This is the forum archive of Homey. For more information about Homey, visit the Official Homey website.
The Homey Community has been moved to https://community.athom.com.
This forum is now read-only for archive purposes.
The Homey Community has been moved to https://community.athom.com.
This forum is now read-only for archive purposes.
Comments
Hi all,
It’s Monday February 26th which means this is Kaj and Thijs with a new update on Slide. Our previous (super-long) update focused on our time line for DFM and production. Today, we’ve got a few update bits but mainly focus on the servo-less design – and have also made a nice video of that. Additionally, we’ll talk briefly about a few other steps we’ve made this month. Let’s go!
Eliminating the servo (and why it’s such a big deal)
In the video below, Kaj walks you through the engineering we’ve done to eliminate the servo. Essentially, servos are small DC motors combined with gears and a controller, and they’re used in cases where you need to move an object to a specific position. In the case of Slide, we used the servo to position the coupling mechanism. This coupling mechanism essentially determines which way your curtains are moving. It also makes sure that you can still move your curtains by hand when the motor is not turning.
However, we tested countless servos and none of them delivered the performance and precision we needed. At the time of our beta test, we were using a servo that was accurate, but it still tended to make a low ticking noise, even when Slide was not working. Not ideal – and unacceptable in the bedroom. So after the beta test, as we wrote in our earlier update, we definitively decided to take it out. Now we have what we call our servo-less design.
As you can see, what makes this such a good alternative setup is the fact that we re-use the motion of the DC motor. We also investigated other options like solenoids, but it always came down to adding more complexity and parts. The servo-less design only adds a few mechanical parts: a coupling (like the one from before, but simpler), a friction/slip clutch, and a secondary axis called a spindle. We also added a third magnetic sensor to the sensor board, that tracks the location of the coupling. We developed a software algorithm that automatically brings the coupling back into neutral at the end, so you can keep moving your curtains by hand.
TL;DR: In summary, by removing the servo, we’ve finally eliminated the last big internal bottleneck of Slide (that followed from the first beta test), so we’re very happy with this.
Software: connectivity update, and Apple HomeKit possibilities
For the beta test, Valentin has rolled out the OTA update to almost all backer testers and it seems to have tackled most connectivity issues. There’s still a few cases we’re looking into, typically where the default modem has some odd multicast settings that affect performance of Slide (as well as other IoT devices, from what we understand). Overall, pretty good progress so far.
In other exciting news, the Espressif hardware architecture we use (essentially, the brain of Slide) has now been certified by Apple as HomeKit compatible. Espressif has also released an accompanying SDK. This does not mean we can commit to HomeKit support straight out of the box, but it’s a lot more likely than before. Another piece of good news is that we’ve obtained an MFi Developer license from Apple that allows us to investigate how we could implement Apple HomeKit. To be clear, this is not a priority for us right now and we’re focusing on the core of Slide at this moment. But we’ll continue this exploration on the side, and it would potentially mean we could add HomeKit support through an OTA update a few months after the release.
Miscellaneous points
Planning progress In our last update we shared an internal time line, we’re currently on track there with no significant changes. We did integrate the servo-less design into our main engineering model a few revisions earlier than planned due to good performance and to simplify the pre-DFM, which will take up the bulk of our focus in March. We finished V3 of the second connector (for people with double curtains), which now consists of two parts you ‘snap’ into each other, eliminating the need for screws. We’re making good on our promise that you won’t need any tools or equipment to set up your Slide. :-)
New Slide design
We've got a lot of comments on the new Slide design, which people are generally pretty happy about, which makes us happy too! Some of you did express concern over the use of fabric, and the possible downsides of that. We hope to alleviate those concerns with the following pointers:
We’ll use the next update to talk a bit more about the design, and explain how it’s both an aesthetic improvement, but also serves a few key functional purposes.
Massive Mount Test (the MMT)
Already mentioned in our last update, we’re going to send – in phases – 100 Slide mounts into the field to test compatibility with your curtain systems. By doing so, we want to collect more data to make sure it works with virtually all curtain systems. We’re starting with curtain rails first and have two focus groups within that category. We’re sending 25 mounts to people that self-reported (in the Backerkit survey) that they have an exotic system. These are the ‘known unknowns’ – cases where we expect the current version of the Slide mount not to work perfectly (and where we want to learn). But we’re also sending 25 mounts to people that reported to have a common system – to see if it’s as common as they think. These are the ‘unknown unknowns’, where we don’t know what results to expect – so just as important to test. By next update, we expect to share some first results with you.
