Nitinol project by 3 Ats Members part deux

So I was originally going to put this in the old thread but I've decided since it's a totally new application it deserves a thread of it's own.

Some of you may recall our earlier thread entitled nitinol energy generator by 3 ats members. Well this thread is a follow on to that thread with one key wrinkle. Instead of Imamarty's amazing design as a generator we are instead using it as a motor to drive a bicycle.

This is the basic design of the mechanical actuator. The only difference being there is no generator in the loop now just the double drive screwdriver which normalizes the back and forth motion into rotational energy.

Here is a review of the Double drive screwdriver for those who are unfamiliar with this technology. Basically what the double drive screwdriver does is turn the back and forth motion of the nitinol actuator into 360 degree rotation instead so effectively the double drive is the drive shaft coming off of the actuator.

Finally here is the VERY ROUGH first revision design for the unit mounted on a bicycle.

As you can see here there are 16 individual nitinol actuators in a line separated by spacers that also guide the bending of the nitinol rods on the unpowered stroke. Not pictured are the electrical contacts that will be in position at the farpoint of the bend. These electrical contacts will serve to heat the nitinol wire which will induce it to phase change and violently straighten powering the actuator. Also of note is that each nitinol rod will be encased in a coolant sleeve per member sotaxme's suggestion for how to make nitinol phase change quickly reliably. I haven't worked out how we'll exchange the fluid yet but that will come to me in time.

So the basic cycle of operation is as follows
1. the bent nitinol rod will be heated using direct electrical stimulation through the contacts on the top or some other method.

2. Once the nitinol rod is sufficiently heated to induce phase change it will violently straighten. This violent straightening of the nitinol rod on one side will forcibly bend the now cool nitinol rod on the other side down to where it comes in contact with the electrical contacts. While it's pulling the rod down the syncromesh strand between the two nitinol rods will be pulled across the gear spinning the handle of the double drive screwdriver. This motion even though it will probably only be a 30 degree turn of the handle will be translated into straight rotational energy by the double drive mechanism.

3. Now the other nitinol rod is in position to be heated by the electrical contacts. Once it's hot enough to induce phase change this rod will straighten violently once again inducing a 30 degree turn of the double drive screwdriver handle in the opposite direction, but due to the gearing in the double drive screwdriver again this rotational energy will be converted to unidirectional rotational force.

4. Steps 1 through three repeat

As you can see even though the actuator will just be popping back and forth we'll turn that see saw motion into wheel turning unidirectional rotation.


Some things I'd like to point out here are:

1. Please don't laugh at my bike frame and crankset I'm somewhat new to autodesk inventor and was just trying to get a semi accurate representation done quickly so we could gather input from the ATS community at large.

2, In this drawing there are sixteen individual nitinol actuators side by side. WE have no way of knowing how many we'll actually need to give us a good powerful consistent turn of the wheels so I just stuffed in as many as would fit.

3. The power system for this is still undecided and is not included in the drawing

4. THis is still a very rough idea in it's infancy. It's liable to change quite drastically as we figure out how to get the system to work as efficiently and quickly as we'd like it to.

5. The sizing could also change very drastically depending on how much power we can get out of how small of a package.

6. The reason we have our motor idea depicted on a bicycle is that it's a high torque low speed application which is perfect for playing to the strengths of Nitinol.

Please feel free to let us know what you think and or any suggestions you may have.

Thanks in advance,

I'm a Marty, Omega Logos, and Roguetechie

Originally posted by roguetechie
Not pictured are the electrical contacts that will be in position at the farpoint of the bend. These electrical contacts will serve to heat the nitinol wire which will induce it to phase change and violently straighten powering the actuator. Also of note is that each nitinol rod will be encased in a coolant sleeve per member sotaxme's suggestion for how to make nitinol phase change quickly reliably.

What are you going to use to power the "electrical contacts"?
What are you going to use to keep the "coolant" cool?

