Jump to content


  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by jpnorair

  1. There was a company called "Infinite Power Solutions" that used to make a battery like this, and which pioneered the technology. Actually, the technology was pioneered at Oak Ridge National Laboratory, but the guy from Oak Ridge founded IPS. IPS raised enough money to build a nice factory in Colorado (I've been there!), but things didn't really work out for them. For ST to license the tech from Oak Ridge and then to spend 3 years developing it (this project has been going for 3 years, but you couldn't buy it until now), indicates to me that they have some particular customer in mind. Ma
  2. Proto PCBs are sent to OSHpark. Pluto is basically a Teensy 2.0 with Cortex M0+. I tried to keep the I/O as similar to the Teensy as possible, although it is not quite identical. I also added some new things: A load-switch that allows a coin cell (or whatever) to operate as the power source, automatically, when it is not plugged-in. A raw Vdd port for attaching a super cap. Pink LED. Go pink. Charon is a tiny USB-dongle type thing. It is extremely small, the USB plug is the biggest part of the assembly. There is a 10 x 12.7mm header on one end that exposes SWD and a com-bus. SP
  3. Thanks, everyone, for the encouragement. Look for a kickstarter in the next couple of weeks. The project is thus dubbed, "Pluto." The base goal will be pretty low, for the hardware and basic firmware, and then I will do a series of stretch goals for additional firmware and software features. I'm thinking maybe $18/unit in order to pay for all the firmware time. COGS will be tiny, certainly under $4 in low/moderate volumes. Freescale makes nice parts, but: 1. Kinetis M0+ is more comparable to STM32F0, which is also really cheap 2. Both STM32F0 and L0 can do USB without a crystal
  4. Question 1 is something that you can figure-out just by reading marketing brochures on Semtech's website. Question 2: the answer is NO. However, I believe it is possible to license the LoRa technology from CSEM, if you want to build your own chip with LoRa. Question 3: only Semtech or CSEM can give you this.
  5. Small. I'll have a prototype soon. It will be able to fit inside a USB cable, I think. Possibly. In any case, the project is to be open source, so you can add features however you want. I had not planned on making this into a Teensy competitor, but perhaps there is reason to do so. I could one-up Teensy by offering multithreaded Wiring (i.e. multiple sketches running simultaneously) with automatic low-power usage in the background -- the RTOS underpinnings are already there. This would require a couple months of my time, and therefore would necessitate a successful kicksta
  6. Ring buffers are not good solutions for fast performance. For that, you want to use an A/B buffer. Yes, you need double the memory, but that's the tradeoff. To get "best of both worlds" I've found that, quite often, it is faster to do a DMA memcpy to shift the buffer (only happens when limit is reached) than it is to do the ring-buffer logic on every byte entered or removed. On MSP430's with DMA you can move 2 bytes in 1 cycle, plus some setup, whereas all those if's probably burn 20-30 cycles for each byte entered/removed. So do the arithmetic and figure out what is the best choice base
  7. Just a note: I'm building an open source project that utilizes STM32L0 as USB-UART gateway. This chip has some interesting attributes, and it can create a low-cost, small geometry USB interface that is also programmable. I'm building the CDC-ACM to UART version first, then an Android version that does the necessary authentication, and then finally a USB-Ethernet version so you can telnet to it (ssh is too much overhead). The idle current will be in the area of 2uA, so it's a good USB bridge for any low power project you may have, unlike most off the shelf bridge chips which are nowhere n
  8. There isn't much reason to use an MSP, because the energy cost of the LEDs is so high that optimizing power on the MCU seems like a drop in the ocean. You're probably better to use one of the many open source projects that implements frame buffering for LED matrices.
  9. There are patent holders in Bluetooth for sure, because there really isn't any such thing as a wireless technology without patents. Most wireless technologies are also not royalty-free (it is one thing to have a patent and another thing to license it). A lot of the time, such royalties get included in the price of the chip. So, if you are buying Bluetooth or WiFi modules from China which use chipsets coming from member companies in Bluetooth SIG or WiFi Alliance, you are probably fine (unless they are forgeries, which is another issue entirely).
  10. Note to yagi-uda builders: you can get more directivity by adding more elements to the front. There are already two in the design above. Just cut more wires of the same length and space them at lambda/4. A straight dipole or monopole will have no problem covering the entire band 2400-2500. These antennas are 75 or 37 Ohms impedance respectively, and the RF source of this device is 50 Ohms, so this is easy. I recommend this approach. I also recommend experimenting with trimming the antennas and testing RSSI against a reference because the wire you use will have some inductance (
  11. On a sour note: Broadcom has patents that must be licensed in order to market WiFi devices in most of the world. China does not have much in the way of patent law. If you're thinking about building a product to sell, I would avoid this module for legal reasons. If not, then buy a bunch -- they might not be available at this price forever.
  12. Alternate strategy: order 0.8mm instead of 1.6mm. Then use a good set of shears, or simply a low-end lever shear. http://www.harborfreight.com/18-x-15-paper-cutter-60367.html
  13. Cool, thanks for the tip. I made use of this to spin some antenna test boards.
  14. Ha ha... believe me, I think black and purple (or whatever) boards are awesome, but if I can save money, it doesn't take much to convince me to go with basic green. If you are putting together a maker-focused kickstarter I can understand the point of going for a custom look. Other than that, go for cheap. Incidentally, adding cost to go for ENIG finish is not always a bad idea. It bonds to solder nicely, certainly nicer than the lousy HASL finish.
  15. In radio modulation, which is a course of study with a mountain of research behind it, there is this thing called SNR (signal to noise ratio). If I decrease the data rate, not only do I increase the amount of energy per bit (the bit is longer, hence it is a bigger pulse), but I also can reduce the amount of noise because the signal is narrower. So I get bigger SNR. This is like increasing the duration of the sample. For the most part, it works great. In certain cases, however, you will find that there is *frequency dependent interference*. In other terms, spurious readings you can'
  16. Just FYI: this is only a maker thing. Still, most boards are green by default and no-one cares too much.
  17. The way I think of IR, is that IR is just a form of ASK (analog shift keying, or digital AM) using a ~300 THz (Terahertz) carrier frequency. The antenna for IR can be very small due to the very high frequency. however, to go a certain distance, the power of the transmission must also be very high. Interference from daylight poses an additional problem. Instead, a lower frequency band can be used. Since the helicopter already needs some from of control interface, it makes sense to me to just use the same interface for distance measuring. If we really care about accuracy, it is easier t
  18. There is no point of using BT Class 1 if the goal is to connect to the phone, because the phone does not have BT Class 1. The range of Bluetooth to a phone is always the same: usually 15m or less. BT Class 1 is not a single-chip solution. The cost is less and size is less to use a sub-1GHz long range SoC in addition to a regular Bluetooth SoC (or BLE SoC).
  19. The first thing you need to know is that these demo video very likely were done using a human operator of the drone. There are many ways to get a solution for this problem. In my opinion, a well-thought image processing stack on the drone could do 90% of the job required, here. This is probably the part of it that has the most work complete. Given the large size of the wrist unit, I'm guessing there are quite a few components in it. It is indicated to contain a Bluetooth Class 1 transceiver, which also necessitates it containing a large battery to source the heavy current demand. B
  20. jpnorair