Onwards
So that’s all for today – we hope you enjoyed the update and continue to read along with us on our journey. Always appreciate your feedback, so let us know if you have any comments or questions. Our next update will be out on March 19.
Until then, wishing you a good week from an increasingly frozen Amsterdam,
Thijs & Kaj
Hello everyone!
Today marks Monday March 19th, and it’s Kaj and Thijs here with the latest news on Slide.
Our last two updates centered mostly on one of the more technical aspects of the work – how we eliminated the servo, the last internal part that was causing us trouble. In this update we’re leaving the internals for what they are and show more of the user experience of the final Slide design.
Mounting Slide on rails
Some of you may remember our first Slide rails install video based on the initial design of Slide. Much has changed since then and installing a Slide has become a lot easier. In the initial prototype, users had to fiddle a lot to properly fit the wire into the connector, with lots of risk of human error. During the follow-on R&D we discovered this system was not suitable for mass production. First, because it wasn’t durable enough for multiple years of use, and secondly, because it didn’t fit with all types of runners.
We’ve solved these issues in the final design of the mounting system. Today, we want to show you what that looks like for rails-based systems. We’ve recorded a quick video with our office demo setup to showcase just how simple it is. Check it out below:
www.youtube.com/watch?v=j0S3xWzeiF0As you can see, the various mounting parts have become much more robust to stand the test of time – the result of extensive testing. You still fit two mounts onto your existing rails, but they’re now placed directly underneath your rails. In this video we’re using the existing curtain stopper (and a runner) to fit the mounts. During the beta test we discovered the decades-old stoppers of some people could no longer be tightly screwed, which is why we’ll also be shipping an extra set of universal stoppers for narrow and wide-type rails with Slide – the choice is yours.
Both the Slide main unit and Slide pulley are easily connected to the two mounts. After that, all you need to do is magnetically snap the Slide Connector onto your runners and re-attach your curtains, and you’re ready to go. The Slide Connector is already pre-mounted with the wire, so you don’t need to do that manually. If you have a double curtain setup, simply repeat with a second Slide Connector (not pictured here).
We’re currently conducting our ‘Massive Mount Test’ to verify the mounts pictured here work with all rails. If based on this video, you’re not sure it’ll work for your rails (and we haven’t yet been in touch about testing), send us a message to see if we should include your case in our test. If you have a rod-based system: we’ll also add a video demonstrating the mounting system for rods further down the line.
Planning and timeline
In last month’s update from February 6th we posted a detailed timeline from then to mass production, with first scheduled shipments for end of June/early July. Right now, we are on track with this timeline. What that means is that in February we focused almost fully on bringing the servo-less design from a proof of concept to a functional prototype. And it means that this month, we’re on schedule in integrating those channels into our main internal build. Right now, we’re finalizing this build to be transferred to our production partner next month for DFM – as planned.
In closing
This current phase - optimization for DFM – is hard to capture in an update because it involves many incremental (micro-)adjustments to the builds, resolving small bugs and, as the name says it, optimization across the board. That’s why this update is a bit shorter – but hopefully still as valuable! We know it’s been a lengthy road – but we’re getting there.
One more thing. As Easter is around the corner, it seems appropriate to throw in a little Easter egg video. Since we started with Slide, people often ask if they can still control their curtains by hand. You certainly can: it’s a key feature of Slide (and one we didn’t communicate clear enough
). But for those looking for more of a luxury vibe without grabbing their phone or the remote control... Slide also has a nifty little function called ‘Touch & Go’. Check it out!
https://www.youtube.com/watch?v=oSAiWcaHinAOur next update is out in three weeks on Monday April 9th. Stay tuned!
Cheers from Amsterdam, Thijs & Kaj
Hello everyone,
Team Slide with our latest tri-weekly update on.. Slide! :-) In our last update we focused on demonstrating the (significantly improved) mounting system for rails as well as showcased the much-requested Slide Touch & Go feature.
Today, we’ll give you a quick recap of what we’ve been up to these past few weeks and what’s coming next for Slide. This is a very exciting week as we’re handing over the Slide project to our lead production partner who is going to take over the final pre-production (DFM) work and prepare the production and assembly lines. In other words – a major milestone.
“How does Slide work with a double curtain system?”
