New Posts  All Forums:

Posts by TranquilTempest

What, they couldn't enough HBM2?
I'm going to ask you to actually read my post again, instead of choking on extremely bad ways to build one off prototypes: http://www.overclock.net/t/1606062/why-are-diy-mice-so-rare/10#post_25355043 Logitech isn't in this to build one mouse, they're in this to build millions of mice. Of course they're going to use CAD, they're going to have different people building different parts without looking over each other's shoulder, and they need tooling made for injection...
Which part of a mouse needs tolerances that tight? The sensor height isn't that sensitive, and pin plunger microswitches have a throw about 5x that.Also, if you knew anything about high precision, you'd know that humans can hold MUCH better tolerances than that. Many high end machine and measurement tools use hand scraped ways and surfaces, and in the context of making a plastic shell fit the switch placement, you could get that tolerance in a few minutes with some...
Yeah, I took a look at the micros(); function, and figured if I take the time understand the underlying timers well enough to mess with it, then I can just use the underlying timers instead of messing with it. I can probably get better precision by modifying prescalers, but it would also probably break a bunch of other stuff that relies on the timers.
A DIY mouse is by definition prototyping. It's also easier to manufacture to fit than to manufacture to tolerance, at least in very small production volumes(it's much easier to make part B match part A than to make both part A and B match independent reference C).Shim stock and sandpaper go a long way.
I posted this in another thread a while back, but you can use a microcontroller with native USB to measure the polling interval. I have an arduino micro(this code should work with any arduino that has native USB, like a lenoardo or due. With an arduino, you have to make a couple changes to the library code in order to get the information you want. I'm going to detail the changes I made instead of just posting the modified files because there have been updates to the...
I take it you didn't read my other post, you use your sculpted mouse as a mold for vacuforming, which is your functional shell. High volume mice need injection molding to meet their cost constraints, but nobody's going to spend tens of thousands of dollars on injection molds for one mouse, or even a run of 1000 mice. Vacuum forming is ideal for small runs, and is easy enough to DIY. You also get a better result than the vast majority of 3d printers.
The compromise in the test method I suggested isn't in precision or accuracy, it's in how long it takes to thoroughly test the mouse. In fact, I wouldn't be surprised if it produced better data than whatever logitech uses. Just looked at a datasheet for a somewhat cheap linear encoder, specified accuracy is 15 micron over 1 meter. As in, a 2000 DPI mouse would have to be accurate to 1 count over 1 meter to need better test equipment. That leaves angle. If you want to be...
going to need more information: length, width, number of conductors, pin pitch, termination style. A photo would help.
Making the mouse shape isn't harder than the electronics, it's just a totally different skill set.3D printing isn't going to be a good way to go about it. You're making a physical object that needs to feel good in your hand. Why wait hours/days for your 3d printed prototype when you can sculpt it out of clay and immediately feel what you need to fix?
New Posts  All Forums: