Love to see another version of a DIY SpaceMouse, I'm glad Teaching Tech has given the topic a lot of momentum. I just have to see what version will be out when I'll have the time to build one 😆
This looks like a really nice version. I've been using the Spacemouse for a few years and dread ever having to get a new one due to the cost. Projects like this one tell me that when it's time, I won't have to break the bank. Thank you for your hard work and for sharing it with us.
I've been using the my Space Mouse Wireless for 7 years it is still working perfectly. Considering that the current replacement price for that model is $175, my Space Mouse to date has cost me less than a bottle of Coke per month to use. If you can't afford that then you probably can't afford the material to build this project. And if you consider your time to be worth anything, then this unit is way more costly than the 3d Connexion version.
Unless the tech in the Spacemouse has changed, which it's been a year or two since I went patent diving and last time I did I had issues finding the current patents (mostly due to never finding specifically who holds the current patents, plus patent language and naming is asinine at best), a large part of the cost is the sensors, insanely expensive little bastards due to being low volume components. Then of course you have mark ups and the Logitech specialty product tax on top of everything else. Last time I went looking for those sensors themselves, which they're also an utter pain to find from any vendor assuming you can even find a manufacturer and attain part numbers, the cost of a wired Spacemouse was justifiable by comparison, again they're low volume components with highly specific uses in a narrow range of products, they're expensive due to basic economics of scale and demand, something big companies see as pocket change but amounts to hyperinflated BOM cost for individuals. That's why all of these projects deviate in sensor type, making things affordable component-wise. Another part of the cost of the product is the manufacturing processes, there's a reason why disassembling one is a point of no return and why you absolutely will damage the product opening it up, other than being unfriendly to repairs the processes to manufacture this are just more expensive in comparison to making a simpler assembly. I think the situation is outrageous and that these components should otherwise be far cheaper, but realistically everyone is getting screwed on parts costs.
I'm dying to have the time to build this, I have all the parts. I'm using the space mouse mini from teatching tech, but I feel it's too stiff to use, this can surely be tweaked to be more comfortable. Thanks for this amazing open source project!
I have built the Kempy version with hall effect joysticks and steel balls. It feels flimsy and unprecise. Will give your TPU version a try, hopefully it will be more solid.
@DoTheDiy Yes, but the maths is more complicated. Using 8 means the operation is easier to understand. Just pluses and minuses involved. I have looked at the original Space mushroom by Shiura, which uses 3 joysticks, and the translation matrix he uses looks simple. I'm sure it translates to the trigonometric equations involved, but 2 more sensors make things simpler.
This is nice project. I own 2 spacemice but would love to get another one for the workstation next to my milling machine. This project looks nice so I am going to build (at least) one for my other workstation
I've clearly changed. Once I would have loved to build this. Now I find myself thinking that the £150 for the 3Dconnexion doesn't seem as bad now I've seen all the work that goes into it 🤣
@HughLyon-Sack I got mine many years ago as part of a bulk pack, but I can't remember where from. The nearest cheap match I have found are www.aliexpress.com/item/1005002694387664.html I have not tried these, so I don't know how good they are. You can also try the TPU springs discussed on the GitHub pages if you have any TPU filament to spare.
@pitacotavo5938 Thanks for your comment. I was wondering the opposite. Are the springs too weak? It is easy to move but quite sensitive to the movement. I have 0.4mm wire springs, and they are definitely too strong. I have 0.3mm ones on order, but between the 0.4mm and the 0.2mm ones the later is the best. It is up to each builder to make their own choice as to what works for them from what they can get hold of. I have had the 0.2mm ones for many many years so it was an easy choice for me. But now I am buying more springs to see if I can make it better even though I think it is pretty much perfect as it is.
