EE project ideas

[flyingcheesehead]

(EDIT: If you're seeing this thread for the first time, the project I'm talking about in this first post is complete and posted in post 22. Now it's time to do another one - See post 23 for details!)

So, I need to choose a final project for my Integrated Circuits class (EE 541). I'd like to make it something aviation-related. It needs to cover new developments in technology, with information coming from things like IEEE papers. The format is simply a 20-minute presentation followed by 6-7 minutes for Q&A, plus an accompanying written project report.

The class has basically covered analog CMOS design. So far, we've covered differential amplifiers, op-amps, bandgap references, and switch-capacitor circuits. Still to come: Oscillators and phase-locked loops. So, it should hopefully relate in some way to that sort of thing as well.

The actual project requirements:

1) Topic must be within devices and circuits area.
2) Survey paper has to cite works from at least 2 IEEE transaction or journal papers (not conference papers) and must be at least 8 pages long.

So, it's fairly open. I'm looking for suggestions for things that are in some way related to aviation and/or avionics.

Thanks in advance for any suggestions!

 

[Dr. O]

Collision avoidance detectors such as ZAON need improvement... This is an area that needs fresh ideas...

BTW, average pilots - like us - don't even know how a transistor works, so you are really grasping at straws here... My suggestion is to survey the recent IEEE literature on MMI devices...

denny-o

 

[vontresc]

Power FLARM might be an interesting thing to look into as well. Don't know how much tere is on research on it though.

 

[COFlyBoy]

I was going to suggest looking at integrating consumer grade inertial systems with GPS (There should be lots of IEEE documentation on this), but this may fall more into the digital filters realm than analog CMOS.

Maybe you could look at the analog parts of consumer grade inertial reference systems.

 

[denverpilot]

Electrical and RF interference within and external from avionics, and mitigation techniques?

 

[CJones]

You should investigate the potential for making the flux capacitor more 'green-friendly'. Lots of advances in technology since Doc Brown and Marty McFly. There HAS to be something that will perform the same work on less than the original 1.21 gigawatt requirement.

 

[Ted]

You should investigate the potential for making the flux capacitor more 'green-friendly'. Lots of advances in technology since Doc Brown and Marty McFly. There HAS to be something that will perform the same work on less than the original 1.21 gigawatt requirement.

Well, it's 2011. Within 4 years we should have flying cars, hoverboards, and Mr. Fusion to solve the problem.

 

[flyingcheesehead]

Collision avoidance detectors such as ZAON need improvement... This is an area that needs fresh ideas...

Unfortunately, this isn't a design project. The stated purpose of it was that since we can only learn so much in school, and we'll need to know how to learn what we need to learn in the real world, that he wants to show us how to do that now. Very cool, really. Most profs don't seem to care about the real world. This guy's been there - Used to work at Sun and AMD.

BTW, average pilots - like us - don't even know how a transistor works, so you are really grasping at straws here...

Well, I'm hoping that Lance or some of the other EE's here will have maybe seen something from IEEE in the last several years that might fit this mold.

My suggestion is to survey the recent IEEE literature on MMI devices...

My Google-fu is weak today. All I can find about MMI is that there are a whole lot of different things that you could be referring to - Can you clarify? Thanks.

I was going to suggest looking at integrating consumer grade inertial systems with GPS (There should be lots of IEEE documentation on this), but this may fall more into the digital filters realm than analog CMOS.

Maybe you could look at the analog parts of consumer grade inertial reference systems.

Hmmm... iPads in the cockpit? That's right up my alley.

Electrical and RF interference within and external from avionics, and mitigation techniques?

That would relate to the common-mode rejection characteristics that we've talked about, but it's not really something "new".

Well, it's 2011. Within 4 years we should have flying cars, hoverboards, and Mr. Fusion to solve the problem.

Naah... It's 2011. We were supposed to have all that crap 11 years ago!

 

[Ghery]

Electrical and RF interference within and external from avionics, and mitigation techniques?

Well, there's a bit more than common mode rejection involved in EMC. You can check the Transactions of the IEEE EMC Society for archival papers on the subject. But, EMC is more of a design requirement, as opposed to selecting a product you want to design. DO-160 would be the standard to worry about for the hardware.

 

[tonycondon]

kent not sure if you can work something into this idea but perhaps it will give you some ideas.

http://www.d.umn.edu/~cprince/PubRes/FbF/

 

[flyingcheesehead]

kent not sure if you can work something into this idea but perhaps it will give you some ideas.

http://www.d.umn.edu/~cprince/PubRes/FbF/

Wow, that's interesting! Reminds me of something else I've heard about, that was in regards to instrument airplane flying - Remind me next time I see you and I'll tell you all about it. Very cool stuff.

Since we're supposed to be concentrating on "devices" rather than "systems" I'd be looking at something like the sensors themselves.

Thanks!

 

[Ted]

Naah... It's 2011. We were supposed to have all that crap 11 years ago!

According to Back to the Future, all we know is that by 2015, this stuff was everywhere. So that would've meant Steve Jobs introduced it in about 2014.

 

[wsuffa]

"Overclocking an iPad..."

Had to get that in...

Seriously, something related to high-reliability performance in hostile environments... such as unpressurized aircraft or high-altitude operations. Or impact of radiation (cosmic, nuclear, etc) on CMOS. Etc.etc.

 

[gismo]

A common problem with turbine fuel flow sensors is that the output seems to fluctuate a lot. The typical "cure" is to filter the flow rate in the time domain but this results in a significant lag from mixture adjustments to the resulting change in indicated fuel flow. I suspect that a PLL based implementation would work much better assuming the lock bandwidth was set properly. This could be done entirely in the digital domain since the actual data rate is pretty low but it could also be accomplished with a mostly analog design. Another application that I think begs for a PLL implementation is ANR where the PLL tracks the fundamental frequency produced by the prop, but again the frequency is low enough for a digital approach.

The PLL designs I did most often were for tracking the sync of an analog TV signal in digital video interfaces. In that case a purely digital approach had objectionable jitter in the output phase which manifested in perceptible visual artifacts.

I should also point out that analog PLLs, especially ones with high clock to sync ratios (e.g. 27 MHz pixel clock vs 15.7khz sync) are some of the most noise sensitive circuits known to man.

 

[RJM62]

A common problem with turbine fuel flow sensors is that the output seems to fluctuate a lot. The typical "cure" is to filter the flow rate in the time domain but this results in a significant lag from mixture adjustments to the resulting change in indicated fuel flow. I suspect that a PLL based implementation would work much better assuming the lock bandwidth was set properly. This could be done entirely in the digital domain since the actual data rate is pretty low but it could also be accomplished with a mostly analog design. Another application that I think begs for a PLL implementation is ANR where the PLL tracks the fundamental frequency produced by the prop, but again the frequency is low enough for a digital approach.

The PLL designs I did most often were for tracking the sync of an analog TV signal in digital video interfaces. In that case a purely digital approach had objectionable jitter in the output phase which manifested in perceptible visual artifacts.

I should also point out that analog PLLs, especially ones with high clock to sync ratios (e.g. 27 MHz pixel clock vs 15.7khz sync) are some of the most noise sensitive circuits known to man.

That's pretty much what I was thinking, too.

-Rich

 

[flyingcheesehead]

A common problem with turbine fuel flow sensors is that the output seems to fluctuate a lot. The typical "cure" is to filter the flow rate in the time domain but this results in a significant lag from mixture adjustments to the resulting change in indicated fuel flow. I suspect that a PLL based implementation would work much better assuming the lock bandwidth was set properly. This could be done entirely in the digital domain since the actual data rate is pretty low but it could also be accomplished with a mostly analog design.

Well... The project topic that I got approved (including approval of a couple of specific IEEE papers) is "Integrated Circuit Gyroscopes in General Aviation Glass Cockpits." I'm covering both vibrating gyros and ring laser gyros and the developments that have made them feasible for use in small aircraft. Really fascinating stuff, and I've learned enough that when I had a question on it, I managed to stump both the professor for this class, and the other professor he suggested I talk to.

Lance - When I re-read your message above, AFTER doing the research I've done on this project, I thought "That's just asking for a Kalman filter!" So I guess I've learned something along the way. And since a PLL can be used as a very simple form of Kalman filter, I think we're probably in agreement. BTW, this is what stumped the two profs. I wouldn't say I have a complete grasp of it all yet, but I get the concept at least.

 

[Graueradler]

Adapt ultrasonic wind detectors into an angle of attack meter.

