Week 9 - Prototyping Meeting (Lever Mechanism Testing)

Today we met as a team to explore the possibilities for igniting and extinguishing the gas flame. We thought it would be beneficial for all members to be present for this experimentation as this forms a large part of the product output interface and all needed to understand the mechanism and be happy with the output produced. So we met at John's house and began the process of mocking up the mechanism.

So leading on from my last post, the next step in prototyping was to test the initial gas ignition/release concept that I had sketched on the run. As you can see in this sketch it involves a pivoting wedge that is controlled via a single rotary arm fitted with a sprung wheel. As the wheel travels past the pivoting wedge it moves the wedge in such a way that depresses the gas release lever. This mechanism is repeated for each gas outlet around the flame dial. An electronic igniter controlled via the circuit ignites the gas which enables the rest of the flame dial to be ignited sequentially. This is crucial to the interface that needs to display the percentage of the group that is working on the project. 

Initial diagramatic sketch of the gas release mechanism

This diagram only shows the top view of the mechanism so a number of elements needed to be constructed to control the depth of elements within the mechanism. These elements include the pivoting wedge, the supporting frame and the rotary arm that supports the wheel (the aim is to control this with a single 360 degree servo motor).

The first part constructed was the wedge as the combined height of the lighter and the wedge determined the height of the frame and rotary arm.

A balsa wedge was used to achieve the correct shape for the wedge

Next element to construct was the frame. Using some salvaged meccano parts the frame was constructed to support the wedge and the lighter. Here you can see the wedge rotates with the screw a a pivot point. The lighter is in position with the gas lever ready to be depressed.

Lastly, rotary arm with the sprung wheel was fabricated. Again, more meccano parts were used along with a spring from a retractable pen. This actually provided too much pressure so the flex in the frame ended up providing the pressure required.

Side view of rotary arm

Here you can see the wheel retracting and returning to full length under the springs pressure.

The last thing that we needed to test was the length of time that the lighter can remain lit for. The packaging warns against the lighter being lit for longer than 30 seconds however we needed to test to see what would happen over an extended use. After a few minutes the  surrounds of the lighter top began to melt and catch fire so we decided that the duration of the prototype demonstration needed to be kept to a minimum while still proving the concept. 

This raised concerns about the materials that would be used in the final product however John will address this later in the design of the final object.

Today was a very productive session as we finalised the flame dial mechanism and allowed both the Object and Behaviour sides of the project to continue to move forwards knowing that the design was possible with the use of a single electronically controlled part, the servo motor. Unfortunately the prototype was not recorded in action so I will be posting up a record of its operation over the next few days. Again... watch this space!

Week 9 - Prototype Testing (ignition mech)

Today's excitement came in the form of a brainwave for the gas release mechanism. Initially we had been discussing the use of a solenoid for each of the 20 lighters to achieve the desired output from the interface. I think I now have a couple of ways of working around this. Each involves a mechanical switch however they only require a single 360 degree servo motor.

Leading up to this concept wer a number of ideas that centred around a series of levers and cords that ultimately could not work. Here are a few initial sketches:

 

So I switched the idea of a lever with a wedge to depress the gas release lever. A wedge could be wedged between the top plate and the lever to release the gas however removing the wedge to extinguish the flame was the difficulty. This is what I came up with:

Initial diagramatic sketch of gas release mechanism

The sketch on the top page is my documentation of the idea. It's clear in my head however I think only a prototype will show how it works clearly enough for anyone else to understand. This concept seems promising however I thought I'd come up with an even better idea. The switch mechanism on a ball point pen is a linear push to turn on-push to turn of mechanism. By using the linear motion to drive a wedge onto the gas release lever I was hoping that the servo could activate and deactivate each switch in sequence around the circle. Anyway... more pictures to demonstrate the progress that I made:

The work station

Dismantling the ballpoint pen to retrieve the central mechanism

Replacing the ink reservior with a wooden skewer for rigidity

Reassembling the mechanism

The ballpoint pen reduced to a simple linear switch (off position)

The ballpoint pen reduced to a simple linear switch (on position)

The linear switch in the off position mounted on a bracket to activate the gas release lever

The linear switch in the on position mounted on a bracket to activate the gas release lever

Unsuspecting lighter awaiting their fate (hopefully not impending doom!)

Lighter ungergoing testing... unfortunately the mechanism isn't strong or rigid enough to depress the lever consistently.

Unfortunately this mechanism hasn't worked this time round. For a proof of concept model I'm not sure we'll have the capabilities to successfully manufacture/hack a small enough and strong enough mechanism for the job.

Next is to revert to the initial brainwave idea with the mechanical wedge switch from the original sketch.