@@johncrombie2771it looks too "wobbly" to my eyes. I have a space mouse, and the hat doesn't seem to start to wobble move if I shake the base. I set it sideways down on a scale, and it takes about 250 grams of force to rotate the hat fully to the sides (pitch/roll). Up and down (z) was a bit more fiddly to measure since it's not connected to my PC at the moment, but the whole thing weighs 455 grams, and it wouldn't surprise me if it takes 300-400 grams to fully lift or press the hat. I can't think of an easy way to measure the twisting force (yaw) with the tools I have available. I just have a scale 😅. It would be simple if I had one of those spring-in-a-tube force meters we used in physics class. I guess I could wind a string around it and see how much weight it takes to twist it, but it's going to be fiddly. I can do it if you want. I'm also not sure how I would measure x/y without also getting pitch/roll. Tldr; the force for roll and pitch should be about 250g.
@phizc with the springs I've used on mine, it certainly needs a lot less to move it than 250g. I like the feel and the ease of movement. It is certainly possible to put in stronger springs, but you may need to add a weighted base. I haven't done that on mine even though I did design a base plate to hold something heavy. As you say, the original spacemouse is heavy, so being more difficult to move isn't a problem. I have tried different springs, as I mentioned on github, but I keep going back to the original ones. Because I have never used a 3Dconnexion mouse, I don't know what it should feel like. Then again, I'm trying to replicate the functionality, not the feel.
@joshuadonavan1966 I did a quick Google search, and that says the nano won't work. The nano has a USB controller chip, which prevents it from being used as an HID (Human Interface Device) . I had the same problem using the XIAO ESP32C3 with my cyberpuck mouse design, but it has bluetooth, so I was able to modify the design and heavily modify the code to make it a wireless device. I would suggest getting the Pro Micro. We know it works.
Hi i have a problem, when i try to uppload the coad on to the ardiunu i get an error saying "Compilation error: a function-definition is not allowed here before '{' token" do you now how to fix this?
@pavelo1686 I have no problem compiling or uploading the code. You should get a line number where the error occurred. Send me that together with the code you have just before and after that. About 10 lines either side and I'll see if I can spot anything that could cause your error. Possibly a ; in the wrong place?
@@johncrombie2771 thanks for your time but i already fixed that isiue by deliting the orginal code tha is writen in eavry new sketch in arduino ide and the pasting your code.
@yunghiga7699 According to the 3Dconnexion website, their driver works on Windows, Mac OS and LINUX, so this should also work on any of those. Check out their website for details on compatible OS version numbers.
@youtubehandlesux You need a certain amount of deadzone to cope with jitter. That annoying instability in the numbers from the Analog to Digital converters. The 3D ConneXions driver does allow you to reduce the sensitivity of the Spacemouse. This also let's you reduce the deadzone as it ignores the jitters. If you do decide to build my version, please share photos and give feedback.
@@johncrombie2771 It's more of the combination of non-linear actuation force of the joysticks and big deadzone (baked into the stick itself rather than being a software filter), and each of the joysticks having a slightly different non-linear response. I abandoned 3d printing for a few months due to personal reasons, but im now coming back to this. After comparing a lot of projects online i'm considering building either a miniature Stewart platform with 6 hall effect sensors, or open up 3 hall effect joysticks, remove the centering mechanism and lubricate them a lot.
Love to see another version of a DIY SpaceMouse, I'm glad Teaching Tech has given the topic a lot of momentum. I just have to see what version will be out when I'll have the time to build one 😆
The problem with the spread of diy is that today we use the connexion software, but they can block it if they see that it is too popular...
This looks like a really nice version.
I've been using the Spacemouse for a few years and dread ever having to get a new one due to the cost.
Projects like this one tell me that when it's time, I won't have to break the bank.
Thank you for your hard work and for sharing it with us.
I've been using the my Space Mouse Wireless for 7 years it is still working perfectly. Considering that the current replacement price for that model is $175, my Space Mouse to date has cost me less than a bottle of Coke per month to use. If you can't afford that then you probably can't afford the material to build this project. And if you consider your time to be worth anything, then this unit is way more costly than the 3d Connexion version.
