to allow support of all three launchpads. looking forward to this. Me, too. This may make me buy one of the faster LaunchPads. Quote Link to post Share on other sites
TomKraut 17 Posted August 24, 2012 Share Posted August 24, 2012 This project just showed up on the official TI MCU NewsFlash for August. Congratulations! Quote Link to post Share on other sites
thanhtran 10 Posted October 19, 2012 Share Posted October 19, 2012 This looks like a clever design. I normally think of using a fast MCU to read the GPIO pins as quickly as possible to capture the data. Am I right that the maximum length of capture is limited by the SPI RAM? For example, the max time it can capture is 256K bit = 262144 samples * 1/sampling rate? i.e If I'm capturing at 16MHz, the max time I can capture is: 0.016.38 seconds (or 16.38ms) Can the SPI RAM act as a circular buffer providing that the MSP430 is off loading the data fast enough? Thanks -Thanh Quote Link to post Share on other sites
t0mpr1c3 91 Posted October 31, 2012 Share Posted October 31, 2012 Kevin - what values should I use for C1-C6? and R1=jumper, C8=omitted? Thanks. Tom Quote Link to post Share on other sites
jazz 209 Posted October 31, 2012 Share Posted October 31, 2012 This is great, but I will done it on much simpler (only MSP430F550x) way. uC will collect the samples on high speed (just storing byte from P4 to USB endpoint buffer on 25MHz) and (when the buffer is full, 64 bytes) sending it to PC over USB on 1 MB/s. After measuring is done, on PC side some (dummy) data from beginning of measured stream will be rejected due to USB bus waking up. No need for RAM, and no limitation in length of samples/measured data. Number of channels * Sampling speed = 1MB/s, so if the higher sampling speed is needed, number of channels can be decreased, and opposite. Quote Link to post Share on other sites
oPossum 1,083 Posted October 31, 2012 Author Share Posted October 31, 2012 C1 to C7 are all 100 nF (0.1 uF) - the typical bypass cap yes, R1 is jumper, and C8 ommited Quote Link to post Share on other sites
t0mpr1c3 91 Posted November 2, 2012 Share Posted November 2, 2012 Thanks. I soldered it up the other night. I didn't realise until I had the chips in place that you sent me a spare '573! That was very kind. It will be a head start if I decide to have a crack at your 23LC1024 based version. :thumbup: Quote Link to post Share on other sites
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.