The Challenge Suggestion Thread, - submit your suggestions HERE!

A few more ideas:

Isolated Current Sensor
This one has obvious practical uses, but can you come up with a good, low cost, high accuracy, wide-band, DC capable, ect... sensor. Perhaps not all. for AC line voltages applications, the sensor would not need to be particularly widband or DC capable. for a SMPS, it would need to be DC and somewhat wideband capable, but not necessarily accurate.

or more generally:

Isolation Amplifier
a DIY isolated amplifier. One obvious method might be a VCO on one side of a transformer, and a PLL on the other. thus using FM to transfer signals across the barrier. or


isolated line communications.
eg, for SPDIF, JTAG, I2C, SPI, USB, ect... not necessarily wireless, but some form of galvanic isolation.

For that last one,maybe some of the pulse transformers that are used on Ethernet hardware? Add a buffer,and go?

Dunno,just a thought.

well Flix it seems like you might be able to get away with one SC710 dual schottky, one SC70 dual transistor, 2 0805 caps and two 0805 resistors (or smaller)

each has a different challenge. some are bidirectional, some require DC, USB senses the resistances on the data lines to determine if it is hooked up to a host/device/hub, i2c's open-drain architecture adds a challenge.

Not sure of the details as I don't know much about the technology at the moment. Feel free to change the requirements.

#1. Wearable energy harvesting.
Some kind of wearable device that generates electricity. There are a number of different methods, the winner should probably be the one that generates enough to do something useful.

#2. Generate power from renewable sources.
Obvious really, but it must be practical in the surroundings. ie if you live in a flat, with no windows, solar is out for you. Winner will have the most ingenious solution that fits their situation.

A helmet-mounted lightning rod...?

How about a small desktop generator, powered by moving air or perhaps a heat source? Maybe light up an LED or two with it.

Just a thought.

Hey guys, we need a new challenge!!!!
we need to do some activities!!!! also a strong competition.

Abdulla

This thread:
http://www.dutchforce.com/~eforum/index.ph...84&#entry311884

Gives me some ideas. Although leaning toward the mechanical aspects of engineering ... coming up with the a device to generate power from a human.

Mulling around the 'rules' and spirit of scoring... so by no means take any of the following as gospel. If I'm not mistaken isn't Tommy kind of the default voice of reason here?

Er, Any method of generation acceptable (bioelectric, thermal. mechanical, chemical) so long as all of the input is the result of one human being acting on or influencing an apparatus with no other source of energy input. Probably no need to clarify that with a grade 8 physics textbook, but it is a pretty open interpretation - pushing grandad's hudson over a cliff while attached to a genny via spool... Huh. Lifting a spool attached rock overhead repeatedly and dropping it at your feet. Pointless, but legal.

Storage devices (cap, batteries, mechanical devices) should be allowed so long as they start stone cold 'dead'. But what is a dead battery. Hmm.

Determining winner by power? Either not so easy or I'm overcomplicating it. We vary in size, strength, and stamina. Should the generation time be limited to so many seconds (still a brute strength bias)? Handicapped by some means? Or just nuts out see who can light up a stove top?

Input?

Truthfully, I'm tempted to simply stick an 18" length of steel tube with a weighted end to the shaft of a bike dynamo and spin it like a noisemaker by it's body out of curiosity to see how big a bulb I can light and how brightly, but this is of little comparative value.

Any thought on cleaning this up?

Riding a dynamo equipped bicycle will produce a fair amount of electricity.
I guess whatever is made has to be genuine.

Not entirely clear about this, but I assume you mean a live human being? Or does it matter?

I'm thinking compost & decomposition here. Generating methane to power a boiler and steam turbine.

Gotta dig out my Soylent Green DVD and watch it again. May get some ideas from that.

interesting idea, but as you mentioned, it would basically come down to who is the most active/noisy/luminescent.. someone arc welding all day wearing a solar panel vs someone on a jackhammer laden with Piezzo crystals would be interesting though..

Hi guys,

Feeling bored. Anyone else up for a challenge?

I like it!

Proposal:

An isolation amplifier is a device which accepts a signal on one side of a barrier, and transmits it to the other, with no galvanic connection (DC current path) between the two sides.

Galvanic isolation is hereby defined as a resistance between sides of more than 1 gigohm (i.e., <1uA leakage for a 1kV difference, etc.). Note that real resistors over 1G are easy to find these days...

Although galvanic isolation per se is DC, having a high impedance at AC is also beneficial. A capacitor might provide DC isolation, but allows both signal and noise through at higher frequencies.

For standardization purposes, the signal shall be a control voltage supplying less than 50mW of power. The preferred implementation is a 0...5V input, within +/-10mA current draw, but no restrictions will be given on the actual voltage or current range aside from total power, since it's valuable sometimes to have an oddball input. The output can have any offset and gain with respect to the input, as long as it's fixed (reasonably stable gain, offset, linearity, etc.).

It is not necessary for the circuit to supply power across the isolation barrier. If power is required for circuit operation, the nature of the power sources is unimportant; it is presumed that, were one to build the circuit for practical use, the power source would be designed with similar isolation rating. After all, you can always use a battery and have as much isolation as you want. If the power supply is not an important part of the circuit's signal path, it doesn't have to be constructed carefully for the contest.

Figures that are difficult to test (requiring special test equipment, large statistical batches to prove variation, etc.) shall be justified with a short analysis, followed by the judging panel's rebuttal, if any, to establish those figures. If datasheet values are used in the analysis, only the worse-case MIN and MAX parameters can be used. Sorry, no "typical" data or graphs without error bars (most of them!).


Points shall be awarded based on the following categories:

- Isolation barrier quality: voltage standoff, insulation thickness / distance, leakage current

- Bandwidth, in terms of -3dB cutoff frequency (analog) or maximum baud rate (digital). (Generally, an analog channel has a baud rate equal to twice the -3dB frequency; points will be weighted accordingly.)

- Both analog and digital are allowed; since digital is only 0 or 1, it's really just an analog isolator with really bad accuracy. (This is easily fixed with, say, an ADC/DAC pair, at some expense to bandwidth. Bandwidth-accuracy tradeoff will also be included in the scoring.)

- Precision, accuracy, stability, drift; linearity (analog), or propagation delay and skew (digital); pathological features, if any (nonlinear time dependence, etc.). Uncalibrated (as-manufactured, no adjustments) and calibrated (best case). These can be very hard to test, so do a short analysis if possible.


Basic ideas:

Isolators are all around us. A regulated switching power supply is something of an anti-isolator: the output is designed to be independent from the source! A power supply might be easily modified to perform differently, however.

Transformers are great, but limited in isolation and bandwidth due to core size and winding separation (the further apart the windings are, the worse the leakage inductance is, reducing bandwidth).

Optical couplers are fantastically cheap and common, but typically have pitiful accuracy (the average phototransistor type might have a CTR from 20% to 200% over operating point, temperature, aging and product variations!).

Some special optocouplers are built with two photodiodes, one on the LED side, one on the remote side; the photodiodes aren't very well matched, but their stability is good. If you adjust the LED current for, say, 100uA in the local side, the output might be 50uA or 200, but it will always be the same. So the as-built accuracy is crap, but after trimming the error, it remains pretty good.

The most dramatic examples of isolation, of course, are fiber optics and radio (arguably, similar subjects!). Fiber optic cable can be almost as long as desired, with minimal signal loss. Bandwidth can be huge. The transceivers are a little pricey though. Radio tends not to have quite as much bandwidth, but hey, if you want to, say, talk to a little robot you shot at Mars, you can't do any better than that. 100 Million Miles is, what, circa 10^14 volts isolation?...

Esoteric ideas: just a few that come to mind include really big resistors, abstract light transducers (LEDs and LEDs, incandescent bulbs, neon lights, etc.), heat (eww, thermal inertia = low bandwidth!), acoustic, etc.


Proposed timeline: all of October. Halloween themed projects perhaps?

Tim

Hi all :-)

I'm Kris in Wellington, New Zealand.

I've just joined, on a suggestion from Sch3mat1c.

I have a suggestion. This is a practical circuit for a real-world application. I've designed a solution myself (I know I can't enter my design, since it wasn't created for the contest) but not constructed it.