 

[COFlyBoy]

Well... The project topic that I got approved (including approval of a couple of specific IEEE papers) is "Integrated Circuit Gyroscopes in General Aviation Glass Cockpits." I'm covering both vibrating gyros and ring laser gyros and the developments that have made them feasible for use in small aircraft. Really fascinating stuff, and I've learned enough that when I had a question on it, I managed to stump both the professor for this class, and the other professor he suggested I talk to.

Lance - When I re-read your message above, AFTER doing the research I've done on this project, I thought "That's just asking for a Kalman filter!" So I guess I've learned something along the way. And since a PLL can be used as a very simple form of Kalman filter, I think we're probably in agreement. BTW, this is what stumped the two profs. I wouldn't say I have a complete grasp of it all yet, but I get the concept at least.

Mmmmmmm! Kalman Filters: so simple to concieve, so difficult to implement, and tune, and verify.

Sounds like a cool project. I wish I was back in school.

Ring laser gyros used to be so expensive they were only used on high-end satellites. The physics behind these gyros is really cool.

 

[SCCutler]

"Shook Avionics" releases its first product, the Kent 1500 Integrated Panel Display.

 

[flyingcheesehead]

MRing laser gyros used to be so expensive they were only used on high-end satellites. The physics behind these gyros is really cool.

Yup. Part of what I'm covering is how they got inexpensive and small. I saw pictures of the Delco Carousel RLG's that were used on the 747-100 - Huge! Now, they're building 'em on an integrated circuit - They discovered around 1998 that they didn't need an actual optical sensor - The laser diode they're using to light up the ring inside the IC has somewhat of a photovoltaic effect - So they drive the laser diode with a constant current source, and the beat frequency shows up as voltage changes on the terminals that drive the diode! Cool, huh?

I'll probably post the paper and presentation here when I'm done with 'em.

"Shook Avionics" releases its first product, the Kent 1500 Integrated Panel Display.

It's free, as long as you let me fly your airplane!

 

[denverpilot]

I'll probably post the paper and presentation here when I'm done with 'em.

I read somewhere where some schools now claim intellectual property rights on student's papers.

I doubt you're at one of those, but your statement that you'd self-publish reminded me of it.

Kinda jacked. Of course there's always the whole "Federal or State funding means student works at non-private institutions are 'owned' by the general public" topic too.

Messy.

 

[flyingcheesehead]

Well... Time for a thread bump, which I'll post about in a minute, but...

I'll probably post the paper and presentation here when I'm done with 'em.

Fail! Well, until now. They're attached. Some of the stuff in the early parts is very basic for pilots, but was added for explanation of the problem to engineers. Some of the later stuff may be over the heads of many of you, but I did try to keep it reasonably easy to understand.

It was a lot of fun, I had always wanted to know how the AHRS worked in glass cockpits. The presentation went well, I answered all of the questions the prof posed afterwards well, and I think it was better real-world subject material than anyone else covered. It certainly boosted my grade a bunch.

Enjoy!

 

[flyingcheesehead]

Aight folks, time to do it again - This time, for my VLSI Design class. Same professor, same idea. I'd love to do another aviation-related project - It definitely helps with motivation.

Once again, the specific requirements:

1) Topic must be within devices and circuits area.
2) Survey paper has to cite works from at least 2 IEEE transaction or journal papers (not conference papers) and must be at least 8 pages long.

Whereas the previous class covered analog CMOS design, this class has covered digital... But mostly still from the perspective of "how can we make a digital circuit perform better by focusing on it from an analog perspective," or so it seems to me.

For example, here are the sample topics listed, they should give you an idea of the types of things we've covered in the class:

1) Threshold voltage adjustment to speed up digital circuits
2) Multi-bit flash memory cells - Operation and reliability
3) High performance flip-flops, latches, register cells
4) Clock and power distribution networks
5) Memory designs
6) low power designs
7) layout designs

Now, to relate this to aviation - I could do another glass cockpit related topic, but I don't know what specific advancements were required in the areas we've covered to enable that, and I think that they're likely not specific to aviation like the AHRS gyro topic was, so it'd be hard to put together.

I am, however, curious if maybe NextGen would require some of these advancements, but I don't really know enough about it to know if that would be a dead-end search too.

Any suggestions are much appreciated - Thanks!

 

[denverpilot]

How about design of memory checking and single-event-upset automatic repair and recovery algorithms for spaceborn flash and or SRAM that isn't radiation-hardened? And how many upsets it would take to cause data loss? Fun math there.

Might as well put some space in that aerospace.