December 5, 2012 at 20:37Wooden mouse. Project history
Have you ever wondered how things are arranged, what way they go from an idea to implementation, how simple things are? How easy is it to make a comb? What about a computer mouse? And a wooden computer mouse from a solid mahogany block with an LCD screen, with its own electronic filling and a cable made and braided specifically for it? I think you will be interested in my path, which I went through 2.5 years of creating my mouse.
Design, construction, modeling
Since in design I was a complete zero, then I approached the matter as a complete ignoramus. I bought plasticine and began to sculpt the mouse of my dreams.
At first I made a mouse, which is ideal for me to work on the desktop. She is large dark gray in the photograph. Then I dazzled a mouse that would suit me as a mobile (dark gray small). And then I took the piece of plasticine stolen from the children to work, and my colleagues fashioned a mouse that claims to be “national”. She ideally fit the hand of most of the male population of our team (multi-colored in the photo). And what? The result was banal and dull forms that we jerk our hands day and night in every way. Apparently, among the three standard mice, any user will find it convenient. The triumph of the ideal?
As a result, a mouse was simulated at the computer, which, from my point of view, claimed to be elegant and beautiful.
At that moment I really liked her. And without thinking twice, I divided the computer model into parts. The elements of fastening and pairing with electronic filling were thought out. It sounds simple, but in fact hundreds of hours of painstaking work have been spent.
After that, the obtained parts were grown on a 3D machine to verify collection.
The material is polyamide. He sits well in his hand, like a glove. All parts fit together, the technological assembly also went smoothly.
The next step is milling in a tree. Probably, I got a dozen different species of red trees, but I started with a sapele tree, the rest of the species are waiting in the wings.
I did not like the live design. The vertical gaps between the buttons and the case looked bad and untidy. You can see the technological “sores” when working with wood - chips and tree removal. Well and most importantly - the keys did not bend, there was no click.
I thought about the design for a long time. Something was embarrassing, and there was no feeling of satisfaction. Then I realized - the mouse lacks solidity. I decided to return to the original version of the mouse, which I sculpted at the very beginning, only on a professional level and using sculptural clay. Two design options were made in one mouse. Convenient for comparison and decision making.
After receiving the final version, 3D scanning and transfer of surfaces to SolidWorks was done.
The second model was not much more successful than the first. Buttons were not pressed, and it was impossible to fix this in the current model. The marriage of the model was laid at the DNA level. A more comprehensive approach is needed with the simultaneous control of both design and technology. Otherwise, nothing will work. There will be either technological excellence or good design, but not all at once. These characteristics sit on opposite sides of the swing. Therefore, I throw everything in the trash and start from the beginning. Sketch-design-modeling-testing-growing, and so on, but with technological control of critical parameters on the one hand, and design on the other. We are looking for a middle ground.
The third model was already made as part of the classic product design cycle. Started with a sketch.
Outlines were drawn.
And finally, an approved design.
Plasticine mockup.
3D scanner receiving a surface.
Computer model.
Then the process of finishing the hull began. At the CNC machine, the body was sawn, tested, finalized, then sawn again. As a result, only the tenth version of the case turned out to be workable. The biggest problem was making keystrokes comfortable. As a result, in some places the thickness of the tree decreased to 0.7 mm! It took me a year to finalize the case.
A wheel and a connector were also made of wood.
On the wheel I applied laser engraving with the Clickwood brand.
On the way is the eleventh version of the case, where I will make minor changes. I also started developing a wireless version of the mouse. The wireless module is based on bluetooth technology, the optical sensor is laser. AAA batteries, 2 pieces, replaceable. When recharged, the mouse will continue to work. All elements are located very densely, with the layout I had to pretty much break my head. As a container for batteries is a cavity specially cut in a wooden mouse case.
Wooden details
Work with wood begins with the selection of wood. Boards must be of the correct geometry, have a minimum of knots and defects, and have the necessary humidity.
At first, the boards are dried at home. At least six months.
After that, the board is sawn into small bars, which are dried for several weeks at the place of their further processing. At all stages, humidity is controlled by a special device. If we neglect the drying process, the tree loses geometric stability, and the manufacture and operation of the mouse becomes impossible.
The prepared bars are processed on the CNC machine using a specially created program.
From the very beginning of the creation of the part and until the final assembly of the mouse, the parts are rigidly fixed on a metal snap, so that at no stage does the part change its shape and geometric dimensions.
The processing of the upper part of the mouse has to be done with jewelry accuracy, since its profile is designed for a soft click and in some places is very thin. I control the pressure with a gramometer. In ordinary mice, it ranges from 50 to 75 GS. I try to achieve 50 GS.
The greatest difficulties are associated with the tree in my project. Not only is this the most significant part of the cost, but the proportion of marriage here is very high. Wood is an anisotropic material. It can lead, flaws can get caught, chips may appear, and just a mistake in the technology of topcoat can lead to the mouse body being sent to the trash. I admit that I still improve the processing technology, and I'm not completely sure that I found the right one. For statistics: in the first batch of ten buildings, only three reached the finished product. Therefore, part of the process chain associated with the tree is critical to the cost and quality of the finished product. Over it is ongoing work.
In the future I plan to work with bone. In particular, I am already creating a bone wheel.
Electronic part.
I developed the first mouse circuit myself. As a sensor, I took Avago's top-end optical sensor ADNS-3090, Atmel's controller became the brains, the remaining components of branded companies such as Murata, Yageo, Geyer, Omron and Molex.
I paid special attention to the quality nutrition of the mouse, here, in my opinion, it reached its absolute perfectionism :)
In order to understand how the sensor works from and to, I completely translated the optosensor specification into Russian.
Then the first version of the board was developed.
The first working layout.
The second version of the board, the second breadboard.
The third.