Seems to me that all you're doing is taking the power to operate those two things I just mentioned, and running that energy through a woefully inefficient machine to power a bicycle.
Seems to me that you'd be better off scrapping your "engine" and just powering the bike directly off whatever external power source you're going to need to power the "electrical contacts".

reply to post by roguetechie


I will be interested in the outcome of this.

A Nitinol engine has been proposed several times and the suggested efficiency is always quoted to be very high, but too many of these projects just dry up and disappear.

reply to post by roguetechie


I feel so dumb. But, I want to follow this thread since it's fellow ATS members, attempting to change history.

I'll be the one sitting way in the back. If my eyes start to glaze over, give my chair a kick.

Good Luck...

Des

reply to post by alfa1


Well therein lies the rub alfa.

The question is can we design a nitinol power assist that's uses less power than a comparable electric motor?

The answer is we don't really know for sure so we're going to give it the old college try.

The reason I personally believe we can make this work is the fact that you need a vastly oversized electrical motor to power a bicycle because of the torque required. Because of this the motors don't operate very efficiently at all. Now contrast this with nitinol which is all torque and low end grunt, and I think we MIGHT just maybe be able to produce something that beats out the conventional motor configuration.

I'd just like to further say that half the fun is in the experimentation. I never feel more alive than when I'm chasing a solution to a problem or designing and trying to build something.

I'd also like to point out that this is a very rough sketch and the prototype by the time it comes to one will probably look nothing like this. I oversized many of the parts to make them more visible and to see how big it could be made if it had to be.

In addition I"m going to say that nitinol has a potential force during phase change of 55 tons a square inch. What this means is it's not going to take much nitinol at all to spin the cranks on a bike. Hopefully when we start to actually build we'll find that we can go with a much smaller and less numerous actuator array, If so then I think we could definitely give your traditional bike motors which take up to a kilowatt of dedicated power to run.

That's another big thing to consider is that nitinol only needs to be powered until it heats up then it phase changes rapidly so you wouldn't need to power it all the time... This is where I think we might be able to rack up savings over a conventional always on always struggling to power the load electric motor.

I wonder, if a form of radiant energy wouldn't be a better idea... we can control photons fairly well and just like electricity, we can turn it off and on. This might be useful, if there were a solar application... one that focuses light and systematically use said energy to heat the rods. That's where gold and its control and utilization of radiant energy comes into play. You'd have to keep the temperatures down, but... individual shields for each rod could be created as to not allowing the energy absorbed by each individual rod of affect one another... as well, it could divert unused energy back into a loop.

This might not all be possible at the given moment, but from articles I've read... proton emission, and shooting releasing one proton at a time is becoming possible as quantum computing development furthers. If these rods were created in such a fashion that a good amount of torque can be created in a low range of temperatures, there might just be a great idea here.

Granted, it's far fetched... but I don't see any reason why solar energy could not be, or should not be applied to something like this.

while solar energy might work in theory I think using electricity is a much better bet since we're trying to power a bicycle in all weather all conditions.

I appreciate the input though.

Originally posted by roguetechie
while solar energy might work in theory I think using electricity is a much better bet since we're trying to power a bicycle in all weather all conditions.

I appreciate the input though.

What if you could power this bicycle from extreme distances using a laser? Kinda like the theory of powering planes from the ground with lasers... Now that's something! This is obviously limited by bad weather... so yeah, maybe not so applicable here. But if towers with lasers were created, and recognized these engines... I'm thinking more along the lines of trains and commercial use. Although, that doesn't remove the possibility crossing over into daily driving.

I see that this design is crafted using rather 20th and 21st century technologies, and the ability to demonstrate the basic principals is a must. I just feel that this should be expanded as to keeping up with the times. So for now, I like where this is headed... but this idea will get quickly left in the dust, for it is a rather crude, mechanical/electric design. Obviously there's no need for concurrent engineering, but one thing you should consider is rapid prototyping, the experienced gain here will be most beneficial... more so than theorizing, granted there are some great CAD programs that seemingly remove the need of actually producing anything more than a working model.