Unless the tech in the Spacemouse has changed, which it's been a year or two since I went patent diving and last time I did I had issues finding the current patents (mostly due to never finding specifically who holds the current patents, plus patent language and naming is asinine at best), a large part of the cost is the sensors, insanely expensive little bastards due to being low volume components. Then of course you have mark ups and the Logitech specialty product tax on top of everything else. Last time I went looking for those sensors themselves, which they're also an utter pain to find from any vendor assuming you can even find a manufacturer and attain part numbers, the cost of a wired Spacemouse was justifiable by comparison, again they're low volume components with highly specific uses in a narrow range of products, they're expensive due to basic economics of scale and demand, something big companies see as pocket change but amounts to hyperinflated BOM cost for individuals. That's why all of these projects deviate in sensor type, making things affordable component-wise. Another part of the cost of the product is the manufacturing processes, there's a reason why disassembling one is a point of no return and why you absolutely will damage the product opening it up, other than being unfriendly to repairs the processes to manufacture this are just more expensive in comparison to making a simpler assembly. I think the situation is outrageous and that these components should otherwise be far cheaper, but realistically everyone is getting screwed on parts costs.
There are many second hand units available.
Es de los diseños que por ahora más me gustan.Intentaré construirlo.Gracias por su trabajo.
@@Erosgenuino De nada. Me gusta que te gusta mi modelo.
I'm dying to have the time to build this, I have all the parts.
I'm using the space mouse mini from teatching tech, but I feel it's too stiff to use, this can surely be tweaked to be more comfortable.
Thanks for this amazing open source project!
thank you very much, the best design I found so far 👌👌
amazing works, thanks for sharing it !
I have built the Kempy version with hall effect joysticks and steel balls. It feels flimsy and unprecise. Will give your TPU version a try, hopefully it will be more solid.
Neat design, with 6 sensors u can achieve same result.. looking forward for more videos,👍
@DoTheDiy Yes, but the maths is more complicated. Using 8 means the operation is easier to understand. Just pluses and minuses involved. I have looked at the original Space mushroom by Shiura, which uses 3 joysticks, and the translation matrix he uses looks simple. I'm sure it translates to the trigonometric equations involved, but 2 more sensors make things simpler.
@@johncrombie2771 I am also using halleffect sensors but they are 3d sensors using 3x tli493
@@johncrombie2771 aaaand there's a great software to use it with 4 sensors.
Superb, superb, superb!
This is nice project. I own 2 spacemice but would love to get another one for the workstation next to my milling machine. This project looks nice so I am going to build (at least) one for my other workstation
I've clearly changed. Once I would have loved to build this. Now I find myself thinking that the £150 for the 3Dconnexion doesn't seem as bad now I've seen all the work that goes into it 🤣
Where can I get the springs you used in this build?
@HughLyon-Sack I got mine many years ago as part of a bulk pack, but I can't remember where from. The nearest cheap match I have found are www.aliexpress.com/item/1005002694387664.html
I have not tried these, so I don't know how good they are. You can also try the TPU springs discussed on the GitHub pages if you have any TPU filament to spare.
great!!!!👍👍👍
Hi I have a space mouse
I want to custom the original 2 button
Would t be possible to change the button to be left and right click?
Amazing dude, but by you showing it at the end, i appears to be too stiff, maybe weaker springs to make it more easier to use?
@pitacotavo5938 Thanks for your comment. I was wondering the opposite. Are the springs too weak? It is easy to move but quite sensitive to the movement. I have 0.4mm wire springs, and they are definitely too strong. I have 0.3mm ones on order, but between the 0.4mm and the 0.2mm ones the later is the best.
It is up to each builder to make their own choice as to what works for them from what they can get hold of. I have had the 0.2mm ones for many many years so it was an easy choice for me. But now I am buying more springs to see if I can make it better even though I think it is pretty much perfect as it is.