Fourth.
In black, final.
There have also been experiments with various buttons. I have always tried, among others, to pick up a quiet mouse. Well, since I do it myself, I decided to conduct an experiment and make such a mouse and try it out in work. To do this, I replaced the left and right “mikriki” with soft and quiet ones that were used for the central button (noticed that the central button always clicks quieter?). A special version of the board was created, where all three of the same "mikrika" were mounted.
The mouse was in operation for a couple of weeks, and I can share the conclusion: in my case, the absence of a click is not comfortable. The tree partially absorbs the click, and it seems that the wooden keys are tight, and you have to press harder. It is difficult to describe the sensation, but the closest analogy is as if under the finger there is a mouse button from soft to breadcrumbs, and for a confident click you need to squeeze it. The experiment can be considered successful, since a clear result was obtained which should be the response from a mouse click - clear, clear, unambiguously interpreted.
I picked up and bought a batch of gold-plated connectors for the mouse. As usual, in China. I don’t know about the “best contact”, but they harmonize perfectly with the tree.
At the moment, a project is underway to develop a wireless version of the mouse.
Screen, firmware
Carried away by the idea of placing a display in the mouse, I began searching for it among hundreds of suppliers. The requirements were simple: tight overall limitations and the possibility of at least a symbolic display of at least eight familiarities. While picking up, I learned almost everything about displays. They differ in types: symbolic and graphic, in technology: TAB, COG, TFT, OLED, LCD, E-Paper and others. Each type or technology has a lot of varieties, sizes, colors, lights, etc. In general, there was something to delve into.
Half-Internet break, I found out that the size I need is made by only one company in the whole world. All other options are definitely larger in size. And even the display I found barely fit inside the mouse. As an option, a custom display was considered, which I could make according to my requirements, but this is a very expensive option for me (about one hundred thousand rubles). For the first model, a graphic display with a resolution of 128 by 64 pixels is quite suitable, which I chose.
In order to understand how the display really looks and combines with my mouse, I had to order all varieties of this display from the manufacturers. What do these varieties mean? The model name consists of alphanumeric unpronounceable combinations of the type FP12P629AU12. All of them are composed of various blocks and are clearly decrypted in the specification. For example, the above example can be assembled from FP.12.P.629A.U12 blocks, where the type, size, voltage, controller, operating temperature range and other model information are encrypted. And the last block is the trickiest. It can have several tens of values, each of which means a particular combination of such characteristics as the presence and color of the backlight, background color, character color, range of degrees from which the information is clearly read. These parameters were just interesting to me.
As a result, "for testing" I ordered 18 different modifications. The manufacturer agreed, but said that the minimum order is 5 displays for each modification. There was nowhere to go, and I had to agree, knowing that 90% would go into the garbage can. And on one of the cloudy days, the express delivery service brought home a huge box in which you can live a bum of average build. There were 18 smaller boxes in the box, each of which freely placed 5 displays, securely fixed for a long trip to cold Russia. There were so many accompanying packaging that it was enough for my mother-in-law to cover several beds for the winter.
As a result, after careful tests at a specially assembled stand, two displays turned out to be suitable for the series. They differ only in background: gray and yellow-green. It’s them that I will offer to complete the mouse. By default, I plan to set yellow-green, but two more options will be available: a display with a gray background and a mouse with no display at all.
But the main intrigue was what information can be displayed on the screen? I was offered different ideas: the temperature of the air, an indication of the arrival of letters, something else not very original.
My train of thought went a different way. To begin with, there are two significant restrictions on the display of operational information: the presence in front of the user of a huge and high-quality source of any information (monitor) and the need to turn the mouse over to receive information. In addition, the screen is small, the resolution is small, the LED interferes with normal reading. Therefore, my conclusion was one: the information should be only entertaining in nature, the applied value of which tends to zero, but at the same time, the WOW! Effect should be killer.
What information can possess such properties in a mediocre device? Its a little: mileage, time of use, speed, number of clicks and scroll wheel. I decided to refuse the last parameter, since it seemed to me uninteresting. The rest of all parameters are associated with the session (the last time the mouse was used from the moment it was powered up, i.e. connecting to the computer or turning on the computer itself) and to the entire time the mouse exists. For example, a user can at any moment of the mouse find out how many times he pressed the left mouse button or how much his mouse has run in meters for today or since its purchase. The information is absolutely useless, but it will help the curious to understand how much he torments the mouse. If other interesting ideas appear, they can be implemented with new firmware.
Also added general information on the mouse (model, mouse and firmware number, month of manufacture) and the settings screen. It will be possible to choose the language and system of measures (English or metric). To store all this information, I had to add flash memory of permanent storage to the circuit.
To put such a volume of information, I had to split everything into screens. Each type of information is displayed on each screen, and the parameter values for the session and for the whole time are displayed. In total, six screens turned out that change with the mouse wheel.
The first option was implemented in a purely textual manner, for which several font options were even developed.
I made firmware to evaluate how the text looks using the created font on the mouse’s screen. It looks awful what to say.
Now it has become apparent that the screen needs graphics, not a set of symbolic information. Therefore, I hired a designer to work, and together we prepared three options for graphic execution, as a result, the second option was recognized as the most successful.
Of course, such a design required more resolution, so it had to be adapted.
But this is not the end of the story. After I picked up the screen for the mouse, I made an order for a trial batch for mock-ups. As a result, screens came, but for some reason, the number of conclusions is different from what is specified in the specification (datasheet). The manufacturer received a response to the request that, they say, everything is fine, this is a small modification, and it will not affect the performance in any way. Meanwhile, the missing two conductors were responsible for the brightness of the displayed graphics.
