Fred 453 Posted August 26, 2014 Share Posted August 26, 2014 Wow! This device is complex - like it was designed for a specific client application but is now being re-purposed for regular, everyday applications. I agree. I'm guessing it's not used much as nobody knows exactly what they could do with it. Probably why TI thought that a few sample project from 43oh members would be a good showcase. @@Foghorn @@chicken I'd agree when if your goal is getting something built then you'll pick the right tool for the job. However sometimes the goal is to get to know your way around a certain tool / device, in which case fitting the job round the tool is just fine. Quote Link to post Share on other sites
bluehash 1,581 Posted August 26, 2014 Author Share Posted August 26, 2014 @@chicken Your high level description of the ESI module is spot on. I will like to add my two cents to the description. Each AFE has two types of inputs (ESICIx and ESICHx), The main difference is that with ESICHx you can excite sensors (AFE1 and AFE2 use the same excitation logic) and ESICIx inputs are not connected to the excitation logic. In addition to ACLK, SMCLK or 32KHz oscillator you can also use ESICLK which has a nominal frequency of 4.8 MHz but can be tunned to operate between 2.3 -7.9 MHz in ~78 KHz increments. Also, the ESI module allows you to divde your low frequency clock source by a wide number of dividers (refer to Table 28-25 in SLAU367) Correct, you can program up to 127 states and the state transition are driven by output of PPU(PPUS1, PPU2,PPUS3) and bit Q6. Note: if you are using PPUS3 and bit Q6 for next PSM state transition you need to properly configure the ESIPSM register for such operation. Thanks for chiming in and answering our questions @@ivanc! Welcome to 43oh. Quote Link to post Share on other sites
bluehash 1,581 Posted August 28, 2014 Author Share Posted August 28, 2014 End dat coming up. Please note down your submission if you have one. Also have a title that describes your idea. Thanks! Quote Link to post Share on other sites
bobnova 59 Posted August 29, 2014 Share Posted August 29, 2014 Spent some time digging through the users guide / datasheet / whatever TI calls it. I don't have any sensors or projects that would make using the ESI worthwhile. There are a couple that it could be used on, a digital tach/speedo/MPG/etc. dashboard for my car for instance, but there's no benefit I can see to using the ESI to grab rotational data rather than interrupts. Another one is water consumption monitoring, it'd be great for that if I were looking at using good, accurate, expensive, sensors. I don't have a kilobuck to throw down for one though so that's out. In the end, a big thanks to TI for doing this, but I've got nothing for them. Maybe something will occur to me before midnight Quote Link to post Share on other sites
rampadc 29 Posted August 29, 2014 Share Posted August 29, 2014 I would like to withdraw my entry if that's alright. Last time I dealt something with the mains I accidentally shorted the lines and cut power from the whole house and electricians are expensive. Quote Link to post Share on other sites
bobnova 59 Posted August 30, 2014 Share Posted August 30, 2014 OK, I have two three different projects that the ESI could be used with, I'm going to list 'em from least ambitious to most ambitious. The more ambitious the less likely I'll actually manage to do it, but who knows. Possible project one: Digital tachometer, speedometer, and intelligent shift light. Uses the ESI to read either the ignitor's tach output (one pulse per ignition event, so two per engine revolution) or the vehicle speed sensor's square wave (whichever is more difficult to do with interrupts/timers, unknown at this time) as well as the throttle position sensor. Displays engine RPM via either LCD screen or RGBLED bars (or both, plenty of GPIO here), also vehicle speed with LED indicators for common speed limits. Based on throttle position, engine RPM and vehicle speed (and gear, calculated via engine RPM and road speed) it will also have a pair of shift lights, one to indicate for downshifting and one to indicate for upshifting. Example: On the freeway behind someone at 60MPH in a 65 zone in 5th gear, left lane opens up and you can accelerate. If you give the engine a little bit of gas to accelerate slowly, no lights. Stomp on the gas and the downshift light comes on (perhaps blinking, to indicate multiple gears downward are indicated, as full throttle 60MPH is best done in 3rd gear on this car). If/when you downshift, the light goes out. Let off the throttle to maintain your new speed and the upshift light comes on, as 65MPH in 3rd is