Any who, have fun!

Does nitonol not return to the shape in which it was manufactured once it has cooled?

How much energy does it take to make it react when heated? How much energy does it take to bend it back when cooled?

Iam in.

I think your best bet is the pully system, it has a cold/hot water tank that I've seen on a youtube video. That way, you have a more simple time with cooling/heating and drive. You could power it with black tubing over a large surface area, like 3/8" plastic tube over the bike to provide shade. For cooling, dry ice, CO2, big heat sink. If nitinol is 80% eff, you could get 1-2KW of energy depending on how you heat/cool it. You could make the pulleys out of 1/8" thick plexy sheets where one sheet would be a 3 1/4" circle and the next would be 3" and would alternate (bevel the edges).

OK Fractal Chaos I"ve broke down your questions and answered what I hope is all of them in a satisfactory manner. Without further ado here's my answers to your questions.

Powering the bike from extreme distances with a laser:

Since we are just a trio of ordinary guys I don't think we could even get our hands on the right kind of laser let alone use it to power the bicycle. In addition even if we could why would we? By making a product dependent on said laser towers you'd have a bicycle that is literally the macintosh of the road. It would only work on a small percentage of areas where we could beg steal or borrow the money to put in towers, and the rest of the time it would be dead weight. Also in order to power the bike with lasers you'd have to have a solar collector cell and the exact same electrical wiring as we'll need anyway to run off of a battery pack.

Rapid prototyping over designing in autocad:

We fully intend to construct a working prototype to test. I use Autodesk Inventor to help me envision my basic ideas just to see if they're at all remotely feasible. Once we get to the prototyping phase it will turn into the age old game of sourcing parts from anywhere I can find them that will determine much of the device's characteristics and initial look. I believe firmly in sourcing off the shelf technology as much as possible to keep costs down and speed up the pace of builds. So hopefully that answered your question in this respect.

Does nitinol return to the shape it was manufactured in when it is cool:

Short answer is no. Nitinol like other materials when not engaged in one of it's shape memory states will go into whatever shape outside stimulus pushes it into.
The longer more indepth answer is that Shape memory nitinol can be trained to respond to both hot and cold temperatures. So in theory you can have one piece of nitinol that can be trained to assume two different shapes one by heating and one by cooling and a third default shape caused by environmental stimulus. To train nitinol to respond to heat you must have a very hot oven (around 600 degrees I think) and you must have a jig designed to hold the nitinol in the shape you want it to phase change to when heated. You then put the nitinol in the oven for a differing time period depending on what temperature you want it to phase change at. After it's been in the oven for the requisite amount of time you pull it out and let it cool naturally back to room temperature. At this point your nitinol is "trained". Now the cool thing is once your nitinol is trained it's reaction will get stronger the more times you have it phase change. SO unlike an electric motor which gets weaker over time nitinol is known to stay relatively unchanged over millions of cycles.

How much power does it take to make nitinol phase change?

You know that is a really good question and it's one of the questions that I've never found a sufficiently good answer to. I think this will only be determined by experimentation. I plan on experimenting with pulsed DC power as well as conventional DC power and maybe even AC power to see which way most efficiently brings our nitinol to phase change. Once we've determined this we'll be using a contactor system set up in the guide walls that are designed to keep the nitinol shaft from bending in any direction but the one we want it to. This will have the added advantage of self regulating the amount of power going into each nitinol strand because the strand will move off of the contactor as soon as it's warm enough to induce phase change. I'm hoping this self regulation will help keep the nitinol in pristine condition giving the motor a long service life.

How much energy does it take to bend it back when it's cooled? Again that's another good question .... I'm hoping the answer is enough energy to force the syncromesh wire to engage the gearing in a consistent and slip free manner that will transmit an optimal amount of power to the shaft. THis is part of what we're going to have to experiment with to get it right.

reply to post by XL5


In response to your idea about using water to power the bike motor I'll just say we did consider using water to power them somehow but every time we did the hypothetical device quickly became too complex and heavy to be used. Not to mention you'd probably still need electric power to heat the water in the event of riding at night or on cold days. When you factor that in it becomes a less than zero sum game as you'd have even more efficiency losses by heating the water instead of directly heating the nitinol with electricity. Also using dry ice would be cumbersome and inconvenient.