All this was very suspicious. And just as he looked into the water. We remade the board for a modified screen, soldered it, and then it turned out that the screen was completely dull. As if the device had run out of batteries. And it turned out after a long and painstaking work on the search and selection of screens, the purchase of a trial batch of all modifications and their tests. Time, money, and so on.
But the story ended with a good ending. After correspondence with the Chinese, it turned out that the screen can now adjust its contrast directly from the firmware. They treated the firmware, and everything began to show just fine!
Everything is shown as planned: mileage, speed, number of clicks and more.
In the future, the firmware also changed several times: there was a setting for changing the language. Two languages on one screen is bad - readability is deteriorating, an English-speaking user will only be annoyed by Cyrillic gibberish, and in the future support for other languages may be needed. Difficulties began when I tried to align the mileage of the mouse. It seems to be complicated: the optical sensor transmits the increment along two coordinates, which must be reduced to a system of measures and added modulo to the current value. That's the whole run.
But, as it turned out, not everything is so simple. Two people with mice where the same sensor is installed can get dramatically different results! The thing is that the resolution of the sensor (sensitivity) very much depends on the surface on which the mouse rolls. The best result is when the mouse rolls on white paper. Slightly worse on wood and fabric. On the laminate and the film is very bad. The declared passport sensitivity is achieved only on ideal, from the point of view of the sensor, surfaces.
For the end user, this does not matter. It connects the mouse and, by trial and error, exposes a comfortable cursor speed in the operating system. The system remembers this coefficient and uses it to increase or decrease the values of the increment of the coordinates of movement.
But it’s a completely different matter if you are planning to read these parameters directly from the mouse. A mouse on one surface will show the result of a mileage of one meter, on the other - one and a half. Speed will also “lie”. And something needs to be done with this.
To solve this problem, it was necessary to introduce the parameter "Sensitivity", which allows you to individually select the coefficient for each surface. By default, it is equal to one, which corresponds to the surface of white paper. It can be increased or decreased in the settings. You can not touch it at all, everything will work just fine. But for true perfectionists, a leaflet will be given in the leaflet attached to the mouse, from which you can choose the coefficient to the existing surface and instructions on how to independently configure the mouse to show the exact mileage.
When developing the firmware, another side effect of the sensor was discovered. If you take the mouse and just wave it in the air, then the mileage will also change. This is due to the fact that the sensor determines the surrounding space as a certain surface and also tries to obtain the values of the mouse offset. Therefore, you can observe such an effect: you turn the mouse over, look at the mileage parameters and are surprised that they are changing right in front of your eyes. Of course, you can install a tilt angle sensor in the mouse that disables the sensor for the time of its revolution, but it is unreasonable to do this only for the described situation. Perhaps in the next version he will appear, but not now. After all, the mouse is raised only to look at the indicators, and 99.9% of the time it is on the surface and receives the correct information.
Cable The
cable decided to make it as flexible as possible so that it does not interfere with the movement of the mouse and is “invisible” to kinematics. Well, I personally do not like the "spring" cable.
Sometimes it seems that when creating a product, the cable is the most non-essential part of the product. What is easier is to buy in the store the right amount of cable and solder it. Shitty thing. But, alas, not in Russia. Sometimes it seems that our industry is no longer able to do anything more complicated than cast iron irons. Attempts to find a cable resulted in a three-week search and shaking up the assortment of absolutely all manufacturers of Russian cable products. It turned out: our standards do not describe a cable suitable for modern electronic devices. For example, a four-wire microphone cable with a KMM 4x0.12mm2 braid has an outer diameter of 5 mm. This is a lot. Older mice and keyboards have an apparently thick cable, whose outer diameter is only 3.5 mm. The closest analogue on sale was the cable of the German company Lapp Kabel, but he also had an external diameter of just 3.5 mm. Now imagine also a braid on such a cable. Presented? I’ll tell you that I saw a similar cable on the power cords for irons :)
So, it turned out: you can’t buy such a cable in Russia. Point. Well, we are not accustomed to retreat. I’m going into production and trying to order, since in Russia they still make a cable. And for this we decide on my requirements. So, what I need:
Cores - copper, from braided wires (for flexibility).
Number of cores - 4.
Screen - yes.
Flexibility is maximum.
The outer diameter of the cable is strictly not more than 3 mm.
Color - Pantone 4625 C.
Bottom line: I tried to get in touch, probably with a dozen possible manufacturers of cable products, no one is interested in messing with my order. They didn’t even ask what mileage I needed. Bottom line: in Russia such a cable cannot be bought or produced. Sad But we are not used to retreat.
I'm going to Alibaba.com. I find the first Chinese manufacturer that comes across, I write a letter and literally a few hours later I get the answer: we will make any cable for you! I am in shock. I throw him a specification, money for delivery, and in a week I get a sample. Wow! And I lost almost three months, trying to patriotically place an order in Russia. It turned out that the Chinese can quite easily make me a cable with an external diameter of 2.5 mm.
Bottom line: I ordered 4 different samples in China. At first, the scratchability and dullness of the outer shell did not suit, then the cable did not suit the flexibility, then it did not suit the flexibility again, and finally settled on the last sample sent, which I was ready to order. They cannot be more flexible. The cable has a memory. As a result, I accidentally received a memory cable, although I wanted to be as flexible as a rope :)
I ordered a kilometer, after two weeks I had the cable. Total time spent: six months.
Braided his kilometer of cable. It turned out two options.
About 10% of the cable went into rejection. This is the beginning of the bays, where the braid is untwisted and the machine has not yet entered the operating mode. And some places where for some reason loops and knots of braid threads formed.
If the end of the cable is not terminated with heat shrink, then it will fluff at the moment, the threads are synthetic! Therefore, the installation of the cable assembly is hindered by the preventive shrink sleeve.