Initial diagramatic sketch of gas release mechanism

As mentioned at the beginning of this post I will need to prototype the mechanism and video it in operation or take progressive photos to demonstrate the idea. Watch this space!

Week 9 - Prototyping Exploration (Lighter Testing)

Today we also purchased a number of cheap lighters that we could use in our proof of concept prototype. I've been playing around with them tonight to see how we can use the to show how the product works and how the output will look. A picture tells a thousand words and videos can tell even more so here's my progress on this front so far.

Left to Right: Rhino Lighter, Bic Lighter

Removal of flint and gas release mechanism on Rhino Lighter

Gas release lever reattached

Close-up of gas lever attachment slot

Lighter stripped completely to reveal twist adjustable nozzle (twisting the white section increases or decreases the gas release and flame size)

Gas release lever reattached

 

Close-up of gas lever attached

The following video shows the flame on a higher setting... aside from singeing eyebrows it also shows the possibility for flame size control. This however may not be all that advisable as the flame becomes much less stable and controllable at higher settings.

A couple of obervations regarding the use of this lighter in the proof of concept prototype:

• It seems that it requires less pressure to release the gas than the Bic Lighter
• Also seems to need to be depressed less also

This may assist in the mechanism design over the next week. I will continue to think on this over the coming days to try and nut out the mechanism that can activate and deactivate the gas release to ignite and extinguish the flame.

Week 9 - Design Freeze & Round Table Discussion

So today we had our round table discussion where we were required to present our product concept as it is, preferably as a design freeze where we are able to discuss and consolidate our product concept before proceeding to finalise the design.

We needed to discuss the product narrative, as laid out in Kahlia's Scenario Blog entry (Week 9: Narrative, Brief and hard evidence SORTED! )

The following components of our design were required to be demonstrated:

• Fan speed controlled by number of flames
• How this is facilitated through the Magnetic Levitation
• How the gas flames are ignited from eachother 

Each of these elements have been explored through proof of concept models so the easiest method of demonstrating the design freeze was to use video footage from each model. The following video was used to enable the discussion on further development:

As the final notes in the video explain, our next steps are to:

• Get the Arduino Levitation working
• Explore the use of piezo ignitors to light the first gas flame
• Use solenoids to control the output from lighters in the proof of concept model/prototype (I'm not sure that using a solenoid for each lighter is the best way for us to activate each of the lighters... simply the cost of this would render the project inviable... it would be better to try and use a single mechanism that can adjust the number of lit flames on it's own)
• Explore the interaction between social networking sites and the product

Additionally, we need to:

• Purchase the lighters for the proof of concept model
• Nut out the mechanism for the lighter control (without the use of individual solenoids)
• Collect all components for the Arduino Levitation Circuit
• Develop a form for the Object

We did actually manage to explore some options for a form today however John will be exploring this further over the coming weeks. Here are some initial scribbles:

We will also be meeting outside class this over the next couple of weeks to pool our resources in modelmaking tools and parts to nut out the mechanism for the ignition system.

Week 8 - Evidence to Back Up the Concept FOUND!

Exciting news today!!! Kahlia has found concrete evidence to suggest that visual simulation actually improves motivation and attention. See here for Kahlias full description:

Week 9: Narrative, Brief and hard evidence SORTED! 

This valuable information gives our product a whole new level of meaning and further backs up our argument for the positive effect that our product can have on a collaborative team. 

This now puts us in a really good place to freeze the design for tomorrows round table discussion in class.

Week 8 - Arduino Levitation Research

The Magnetic Levitation component of our product is quite critical to the operation of our product (we really need a name for it!). Since we have seen that it is possible with the use of an arduino circuit and programming I have been looking into how it has been done before. The circuit that was used is very straight forward and is explained on this blog entry Arduino Magnet Levitation – detailed tutorial. The programming is however much more difficult however the files that are needed to run the circuit are also posted at the bottom of this the Arduino Magnet Levitation – detailed tutorial.   

http://mekonik.wordpress.com/2009/03/17/arduino-magnet-levitation/

This diagram shows the basic layout of the components involved in the visible part of the magnetic levitation system. 

• The coil is the electromagnet that is adjusted to attract the permanent magnet below.
• This is controlled by the information sent from the sensor that reads the strength of the permanent magnets magnetic field.
• The sensor samples at a very high frequency so that the electromagnet can continually adjust its strength to hold the magnet at a set distance.

These diagrams are the circuit diagrams for the arduino board.

Driver Circuit Diagram - Output (http://mekonik.wordpress.com/2009/03/17/arduino-magnet-levitation/)

Sensor Circuit Diagram - Input (http://mekonik.wordpress.com/2009/03/17/arduino-magnet-levitation/)

The circuit diagrams appear to be quite straight forward though it will be a matter of getting the right components to put the circuit together.