lousy for cruising. Also indicates upshifts based on engine RPM directly, given a 6750 redline and high throttle angle the upshift light would come on around 6500RPM. I may make the MCU learn the rev limits, may not. ESI and FRAM are not specifically needed for this project, but using ESI for either tach or road speed would eliminate timing artifacts that would happen if the road speed and tach signals happened simultaneously, something that is guaranteed to happen eventually. Possible project two: Battery-free bicycle telemetry Bicycle telemetry, read road speed via multiple magnets attached to one rim and a coil attached nearby. Trick is, I want to make said magnet/coil arrangement power the MCU as well as give it a timing signal. Both FRAM for its low power consumption and the ESI module for its ability to grab rotational information while the CPU sleeps would be useful. Once it has this information, display it on low power LCD screen and save it to FRAM blocks for later downloading to computer, possibly via radio module. No radio for normal operation, radio module also contains a battery. Plug the radio module into the bike and it detects the external power and radio, and contacts a home base unit attached to a computer. Possible bonus features include a second sensor on the crankset, and shifting suggestions similar to project #1. Alternatively a battery+SD card for data transfer, done more or less the same way. I'm not positive that enough power can be generated this way, if not then I'll use a more classic power method, but the ESI and FRAM low power abilities will still be useful. Solar maybe, with a small rechargeable backup battery. Most bike riding is done during daytime after all. Possible project number three: Digital tach/speedo/intelligent shift light, with advanced fuel consumption monitoring. Same as project #1, but put the ESI on injector monitoring duties (start timer on injector fire leading edge, stop on trailing edge, calculate fuel injected based on time) and calculate instant and average fuel consumption. Display current MPG, average MPG, cruise MPG and town MPG (buttons to switch between display modes are now needed). Ideally run some calculations internally with some learning ability, to allow the MCU to suggest cruise speeds/gears for better mileage. Also a route mode. Push "start" button, drive to where you're going, push "end". Display tells you both MPG and how much fuel you actually used. This can be used to figure out what the most efficient routes from point A to point B are, as MPG does not tell the whole story (20 miles at 30mpg, is worse than 10 miles at 20mpg, for an extreme, but not unreasonable for direct city vs indirect freeway example). #1 is quite doable. #2 may have power issues, but is doable as well. #3 is further out there, but something that I would like to do and more importantly has a feature that I've not seen anywhere else. Will I actually do any of these if I win? I plan to do them anyway. Free parts help, though. Are the FR6989 chips necessary? Not really, a FR5969 would work just as well, or a LM4C for the automotive ideas. Considering that I can't find a place to buy fewer than 1000 FR6989 chips, I'll either be getting by with free samples or using a different MCU. Certainly don't have $4500 to throw down on 1k FR6989s, that'd be overkill even for me. Will I blog/log about the project(s)? Definitely. abecedarian and tripwire 2 Quote Link to post Share on other sites
Automate 69 Posted August 31, 2014 Share Posted August 31, 2014 I got my project idea from this paper "Single-Point Sensing of Whole-Home Water Activity" https://homes.cs.washington.edu/~jfogarty/publications/ubicomp2009.pdf It's a home water usage sensor with a twist. Instead of using a water flow sensor at every faucet, tub, shower and appliance that uses water, a single sensor is used to monitor the water usage at every location. This is accomplished by learning and recognizing the unique pressure / vibration pattern for each water usage location. In this way you know not only how much water is used but also which location is using the water without the high cost of sensors at every location. In addition the system would trigger alarms (maybe with text message) for specific water usage events. Such as if it detects water flowing continuously for more than an hour or some configurable time. This might indicate that a toilet float is stuck and wasting water.or someone left a faucet open. Quote Link to post Share on other sites