The outer diameter of the braided cable was 3.2 mm, i.e. the braid added 0.7 mm to the cable diameter. It seems a little, but with a normal mouse the cable comes mainly with a diameter of 3.5 mm, and in the era of wireless mice it seems thick and heavy. Recently, not budget mice have begun to be equipped with cables with a diameter of 3 mm, and they no longer interfere with work, they are almost imperceptible. But the keyboard cable can be with an external diameter of 4 mm. And even more. But for the keyboard this is not important.
Next ordered a cable assembly. Cables have two different lengths and two braid colors. So to speak, for every taste of the consumer.
Plastic parts
No matter how I want to make the mouse body parts made entirely of wood, plastic is indispensable here. We need legs, an axis for the wheel, support for the axis and a glass for the display.
Therefore, I had to order a mold from the Chinese.
After each test casting, the Chinese sent me a dozen samples, which I tested on my mouse.
As a result, I worked on the mold three times, until the quality began to satisfy me. The problems were different. For example, after assembly, I got a dust problem that forms between the display and the protective glass. It looks untidy. Moreover, the mouse will scrape over the surface, and dust will gradually accumulate there. I had to convert the glass into a container with sides, where the display will be inserted, after which the circuit will be sealed.
It turned out here is such a detail.
Refining the mold is a daunting task, and changes can only be made in the direction of increasing detail. Therefore, any inaccuracy or error can ruin the whole work. For reference: each revision is a month and a half of waiting for new samples. And the change itself could be microscopic, but necessary.
I will not dwell on plastic parts, this technology is now leading, and I can not tell you anything new and interesting here. I’ll only say about the legs, for which I have been selecting material with reduced friction for a long time, after which I have tested and run the mice in order to determine the winner with minimal friction. I described all this in detail in my blog.
Processing and coating
First, there is thorough work with the removal of pile, sanding and polishing the surface.
I faced a difficult task. It was necessary to stabilize the tree so that the geometry of the mouse did not change depending on humidity, and protect the tree from working in an aggressive environment (sweat and fat by hand).
From the very beginning I refused varnish. Varnish is a surface film that cracks over time, collapses, and the tree turns out to be bare. Sweat and fat penetrate the pores, the tree darkens, and the irreversible process of its degradation begins. Therefore, it was decided to use oil as an impregnation and protection, and wax to give a commercial look.
To make it clear: the tree is thoroughly saturated with pores, which contain either air or the oil of the tree itself (if the tree is rubber). Our task is to fill the pores with our oil as much as possible, which then should polymerize and protect the tree.
In order not to stretch the story, I’ll say that I have tried many oils: linseed, teak, tung, vaseline, Danish. Each oil has its own character. For example, wax is applied very poorly to teak oil, and linseed oil cures for a very long time. Therefore, it is necessary to introduce a catalyst - desiccant.
As a result, I developed two technologies. The first is the technology of vacuum impregnation of wood. It works like this: I create a vacuum in an environment with oil and wood. Air begins to come out of the pores. After removing the vacuum, the pores are filled with oil. As a plus, the tree stabilizes well. As a minus - it gets very dark. It looks good, but not for everybody.
The second technology is a surface oil coating. Oil is applied 1-2 or more times with a non-woven cloth.
Then we process with a felt circle.
Apply carnauba wax.
And rub the muslin circle.
Then, using a hair dryer, I “dissolve” the dry wax residue in narrow and difficult places. In the case of “insoluble” debris I pick up a toothbrush with a stiff bristle, remove the debris and then again repeat the procedure for applying wax locally.
All is ready!
If we evaluate the labor costs of processing, then manual labor per mouse turns out about four hours.
Assembly.
Next is the installation operation, but before it you still need to remove traces of processing from the technological holes. Then, using a special 3M tape, I align and stick the legs (the case can lead to fractions of a millimeter, and this will immediately be noticeable: it will stagger like a lame-footed stool). Then I lay the cable, mount the board, support, install the wheel and also, if necessary, adjust the buttons (there should be no bounce) and the force of pressing. This operation can also take up to four hours.
Logo, website, blog
Such a status wooden mouse cannot do without a cool logo. I will omit the intermediate options and show the result. He pleases me. The combination of leaf (wood), mouse (gadget) and wood color clearly indicates the scope of the company. And the inscription made using the notched font indicates the solidity of the enterprise.
The site decided to do in a low key. The tree looks best on black: www.clickwood.ru .
The main feature of the site that should be mentioned is the configurator. Any user can play around with the options and order a mouse for themselves. You can change the type of wood for each part, its impregnation, the length and design of the cable, the presence and color of the screen. Even if the Chinese copy my mouse, they are unlikely to offer such a service.
For those who are interested in the process itself, I cover every step in detail on my blog: kuliba.livejournal.com . There you can also find the project of a wooden keyboard and case, which I put aside.
Why did I do all this? At first it was a hobby. Then the hobby turned into a very expensive budget item, but there was no reason to stop, a lot was done, it was necessary to move forward. In the end, I made an industrial-replicated author's product. Spent 2.5 years and a large budget. It's no secret that I now want to create a small business. Perhaps my mouse will be bought as an elite gift to the head or just a unique gift. Indeed, until now there has not been a single status product on the computer market. Even if you are a Persian sheikh, you still have to hold a dull plastic craft in your hands. I exaggerate, of course, but the meaning is clear.
What plans do you have for the future? I plan to create all the elements of a workstation: mouse, keyboard, body and monitor. Perhaps along the way I will offer goods from the “desktop” category: pens, coasters, etc., also made using valuable varieties of wood and other exotic materials. Products are status, from the premium segment, but I strive to make them affordable for the middle class, this is the most buying category of people.
I would like to create a small workshop. To feed old age and bring joy to me and lovers of original gadgets. Not everything is completed to the end, not everyone I am satisfied. But I keep moving forward. Improving, perfecting and inventing.