    M2M shield

    "This prototype works, but not fully tested : I need a good alimention to supply 2 Amps bursts from the SIM900 during active communication." Try a LiPoly with a big capacitor.
  21. I was born in 1981, but one of my friend's dad was a programmer in the 80's, and I distinctly remember he had a TI-99. Most specifically I remember playing games on it, namely one called Parsec (I'm one of those people with a very good memory). Very cool, thanks for sharing!
  22. Good luck with that! Vodka sounds like a better plan. Na zdorovya.
  23. I have been working on "otter," which is a POSIX-C app that interfaces a TTY with a relatively generic packet-based, binary protocol. Github: jpnorair/otter With the proper setup, otter can act as a client-side shell, in which the text translations occur on the client rather than the server. The advantage here is that implementing a classic, text-based shell on an embedded device can consume a ton of resources, but the binary protocols used by otter incur only a modest overhead. I've implemented some of them on MSP430 and Cortex-M devices. otter is my first attempt using pthreads, a
  24. I have: (both sold, thanks guys) 36x CC430F5137IRGZ 50x MSP430F5503IRGZT These are both 48 pin QFN devices, and they are still in their cut-tape strips. The 5503 is identical to the popular 5510, except without ADC. It makes an excellent USB platform. Make me an offer -- can be cash or trade. As far as trade goes, my work has migrated mostly to STM32, so I might be interested in stuff containing STM32 devices.
  25. The CC430 can map any signal to any pin, so whatever you need to do CAN be done. You will also need to change the PanStamp RF code to use 433 MHz. I don't know exactly how PanStamp firmware picks channels, but you will need to change the values in several of the RF registers for the CC430.
  • Create New...