pjkim 22 Posted August 31, 2014 Share Posted August 31, 2014 I have been working on a servo motor project. I fabricated a quadrature encoder and affixed it to a double shafted coreless motor with a planetary gearbox. The coreless motor is rather nice because of its efficiency-- turns easily with a single worn out AA battery measuring 1V, but can be driven at up to 24V for more power (for short periods). I want to use this in projects where stepper motors are not a viable alternative because of their high power consumption. The encoder is working and I have proper quadrature signals coming out. I am trying to decide on a microcontroller to read the quadrature encoder, run a PID algorithm, and control a PWM signal to a motor driver. I want to use a microcontroller with a dedicated quadrature encoder instead of interrupts and wonder if the MSP-TS430PZ100 would be a good fit. I am attaching pictures of what the motor assembly looks like as well an oscilloscope screenshot of the quadrature signals. Disclaimer-- I read the ESI chapter in the MSP-TS430PZ100 datasheet and it looks like it is fairly dense to say the least-- possibly/probably a reflection of its flexibility. I am hoping there is a "ESI for dummies" tutorial somewhere. Quote Link to post Share on other sites
greeeg 460 Posted August 31, 2014 Share Posted August 31, 2014 Disclaimer-- I read the ESI chapter in the MSP-TS430PZ100 datasheet and it looks like it is fairly dense to say the least-- possibly/probably a reflection of its flexibility. I am hoping there is a "ESI for dummies" tutorial somewhere. Cool project, I think that's partially TI's goal with this competition is to have some example projects that will show off example cases of how to use the ESI module. Also, correct me if I'm mistaken.. but isn't the chip the MSP430FR6989? and the MSP-TS430PZ100 is just the PCB with an expensive 100pin IC socket on it for development purposes? Quote Link to post Share on other sites
Fred 453 Posted August 31, 2014 Share Posted August 31, 2014 That's what I thought. Those LQFP ZIF socket boards are surprisingly expensive and seem to only be compatible with a few MSP430s (for the programming interface at least). The ESI does seem to be a peripheral that will be hard to get the best out of. I think we're supposed to be creating the ESI for dummies guide. Quote Link to post Share on other sites
abecedarian 330 Posted August 31, 2014 Share Posted August 31, 2014 Cool project, I think that's partially TI's goal with this competition is to have some example projects that will show off example cases of how to use the ESI module. Also, correct me if I'm mistaken.. but isn't the chip the MSP430FR6989? and the MSP-TS430PZ100 is just the PCB with an expensive 100pin IC socket on it for development purposes? MSP-TS430PZ100D comes with two MSP430FR6989IPZ. Quote Link to post Share on other sites
abecedarian 330 Posted August 31, 2014 Share Posted August 31, 2014 I'd like to revise my project idea a little bit, but it's still along the same theme. It would eventually do a few things: - Cap-touch control over lights in my nephew's bathroom and bedroom (they are adjacent); - monitor electricity usage of the above (convenient since the breaker panel is between the bed/bathrooms); - monitor water usage in my nephew's bathroom. Purpose- to help teach him the importance of conservation. Quote Link to post Share on other sites
Citizen 7 Posted September 1, 2014 Share Posted September 1, 2014 I'd like to revise my project idea a little bit, but it's still along the same theme. It would eventually do a few things: - Cap-touch control over lights in my nephew's bathroom and bedroom (they are adjacent); - monitor electricity usage of the above (convenient since the breaker panel is between the bed/bathrooms); - monitor water usage in my nephew's bathroom. Purpose- to help teach him the importance of conservation. Stop monitoring your nephew. Give him some privacy:)) igor 1 Quote Link to post Share on other sites
bluehash 1,581 Posted September 1, 2014 Author Share Posted September 1, 2014 --- Closed --- I'll be listing all the entries in the first post. Will keep you posted. Thanks for your entries! Quote Link to post Share on other sites
bluehash 1,581 Posted September 4, 2014 Author Share Posted September 4, 2014 Final entries: Abecedarian -: Water supply usage Chicken -------: Resistive touchscreen pattern detector Fred -----------: Laser cutter coolant and temperature monitor Greeeg--------: Fitness monitor Rampadc------:Central heating controller/interface Bobnova-------:Digital tachometer, speedometer, and intelligent shift light. Automate------:Single-Point Sensing of Whole-Home Water Activity Will be forwarding these to TI and keeping you all updated. Thanks again for your entries. 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.