UPD: I in no way pretend to be the palm and I am not the first to make a wooden status mouse, but in my post I want to talk about the history of my project, I didn’t set another task for myself. Thank you for understanding.
Design, construction, modeling
Since in design I was a complete zero, then I approached the matter as a complete ignoramus. I bought plasticine and began to sculpt the mouse of my dreams.
At first I made a mouse, which is ideal for me to work on the desktop. She is large dark gray in the photograph. Then I dazzled a mouse that would suit me as a mobile (dark gray small). And then I took the piece of plasticine stolen from the children to work, and my colleagues fashioned a mouse that claims to be “national”. She ideally fit the hand of most of the male population of our team (multi-colored in the photo). And what? The result was banal and dull forms that we jerk our hands day and night in every way. Apparently, among the three standard mice, any user will find it convenient. The triumph of the ideal?
As a result, a mouse was simulated at the computer, which, from my point of view, claimed to be elegant and beautiful.
At that moment I really liked her. And without thinking twice, I divided the computer model into parts. The elements of fastening and pairing with electronic filling were thought out. It sounds simple, but in fact hundreds of hours of painstaking work have been spent.
After that, the obtained parts were grown on a 3D machine to verify collection.
The material is polyamide. He sits well in his hand, like a glove. All parts fit together, the technological assembly also went smoothly.
The next step is milling in a tree. Probably, I got a dozen different species of red trees, but I started with a sapele tree, the rest of the species are waiting in the wings.
I did not like the live design. The vertical gaps between the buttons and the case looked bad and untidy. You can see the technological “sores” when working with wood - chips and tree removal. Well and most importantly - the keys did not bend, there was no click.
I thought about the design for a long time. Something was embarrassing, and there was no feeling of satisfaction. Then I realized - the mouse lacks solidity. I decided to return to the original version of the mouse, which I sculpted at the very beginning, only on a professional level and using sculptural clay. Two design options were made in one mouse. Convenient for comparison and decision making.
After receiving the final version, 3D scanning and transfer of surfaces to SolidWorks was done.
The second model was not much more successful than the first. Buttons were not pressed, and it was impossible to fix this in the current model. The marriage of the model was laid at the DNA level. A more comprehensive approach is needed with the simultaneous control of both design and technology. Otherwise, nothing will work. There will be either technological excellence or good design, but not all at once. These characteristics sit on opposite sides of the swing. Therefore, I throw everything in the trash and start from the beginning. Sketch-design-modeling-testing-growing, and so on, but with technological control of critical parameters on the one hand, and design on the other. We are looking for a middle ground.
The third model was already made as part of the classic product design cycle. Started with a sketch.
Outlines were drawn.
And finally, an approved design.
Plasticine mockup.
3D scanner receiving a surface.
Computer model.
Then the process of finishing the hull began. At the CNC machine, the body was sawn, tested, finalized, then sawn again. As a result, only the tenth version of the case turned out to be workable. The biggest problem was making keystrokes comfortable. As a result, in some places the thickness of the tree decreased to 0.7 mm! It took me a year to finalize the case.
A wheel and a connector were also made of wood.
On the wheel I applied laser engraving with the Clickwood brand.
On the way is the eleventh version of the case, where I will make minor changes. I also started developing a wireless version of the mouse. The wireless module is based on bluetooth technology, the optical sensor is laser. AAA batteries, 2 pieces, replaceable. When recharged, the mouse will continue to work. All elements are located very densely, with the layout I had to pretty much break my head. As a container for batteries is a cavity specially cut in a wooden mouse case.
Wooden details
Work with wood begins with the selection of wood. Boards must be of the correct geometry, have a minimum of knots and defects, and have the necessary humidity.
At first, the boards are dried at home. At least six months.
After that, the board is sawn into small bars, which are dried for several weeks at the place of their further processing. At all stages, humidity is controlled by a special device. If we neglect the drying process, the tree loses geometric stability, and the manufacture and operation of the mouse becomes impossible.
The prepared bars are processed on the CNC machine using a specially created program.
From the very beginning of the creation of the part and until the final assembly of the mouse, the parts are rigidly fixed on a metal snap, so that at no stage does the part change its shape and geometric dimensions.
The processing of the upper part of the mouse has to be done with jewelry accuracy, since its profile is designed for a soft click and in some places is very thin. I control the pressure with a gramometer. In ordinary mice, it ranges from 50 to 75 GS. I try to achieve 50 GS.
The greatest difficulties are associated with the tree in my project. Not only is this the most significant part of the cost, but the proportion of marriage here is very high. Wood is an anisotropic material. It can lead, flaws can get caught, chips may appear, and just a mistake in the technology of topcoat can lead to the mouse body being sent to the trash. I admit that I still improve the processing technology, and I'm not completely sure that I found the right one. For statistics: in the first batch of ten buildings, only three reached the finished product. Therefore, part of the process chain associated with the tree is critical to the cost and quality of the finished product. Over it is ongoing work.
In the future I plan to work with bone. In particular, I am already creating a bone wheel.
Electronic part.
I developed the first mouse circuit myself. As a sensor, I took Avago's top-end optical sensor ADNS-3090, Atmel's controller became the brains, the remaining components of branded companies such as Murata, Yageo, Geyer, Omron and Molex.
I paid special attention to the quality nutrition of the mouse, here, in my opinion, it reached its absolute perfectionism :)
In order to understand how the sensor works from and to, I completely translated the optosensor specification into Russian.
Then the first version of the board was developed.
The first working layout.
The second version of the board, the second breadboard.
The third.
Fourth.
In black, final.
