Thank you everyone for your massive support during the Kickstarter campaign… Remee was able to raise nearly $600K! If you found out about Remee or Bitbanger Labs after the Kickstarter ended, you can still pre-order through our website, www.sleepwithremee.com
Thanks again!
Thanks to a lot of support and due in no small part to the efforts of folks on sites like reddit, twitter, and facebook, Remee has far surpassed our original short term goals!
Thanks to everyone who has supported us so far. The Kickstarter is still live for 17 more days!
We’re happy to announce that Remee, the world’s first comfortable, affordable lucid dreaming mask is now seeking funding on Kickstarter! Click here to see more!
New PCBs just showed up for our second project. They’re looking sexy as usual. At the last minute I swapped out the TQFP atmega footprint for a DIP package. It means I’ll have to go back and make changes to the lard before any kind of final run, but I was getting sick of soldering atmegas onto test boards with no way to salvage them when I go to the next iteration. I guess I could always reheat the pcb and pluck them off, but that’s a project for another day.
This board will be the first time we’ve used a free spinning rotary encoder, should be interesting and make for a pretty flexible interface. We’ll also be running an 8×2 LCD using a shift register to save pins. Now I’ve just got to find time to solder the thing.
The beast is alive. Electronics showed and all sorts of other bits and bobs showed up last week and we’ve been squeezing in a couple late nights to get the mendelmax up and running. You can see from the photo that the coffee pot was an integral part of the process. We’ve got a heated pcb for the build platform but at the moment we’re going to get calibrated printing PLA so we don’t need a heat bed yet.
First calibration cubes off the printer show impressively tight clean layers for a first go and also illustrate that PLA needs a fan for cooling, otherwise you get the drooping and sagging that you can see halfway up the prints.
Starting to look professional now. Got the smaller PCB all put together (you can see little antennae debugging wires in the pic), and its now settled in a nice tight sandwich of three layers of laser cut foam. We’re really digging on the foam as a super low profile enclosure, going to have to look at it for some other projects, especially wearables. In the bottom pic you can see the two tiny little light sensors at the top corners of the board that less us bitbang settings into the mask using any screen.
Bitbanger labs has got a bun in the oven. We’ve decided to finally dive headfirst in the 3d printing reprap world with a shiny new mendelmax. The mendelmax is a relatively new variant of the reprap which relies on slot aluminum instead of rods. This things looks like a beast, and after a few hours the frame went together with only minor swearing. We’ve got plenty of experience with slot aluminum and pulleys from building camera rigs, so fingers crossed that once the rest of the bits arrive everything else goes together as easily as the frame.
We’re bitbanging, and it feels good. We always intended the sleep mask to have parameters that a user could customize, but didn’t really give too much thought as to how that was going to happen until a few weeks ago. Programing the chip with an ISP is feasible but not very user friendly, especially not for the non-arduino user. Then there is the default, use the two buttons on the mask to enter a debug mode and then count blinks and enter some ridiculous code of button presses to alter stuff. That also seemed like a terrible way to spend ones time.
Then the epiphany, we’ll put two light sensors on the mask and then you can blink the data in from any screen. It took a bit to get the correct light sensors (not all are tuned to ambient light), then we had to calibrate them to a computer screen or smart phone. A bit of code to get a website up and a bit more code to set one input as a Clock and one as a Signal and there you go. Now you can change your settings on any web enabled device with a screen and then simply blink them right over to the mask. You can see us debugging a breadboard version, where we used LEDs to signal if the arduino was receiving a bit when it should. White lights up the led, black and they stay dark. We also set up a handshake byte so the mask would know when to start analyzing data. Even though we made the thing, its still pretty amazing to see data get sent this way.
PCBs aplenty. It’s been a while since the last update mainly because we’ve been engrossed in learning surface mount soldering and making tweaks to our PCBs. You can see we’re already through a number of versions and we’re pretty excited with how things are moving along. The topmost smaller board in the image is our final-ish design. We realized by putting the coin cell on the back of the board we can effectively cut the footprint in half. Makes the mask lower profile and more comfortable. We also got some time on a friends laser cutter and were able to make some snazzy foam sandwiches that hold the electronics but stay soft and flexible. The thing is amazingly wearable, even for someone for me, a restless side/face sleeper.
The code is also coming along nicely. The mask works by waiting for a predetermined long delay (say 5 hours or so), at which point the sleeper will be at their highest ratio of REM to deep sleep, then it pulses distinct light patterns at a much shorter interval. This way you are likely to have the lights coincide with REM a good number of times per night without having to have all the bulkier and less reliable components that go with REM detection. We’ve even managed to tune the atmega sleep function so that the chip is only on for two cycles every 8 seconds. It barely draw enough power to register on the multimeter, great for battery life.
If you look close at the new board you’ll see we’ve christened project X with a real name too. Look even closer and you’ll see the T1 & T2 solder pads where our awesome mystery feature is going to end up.
From breadboard to frankenstein protoboard, we’re moving steadily along. This is our attempt to take the breadboard circuit into something moderately wearable and testable. Gotta say I hate protoboarding, seems like so much work for simple connections. Gonna take a look at Eagle CAD and designing my own PCBs after this. Still, in the end it works. This prototype is running off two coin cells through an LDO 3.3v regulator. We used the handy SMD lilpad LEDs to stay low profile. After trying various paddings for the PCB/mask we ended up with craft foam which can easily be cut to exactly fit the boards but molds nicely to the face and feels soft but firm.