An in-line AC-mains switch for use with a 1kW heater, that cycles the heater on and off with a duty cycle that's potentiometer-adjustable from about 5% to 95% (or wider if you want). I need this because we have a grill with a fixed thermostat, that gets much too hot for toasting sandwiches, so I want to be able to reduce its temperature.

Here are some suggested guidelines.

The circuit must connect in line, between the AC mains supply and the heater load. So it has four connections: Phase and Neutral input from wall plug, and Phase and Neutral out to the heater appliance.
The load is non-inductive and is rated at 1kW. It will draw more than that when it's cold, I suppose. Say 1200W (I don't know for sure).
You can't assume that the load is always present, because my grill contains a thermostat, which may switch off if it gets very hot.
AC mains voltage is either 115V or 230V RMS at your discretion. (We use 230V here.) Frequency from 45~65 Hz should be supported, if that's relevant.
The unit obtains power from the AC input supply, either via transformer or some other method. Power consumption must be <20W. (That should be PLENTY.)
The unit must be small enough to mount in a tube of some kind, with a plug on one end and a socket on the other. In other word, no flying leads. Let's say maximum overall volume 20 cubic inches (328,000 cubic millimetres).
No moving parts, so you can't use a bimetallic strip, and you CAN'T use a relay.
Switching need not be synchronised with anything; the duty cycle need not be exact; the total cycle time should be between 5 seconds and 15 seconds (to suit a heating load).
Duty cycle will be controlled using a potentiometer with a value between 100R and 100K (your discretion) with a plastic shaft, for safety.

If you (i.e. the gods of the forums) accept this suggestion for a weekly design contest, and any other questions come up, I'll update this post (and label the changes).

Hi, and welcome to the forum Kris.

I'm in charge of nothing, but, a few things:

1 - We haven't done challenges in a while.
2 - While everyone liked the idea, no one really participated when we did have them.
3 - Challenges would be best selected based on things that are of interest and entertainment to most people.
4 - Your suggestion isn't really a challenge, it seems to be more of a "Not only do my work for me, but give me multiple solutions among which to choose."
5 - Your suggestion is tailored very very specifically to your needs and is narrow in both focus and constraints, so there's not much room for creativity.

But if you have an interesting project you'd like to discuss, or a solution you'd like evaluated, feel free to start it in a thread.

Hi AwesomeMatt,

Fair enough, good points.

Really? I looked at all the other challenges and they weren't of much interest - to me, at least. My challenge at least has a practical application, i.e. it is USEFUL.

I understand why you would think that, because you don't know me. I see that kind of rude request often on electronics forums (I hang out on electronicspoint.com) and I WILL actually design circuits on request for people, if I think they have a good attitude but just need guidance.

But as I said, I've already designed the circuit, and I'll attach it here as proof. I'm not looking for help; my idea was to give a realistic and detailed specification for people to work to, like you would see if you were in a professional engineering environment. I guess that's not the focus of these forums, though; they seem to be more for experimenters.

My main reason in suggesting this project is to see whether anyone could come up with any ideas that would improve my design.

Again, that's because I believe a designer should have a clear specification to work to. And I don't agree that having an explicit specification leaves little room for creativity.

I think this project would have some general usefulness. It could be used to control ANY heating device that didn't have a continuously adjustable heat setting, which applies to MANY heaters. Although I specified a lot of things that seem "tailored very very specifically to my needs", IMO those needs would be appropriate for this project when used by other people too.

Edit3: This image link should work (thanks CWB):

This post has been edited by KrisBlueNZ on December 14, 2012 02:52 am

yep ... we use links to uploaded photos .
at photobucket , hover on the picture , click on the "img code" , it will say "copied" ... paste the link into your post .

My suggestion would have been more-or-less the same as your circuit.
If you're a stickler for isolation,use a transformer instead of the cap/resistor dropper..or,as you mention,make sure the pot shaft is plastic!

I have a IP phone. But it's haven't any battery. So when electricity gone then phone is fully stop. At this moment I need a battery backup for my ip phone. It's have 5v 1a adapter.

Now please help me how can I insert a battery in my IP phone?

posted in wrong section .

"clean up , isle four !"