There have also been experiments with various buttons. I have always tried, among others, to pick up a quiet mouse. Well, since I do it myself, I decided to conduct an experiment and make such a mouse and try it out in work. To do this, I replaced the left and right “mikriki” with soft and quiet ones that were used for the central button (noticed that the central button always clicks quieter?). A special version of the board was created, where all three of the same "mikrika" were mounted.
The mouse was in operation for a couple of weeks, and I can share the conclusion: in my case, the absence of a click is not comfortable. The tree partially absorbs the click, and it seems that the wooden keys are tight, and you have to press harder. It is difficult to describe the sensation, but the closest analogy is as if under the finger there is a mouse button from soft to breadcrumbs, and for a confident click you need to squeeze it. The experiment can be considered successful, since a clear result was obtained which should be the response from a mouse click - clear, clear, unambiguously interpreted.
I picked up and bought a batch of gold-plated connectors for the mouse. As usual, in China. I don’t know about the “best contact”, but they harmonize perfectly with the tree.
At the moment, a project is underway to develop a wireless version of the mouse.
Screen, firmware
Carried away by the idea of placing a display in the mouse, I began searching for it among hundreds of suppliers. The requirements were simple: tight overall limitations and the possibility of at least a symbolic display of at least eight familiarities. While picking up, I learned almost everything about displays. They differ in types: symbolic and graphic, in technology: TAB, COG, TFT, OLED, LCD, E-Paper and others. Each type or technology has a lot of varieties, sizes, colors, lights, etc. In general, there was something to delve into.
Half-Internet break, I found out that the size I need is made by only one company in the whole world. All other options are definitely larger in size. And even the display I found barely fit inside the mouse. As an option, a custom display was considered, which I could make according to my requirements, but this is a very expensive option for me (about one hundred thousand rubles). For the first model, a graphic display with a resolution of 128 by 64 pixels is quite suitable, which I chose.
In order to understand how the display really looks and combines with my mouse, I had to order all varieties of this display from the manufacturers. What do these varieties mean? The model name consists of alphanumeric unpronounceable combinations of the type FP12P629AU12. All of them are composed of various blocks and are clearly decrypted in the specification. For example, the above example can be assembled from FP.12.P.629A.U12 blocks, where the type, size, voltage, controller, operating temperature range and other model information are encrypted. And the last block is the trickiest. It can have several tens of values, each of which means a particular combination of such characteristics as the presence and color of the backlight, background color, character color, range of degrees from which the information is clearly read. These parameters were just interesting to me.
As a result, "for testing" I ordered 18 different modifications. The manufacturer agreed, but said that the minimum order is 5 displays for each modification. There was nowhere to go, and I had to agree, knowing that 90% would go into the garbage can. And on one of the cloudy days, the express delivery service brought home a huge box in which you can live a bum of average build. There were 18 smaller boxes in the box, each of which freely placed 5 displays, securely fixed for a long trip to cold Russia. There were so many accompanying packaging that it was enough for my mother-in-law to cover several beds for the winter.
As a result, after careful tests at a specially assembled stand, two displays turned out to be suitable for the series. They differ only in background: gray and yellow-green. It’s them that I will offer to complete the mouse. By default, I plan to set yellow-green, but two more options will be available: a display with a gray background and a mouse with no display at all.
But the main intrigue was what information can be displayed on the screen? I was offered different ideas: the temperature of the air, an indication of the arrival of letters, something else not very original.
My train of thought went a different way. To begin with, there are two significant restrictions on the display of operational information: the presence in front of the user of a huge and high-quality source of any information (monitor) and the need to turn the mouse over to receive information. In addition, the screen is small, the resolution is small, the LED interferes with normal reading. Therefore, my conclusion was one: the information should be only entertaining in nature, the applied value of which tends to zero, but at the same time, the WOW! Effect should be killer.
What information can possess such properties in a mediocre device? Its a little: mileage, time of use, speed, number of clicks and scroll wheel. I decided to refuse the last parameter, since it seemed to me uninteresting. The rest of all parameters are associated with the session (the last time the mouse was used from the moment it was powered up, i.e. connecting to the computer or turning on the computer itself) and to the entire time the mouse exists. For example, a user can at any moment of the mouse find out how many times he pressed the left mouse button or how much his mouse has run in meters for today or since its purchase. The information is absolutely useless, but it will help the curious to understand how much he torments the mouse. If other interesting ideas appear, they can be implemented with new firmware.
Also added general information on the mouse (model, mouse and firmware number, month of manufacture) and the settings screen. It will be possible to choose the language and system of measures (English or metric). To store all this information, I had to add flash memory of permanent storage to the circuit.
To put such a volume of information, I had to split everything into screens. Each type of information is displayed on each screen, and the parameter values for the session and for the whole time are displayed. In total, six screens turned out that change with the mouse wheel.
The first option was implemented in a purely textual manner, for which several font options were even developed.
I made firmware to evaluate how the text looks using the created font on the mouse’s screen. It looks awful what to say.
Now it has become apparent that the screen needs graphics, not a set of symbolic information. Therefore, I hired a designer to work, and together we prepared three options for graphic execution, as a result, the second option was recognized as the most successful.
Of course, such a design required more resolution, so it had to be adapted.
But this is not the end of the story. After I picked up the screen for the mouse, I made an order for a trial batch for mock-ups. As a result, screens came, but for some reason, the number of conclusions is different from what is specified in the specification (datasheet). The manufacturer received a response to the request that, they say, everything is fine, this is a small modification, and it will not affect the performance in any way. Meanwhile, the missing two conductors were responsible for the brightness of the displayed graphics.
All this was very suspicious. And just as he looked into the water. We remade the board for a modified screen, soldered it, and then it turned out that the screen was completely dull. As if the device had run out of batteries. And it turned out after a long and painstaking work on the search and selection of screens, the purchase of a trial batch of all modifications and their tests. Time, money, and so on.