To further understand how the prototype will work and how it could be implemented in terms of the interface and user interaction I have quickly looked into how the hall sensor works.

I found that this site explains the three different types extremely well: 

bildr.org/2011/04/various-hall-effect-sensors/

I now understand why I need a linear Hall sensor and how it can be implemented to gain a particular set of information/input readings.

Below is a checklist of the components that I require:

• Arduino Duemilanove
• Switched power supply 12V
• Linear Hall effect sensor Honeywell SS19
• Norton operational amplifier MC3401P
• NPN transistor MPSA06
• Rectifier 1N4001
• Electromagnet (I used a coil from a 12V push-type solenoid)
• Resistors 2x 1k, 5k6, 47k, 68k, 330k, 4x 1M
• Capacitors 2x 1μ

I will aim to have all of these within the week so that I can have the circuit together by week 10 to that the prototype can be finished as there is still a lot of work to be done on igniting the gas and the mechanisms behind that for the proof of concept prototype.

Week 8 - Testing the Concept (Group Workshop)

We've had an exciting day today! We hosted a group workshop to test the input and output of the concept that we came up with after Tuesday's discussion. We looked into the Object and the Interface/Prototype Behaviour and decided on technology that could be implemented. We all met at my house with boxes of useful looking things to test the concept described in my previous post.

Gold Christmas Candle Pyramid

To start off with I had a Christmas Candle Pyramid to test our concept on and to transfer the principle to another form.  Unfortunately we were unable to test the difference in speed between a single flame and multiple flames and so an alternative model conceived.

After much thought we constructed a simple frame from which to hang a fan. The fan was sourced from an old computer cooling fan and the initial frame was from a previous project.

 

This is the fan that we used for to test the concept. A cotton thread was tied to the central pin to allow it to spin relatively freely. This type of fixing caused difficulties with balancing the fan so the pin was removed and the thread attached through the remaining hole.

Fan with pin in place

Fan with pin removed

Initial attachment of cotton thread

Improved attachment of cotton thread

Initially the fan was attached to the frame using a light weight cotton. Other materials such as fishing line and stronger cottons were experimented with however they provided too much twisting resistance. However I have made a new frame that sits directly over the shield so that I can more easily make adjustments to the system. The initial attachment method exagerated the wind-up issues that we were having in the first place so I made adjustments to it to simulate a continuous frictionless bearing.

Having ensured that there was minimal resistance to twisting I was then able to properly test the behaviour of the interface that is to become one of our major outputs.

The following video demonstrated how the fan can spin up according to the number of flames and the amount of heat that is being generated.

As you can see in the video the use of one flame begins the rotational motion of the fan at a slow pace. As more flames are added under the fan it begins to rotate faster and as the flames are taken away it rotates more slowly. The major problem with this mechanical system is that there is still an issue of wind up in the cotton thread (also noticable in the video). 

We also made another discovery in terms of the ability to control the direction of the fan. We found that the direction of the fan depends on where the flames are drawing oxygen from. If the shield has a gap at the bottom the air rises causing the fan to rotate in one direction. If there is gap at the bottom the flames draw oxygen down into the shield causing the fan to rotate in the opposite direction. It was interesting to note the speed of this change in direction alsoThe following video demonstrates this discovery.

Returning though to the issue of the resistance against twisting. We looked at designing some form of frictionless bearing however we decided that this would require lubrication and maintenance to keep the bearing frictionless. It was at this point that I remembered seeing a magnetically levitated globe and how it rotated continuously without friction to slow it down. A quick search revealed the Stellanova Magnetic Levitating Globe that suspends a globe with an embedded permanent magnet under an electromagnet.

Another quick search revealed that this can be achieved with an arduino circuit and programming! Looks like we have a winner!

This circuit uses a Hall Sensor to control the levitation height. I will be looking into how we can get this working for the prototype so that we can demonstrate the concept. Not only does this solve our twisting friction issue, it also gives us an interaction interface that is easily performed by the user.

Here are some form sketches of how this interaction and outputs could be designed into an object. The concept now uses a circle of gas flames encased in a glass shield below a magnetically levitated fan. We have also identified the need for separate flame to signify the users being logged on themself.

circle of flames + individual

Based on this new concept the sketch below maps out the required prototype to demonstrate the workings of the product.

Now that we have a product with a particular input and output we were able to roughly map out a narrative for the everyday use of our product. Below is the draft narrative:

Below is a fleshed out narrative that Kahlia will work through to ensure that the product input and outputs match up with the senario requirements.

And now, a list of the things that I need to work out for the prototype and to be able to asses the interaction and behaviour of users with the specified input and output.

Today's group workshop was certainly productive!