I'd also like to point out that while I"d love nitinol to be 80% efficient at converting heat to energy I'm not sure if that's the case or not. I do however think that in this application where you need high torque and a relatively few rotations per minute it has potential to be a winner. I also think it will give a more natural feel to using a power assist over an electric motor. Between these two areas I sincerely hope we have come up with something worth building.

The only problems I see with the current design is the cooling, you could only use forced air and that limits how fast you can cool the nitinol and you would have to shut the air flow down once the wires are heated. Using water in that design would not really let it heat back up that fast. IMO it's better to have a constant hot/cold side because you could run it off propane if you wanted to.

That said, with your current set up, I would use magnetic reed switches connected to relays or mosfets and magnets near the bottom of the nitinol (insulated of course). That way you can set the hysteresis and thus how long the wires heat for.

reply to post by XL5


Actually with the design as it's currently setup there would be a sleeve around each nitinol wire/rod which I Could run a secondary stub pipe off of the top of the square contactor points/ guides that way I could force water or ammonia through the bottom and up through the top with a basic radiator setup if it proves to be needed.

Rather than going with reeds and all of that complication I am just going to go with a pair of copper contact points set at the low point of the rods dipping motion that way it can start inducing current the second it's bent down and will stop pushing current the second it starts to phase change.

With coolent jackets, how do you plan to make an air bubble when the time comes to heat the nitinol? Also, with the contacts, make sure to use a spring with a lot of travel because you don't really want/need high amps for a split second burst that may not heat the nitinol enough. You want lower current for a longer time, research the puekert effect on batteries.

I'm not an electronics expert by any means although I will be taking my basic electronic circuits course in college shortly so that should help.

ANd as far as creating an air bubble I actually got the coolant jacket idea from member sotaxme who suggested it as an optimal way for nitinol to be utilized. I don't think you'll need an air bubble at all if he is right.

You will need an air bubble, otherwise you would be heating the fluid and not the nitinol. It would sort of be like plugging a cars tail pipe to make it less noisy.

So you need to have your wires inside a coolant bath that is kept below the transitional temp of the wire at all times, and apply electrical current just enough to overcome the bath temp. Then you have extremely fast operation.

This is a quote from member Sotaxme in the nitinol thread. Now bear in mind that sotaxme was a lab assistant to one of the two discoverers of Nitinol and I think it holds some weight.

Will heat be transferred to the fluid? Undoubtedly

Does this change my idea at all? Not really since I already knew heat would be transferred to the fluid and hence the fluid would have to be moved through a radiator at some point to keep it cool.

This is why I specifically listed ammonia. With it's low boiling point I could engineer the system so the boiling fluid wants to travel to the top of the enclosure around the nitinol and out of the top constantly making room for new cooled ammonia.

reply to post by roguetechie


The question has arisen as to how efficient Nitinol is throughout its temperature transition. Considering that one mole of Nitinol (weighing 114.5 grams) has a volume of 17.75 cubic centimeters (or 1.08 cubic inch), and an alloy with a TTR (temp transition) of 55 degrees C produces only 551 Joules per mole, the efficiency of Nitinol during a 4% strain would calcutalte out to only 19.8%.

I like the fact that members are putting on their thinking caps in reference to delveloping Nitinol motor applications, but in this case, but your bicycle is going to be the size of a house, and too heavy to move.

Please keep in mind that Nitinol is strongest in its linear applications when a wire contracts. I have witnessed a 4x4 of wood snapped by a wire about 1/8" in diameter as the alloy hit TTR. Of course that wire had only ~6% contraction from its original length, but this demonstrates the power that takes place on a linear scale.

Keep thinking. You are only limited by the limits of your imagination.

73