But the story ended with a good ending. After correspondence with the Chinese, it turned out that the screen can now adjust its contrast directly from the firmware. They treated the firmware, and everything began to show just fine!
Everything is shown as planned: mileage, speed, number of clicks and more.
In the future, the firmware also changed several times: there was a setting for changing the language. Two languages on one screen is bad - readability is deteriorating, an English-speaking user will only be annoyed by Cyrillic gibberish, and in the future support for other languages may be needed. Difficulties began when I tried to align the mileage of the mouse. It seems to be complicated: the optical sensor transmits the increment along two coordinates, which must be reduced to a system of measures and added modulo to the current value. That's the whole run.
But, as it turned out, not everything is so simple. Two people with mice where the same sensor is installed can get dramatically different results! The thing is that the resolution of the sensor (sensitivity) very much depends on the surface on which the mouse rolls. The best result is when the mouse rolls on white paper. Slightly worse on wood and fabric. On the laminate and the film is very bad. The declared passport sensitivity is achieved only on ideal, from the point of view of the sensor, surfaces.
For the end user, this does not matter. It connects the mouse and, by trial and error, exposes a comfortable cursor speed in the operating system. The system remembers this coefficient and uses it to increase or decrease the values of the increment of the coordinates of movement.
But it’s a completely different matter if you are planning to read these parameters directly from the mouse. A mouse on one surface will show the result of a mileage of one meter, on the other - one and a half. Speed will also “lie”. And something needs to be done with this.
To solve this problem, it was necessary to introduce the parameter "Sensitivity", which allows you to individually select the coefficient for each surface. By default, it is equal to one, which corresponds to the surface of white paper. It can be increased or decreased in the settings. You can not touch it at all, everything will work just fine. But for true perfectionists, a leaflet will be given in the leaflet attached to the mouse, from which you can choose the coefficient to the existing surface and instructions on how to independently configure the mouse to show the exact mileage.
When developing the firmware, another side effect of the sensor was discovered. If you take the mouse and just wave it in the air, then the mileage will also change. This is due to the fact that the sensor determines the surrounding space as a certain surface and also tries to obtain the values of the mouse offset. Therefore, you can observe such an effect: you turn the mouse over, look at the mileage parameters and are surprised that they are changing right in front of your eyes. Of course, you can install a tilt angle sensor in the mouse that disables the sensor for the time of its revolution, but it is unreasonable to do this only for the described situation. Perhaps in the next version he will appear, but not now. After all, the mouse is raised only to look at the indicators, and 99.9% of the time it is on the surface and receives the correct information.
Cable The
cable decided to make it as flexible as possible so that it does not interfere with the movement of the mouse and is “invisible” to kinematics. Well, I personally do not like the "spring" cable.
Sometimes it seems that when creating a product, the cable is the most non-essential part of the product. What is easier is to buy in the store the right amount of cable and solder it. Shitty thing. But, alas, not in Russia. Sometimes it seems that our industry is no longer able to do anything more complicated than cast iron irons. Attempts to find a cable resulted in a three-week search and shaking up the assortment of absolutely all manufacturers of Russian cable products. It turned out: our standards do not describe a cable suitable for modern electronic devices. For example, a four-wire microphone cable with a KMM 4x0.12mm2 braid has an outer diameter of 5 mm. This is a lot. Older mice and keyboards have an apparently thick cable, whose outer diameter is only 3.5 mm. The closest analogue on sale was the cable of the German company Lapp Kabel, but he also had an external diameter of just 3.5 mm. Now imagine also a braid on such a cable. Presented? I’ll tell you that I saw a similar cable on the power cords for irons :)
So, it turned out: you can’t buy such a cable in Russia. Point. Well, we are not accustomed to retreat. I’m going into production and trying to order, since in Russia they still make a cable. And for this we decide on my requirements. So, what I need:
Cores - copper, from braided wires (for flexibility).
Number of cores - 4.
Screen - yes.
Flexibility is maximum.
The outer diameter of the cable is strictly not more than 3 mm.
Color - Pantone 4625 C.
Bottom line: I tried to get in touch, probably with a dozen possible manufacturers of cable products, no one is interested in messing with my order. They didn’t even ask what mileage I needed. Bottom line: in Russia such a cable cannot be bought or produced. Sad But we are not used to retreat.
I'm going to Alibaba.com. I find the first Chinese manufacturer that comes across, I write a letter and literally a few hours later I get the answer: we will make any cable for you! I am in shock. I throw him a specification, money for delivery, and in a week I get a sample. Wow! And I lost almost three months, trying to patriotically place an order in Russia. It turned out that the Chinese can quite easily make me a cable with an external diameter of 2.5 mm.
Bottom line: I ordered 4 different samples in China. At first, the scratchability and dullness of the outer shell did not suit, then the cable did not suit the flexibility, then it did not suit the flexibility again, and finally settled on the last sample sent, which I was ready to order. They cannot be more flexible. The cable has a memory. As a result, I accidentally received a memory cable, although I wanted to be as flexible as a rope :)
I ordered a kilometer, after two weeks I had the cable. Total time spent: six months.
Braided his kilometer of cable. It turned out two options.
About 10% of the cable went into rejection. This is the beginning of the bays, where the braid is untwisted and the machine has not yet entered the operating mode. And some places where for some reason loops and knots of braid threads formed.
If the end of the cable is not terminated with heat shrink, then it will fluff at the moment, the threads are synthetic! Therefore, the installation of the cable assembly is hindered by the preventive shrink sleeve.
The outer diameter of the braided cable was 3.2 mm, i.e. the braid added 0.7 mm to the cable diameter. It seems a little, but with a normal mouse the cable comes mainly with a diameter of 3.5 mm, and in the era of wireless mice it seems thick and heavy. Recently, not budget mice have begun to be equipped with cables with a diameter of 3 mm, and they no longer interfere with work, they are almost imperceptible. But the keyboard cable can be with an external diameter of 4 mm. And even more. But for the keyboard this is not important.