@@johncrombie2771it looks too "wobbly" to my eyes. I have a space mouse, and the hat doesn't seem to start to wobble move if I shake the base.
I set it sideways down on a scale, and it takes about 250 grams of force to rotate the hat fully to the sides (pitch/roll). Up and down (z) was a bit more fiddly to measure since it's not connected to my PC at the moment, but the whole thing weighs 455 grams, and it wouldn't surprise me if it takes 300-400 grams to fully lift or press the hat. I can't think of an easy way to measure the twisting force (yaw) with the tools I have available. I just have a scale 😅. It would be simple if I had one of those spring-in-a-tube force meters we used in physics class. I guess I could wind a string around it and see how much weight it takes to twist it, but it's going to be fiddly. I can do it if you want. I'm also not sure how I would measure x/y without also getting pitch/roll.
Tldr; the force for roll and pitch should be about 250g.
@phizc with the springs I've used on mine, it certainly needs a lot less to move it than 250g. I like the feel and the ease of movement. It is certainly possible to put in stronger springs, but you may need to add a weighted base. I haven't done that on mine even though I did design a base plate to hold something heavy. As you say, the original spacemouse is heavy, so being more difficult to move isn't a problem. I have tried different springs, as I mentioned on github, but I keep going back to the original ones. Because I have never used a 3Dconnexion mouse, I don't know what it should feel like. Then again, I'm trying to replicate the functionality, not the feel.
@@phizc you can tie a rope or rubber to the stick, and use a weight on the scale and tie it to the weight.
ended up getting a Nano instead of Pro micro by accident. is it worth/possible getting the pinning figured out or should i just get a micro?
@joshuadonavan1966 I did a quick Google search, and that says the nano won't work. The nano has a USB controller chip, which prevents it from being used as an HID (Human Interface Device) . I had the same problem using the XIAO ESP32C3 with my cyberpuck mouse design, but it has bluetooth, so I was able to modify the design and heavily modify the code to make it a wireless device. I would suggest getting the Pro Micro. We know it works.
What are these like for controlling RCS in KSP?
Hi i have a problem, when i try to uppload the coad on to the ardiunu i get an error saying "Compilation error: a function-definition is not allowed here before '{' token" do you now how to fix this?
@pavelo1686 I have no problem compiling or uploading the code. You should get a line number where the error occurred. Send me that together with the code you have just before and after that. About 10 lines either side and I'll see if I can spot anything that could cause your error. Possibly a ; in the wrong place?
@@johncrombie2771 thanks for your time but i already fixed that isiue by deliting the orginal code tha is writen in eavry new sketch in arduino ide and the pasting your code.
Is there a BOM ie; parts list with quantities?
@@rick6563 Check out the Printables link in the description.
Just wonder if only run in windows and OS
@yunghiga7699 According to the 3Dconnexion website, their driver works on Windows, Mac OS and LINUX, so this should also work on any of those. Check out their website for details on compatible OS version numbers.
I think I'm going to scrap my joystick one and build this instead. Don't like the deadzone on the old one
@youtubehandlesux You need a certain amount of deadzone to cope with jitter. That annoying instability in the numbers from the Analog to Digital converters. The 3D ConneXions driver does allow you to reduce the sensitivity of the Spacemouse. This also let's you reduce the deadzone as it ignores the jitters.
If you do decide to build my version, please share photos and give feedback.
@@johncrombie2771 It's more of the combination of non-linear actuation force of the joysticks and big deadzone (baked into the stick itself rather than being a software filter), and each of the joysticks having a slightly different non-linear response. I abandoned 3d printing for a few months due to personal reasons, but im now coming back to this. After comparing a lot of projects online i'm considering building either a miniature Stewart platform with 6 hall effect sensors, or open up 3 hall effect joysticks, remove the centering mechanism and lubricate them a lot.