Next ordered a cable assembly. Cables have two different lengths and two braid colors. So to speak, for every taste of the consumer.
Plastic parts
No matter how I want to make the mouse body parts made entirely of wood, plastic is indispensable here. We need legs, an axis for the wheel, support for the axis and a glass for the display.
Therefore, I had to order a mold from the Chinese.
After each test casting, the Chinese sent me a dozen samples, which I tested on my mouse.
As a result, I worked on the mold three times, until the quality began to satisfy me. The problems were different. For example, after assembly, I got a dust problem that forms between the display and the protective glass. It looks untidy. Moreover, the mouse will scrape over the surface, and dust will gradually accumulate there. I had to convert the glass into a container with sides, where the display will be inserted, after which the circuit will be sealed.
It turned out here is such a detail.
Refining the mold is a daunting task, and changes can only be made in the direction of increasing detail. Therefore, any inaccuracy or error can ruin the whole work. For reference: each revision is a month and a half of waiting for new samples. And the change itself could be microscopic, but necessary.
I will not dwell on plastic parts, this technology is now leading, and I can not tell you anything new and interesting here. I’ll only say about the legs, for which I have been selecting material with reduced friction for a long time, after which I have tested and run the mice in order to determine the winner with minimal friction. I described all this in detail in my blog.
Processing and coating
First, there is thorough work with the removal of pile, sanding and polishing the surface.
I faced a difficult task. It was necessary to stabilize the tree so that the geometry of the mouse did not change depending on humidity, and protect the tree from working in an aggressive environment (sweat and fat by hand).
From the very beginning I refused varnish. Varnish is a surface film that cracks over time, collapses, and the tree turns out to be bare. Sweat and fat penetrate the pores, the tree darkens, and the irreversible process of its degradation begins. Therefore, it was decided to use oil as an impregnation and protection, and wax to give a commercial look.
To make it clear: the tree is thoroughly saturated with pores, which contain either air or the oil of the tree itself (if the tree is rubber). Our task is to fill the pores with our oil as much as possible, which then should polymerize and protect the tree.
In order not to stretch the story, I’ll say that I have tried many oils: linseed, teak, tung, vaseline, Danish. Each oil has its own character. For example, wax is applied very poorly to teak oil, and linseed oil cures for a very long time. Therefore, it is necessary to introduce a catalyst - desiccant.
As a result, I developed two technologies. The first is the technology of vacuum impregnation of wood. It works like this: I create a vacuum in an environment with oil and wood. Air begins to come out of the pores. After removing the vacuum, the pores are filled with oil. As a plus, the tree stabilizes well. As a minus - it gets very dark. It looks good, but not for everybody.
The second technology is a surface oil coating. Oil is applied 1-2 or more times with a non-woven cloth.
Then we process with a felt circle.
Apply carnauba wax.
And rub the muslin circle.
Then, using a hair dryer, I “dissolve” the dry wax residue in narrow and difficult places. In the case of “insoluble” debris I pick up a toothbrush with a stiff bristle, remove the debris and then again repeat the procedure for applying wax locally.
All is ready!
If we evaluate the labor costs of processing, then manual labor per mouse turns out about four hours.
Assembly.
Next is the installation operation, but before it you still need to remove traces of processing from the technological holes. Then, using a special 3M tape, I align and stick the legs (the case can lead to fractions of a millimeter, and this will immediately be noticeable: it will stagger like a lame-footed stool). Then I lay the cable, mount the board, support, install the wheel and also, if necessary, adjust the buttons (there should be no bounce) and the force of pressing. This operation can also take up to four hours.
Logo, website, blog
Such a status wooden mouse cannot do without a cool logo. I will omit the intermediate options and show the result. He pleases me. The combination of leaf (wood), mouse (gadget) and wood color clearly indicates the scope of the company. And the inscription made using the notched font indicates the solidity of the enterprise.
The site decided to do in a low key. The tree looks best on black: www.clickwood.ru .
The main feature of the site that should be mentioned is the configurator. Any user can play around with the options and order a mouse for themselves. You can change the type of wood for each part, its impregnation, the length and design of the cable, the presence and color of the screen. Even if the Chinese copy my mouse, they are unlikely to offer such a service.
For those who are interested in the process itself, I cover every step in detail on my blog: kuliba.livejournal.com . There you can also find the project of a wooden keyboard and case, which I put aside.
Why did I do all this? At first it was a hobby. Then the hobby turned into a very expensive budget item, but there was no reason to stop, a lot was done, it was necessary to move forward. In the end, I made an industrial-replicated author's product. Spent 2.5 years and a large budget. It's no secret that I now want to create a small business. Perhaps my mouse will be bought as an elite gift to the head or just a unique gift. Indeed, until now there has not been a single status product on the computer market. Even if you are a Persian sheikh, you still have to hold a dull plastic craft in your hands. I exaggerate, of course, but the meaning is clear.
What plans do you have for the future? I plan to create all the elements of a workstation: mouse, keyboard, body and monitor. Perhaps along the way I will offer goods from the “desktop” category: pens, coasters, etc., also made using valuable varieties of wood and other exotic materials. Products are status, from the premium segment, but I strive to make them affordable for the middle class, this is the most buying category of people.
I would like to create a small workshop. To feed old age and bring joy to me and lovers of original gadgets. Not everything is completed to the end, not everyone I am satisfied. But I keep moving forward. Improving, perfecting and inventing.
UPD: I in no way pretend to be the palm and I am not the first to make a wooden status mouse, but in my post I want to talk about the history of my project, I didn’t set another task for myself. Thank you for understanding.