The video included in our last update demonstrated how we’ve simplified the mounting mechanism for Slide on rails-based systems, compared to our initial designs. However, in that video we used a single-curtain setup that left quite some of you confused: will a single Slide still work on double curtains? Rest assured: it absolutely does. Out of the box, the Slide Pulley and the 1st Connector are magnetically snapped to the Slide main unit. For a single curtain, that’s all you need. And in order to set it up with a double curtain, all you need to do is snap our 2nd Connector onto the Slide wire. The 2nd Connector is included in the box of every Slide we ship.
Check out this one-minute demonstration of what this looks like in practice:
The 2nd Connector looks different from the 1st Connector because we needed to create something you could easily snap onto the Slide Wire, without actually damaging that wire. We also needed something that was easy to attach, especially if you’re standing on a ladder or chair trying to reach the top of your curtains. The second connector consists of two matching parts with specially engineered elliptical curves on the inside that fully block the wire from moving. This means the wire is fixed in place after snapping the parts together, ensuring a robust system and solid user experience.
What we’ve been up to in the past three weeks
In our February 6th update we included a detailed timeline from now till the shipment of Slide from July onwards. In line with that planning schedule, what we’ve done in these past few weeks is bringing all the isolated design and engineering parts of Slide together in one model, the final internal build before hand-over.
After our beta test among backers this winter we identified a number of improvement points, including the use of internal bearings for smoother hand movement, the elimination of the servo, and the recessed mount. To speed up development all these improvements were engineered in parallel, but isolated from each other. As our final step in engineering, we’ve now brought all these elements together into one unified build. We’ve also tested that finalized design, to make sure it still meets our quality and durability requirements. It does, and all these subsystems pass our standard 5-years simulated-use test runs.
Handing over Slide to our production partner
This week is an important one because we’re handing over the Slide designs to our production partner, as planned. It’s a big milestone and also quite thrilling as we’re transferring the day-to-day work on Slide for this next stage. As you know we’ve gone through a lot of effort to find a suitable production partner with a proven track record in advanced mechanical designs. We’re working with a Netherlands-based manufacturer with their own on-site production facilities in Shenzen, China. This means we have the advantage of working with a local partner that is just a short ride away, but still retain all the benefits of production in China (in terms of cost, quality and speed).
We’ve been working together for the past few months, but previously they were advising us and we were leading the engineering. With this handover, those roles reverse. They have a lot of experience in complex products with moving parts, and we’re happy to have them on board. Their first step, called DFM or Design-for-Manufacturing, means they’re optimizing all parts for tooling. The second step is tooling itself: creating the molds that will be used for producing Slide’s plastic parts. After that, they will set up the full production line for Slide, incorporating assembly of all plastic parts, electronics and quality testing. In just a few months from now this summer, we’ll be able to ramp up production and ship Slide to you in volume. We’ll start shipping Slides to backers from July onwards.
We’re aware it’s been quite a while – but we’re moving towards the finish line and can’t wait to get Slides in all of your hands. :-) It’s been a hard but very worthwhile journey and your continued support and encouragement has been pivotal in that!
--
That’s all for now. As always – reach out to us with any questions you may have here or at thijs@goslide.io. Our next update is out on Monday, April 30th.
Cheers from Amsterdam,
Thijs & Kaj
Anxiously awaiting delivery of Slide :-D
I miss my Slide everyday .... I
And another update from the Slide team to KS backers:
Hey everyone!
Kaj and Thijs here with the latest news on Slide. In this update, we want to give you a quick briefing on what happened in the past three weeks. Most importantly, we handed Slide over to our production partner, a great milestone. As part of our ongoing video demo’s, this time we’re including a short video demonstrating how an a-symmetrical double curtain system works. And to close, we’ll discuss our next plans for Slide as well as a note on our updates. Let’s go!
A-symmetrical double curtain setup
Previously we shared videos of Slide mounted on single- as well as double-curtain systems. Today, we’re including a short video for a-symmetrical double curtains. This is relevant for those of who you have two curtains that are not equal in width. In these cases, you can use two Slides to automate both curtains independently. The video explains how this works.
What happened in the past three weeks
A little over two weeks ago we reached one of the key milestones in this journey: handing the Slide project over to our production partner. To keep this update concise, you can read more about them and what they do in our previous update.
They have now taken over the lead as we gear up towards mass production. They’ve started DFM (finalizing the model for production). For Slide, we work with different suppliers, including for the motor, the electronics (PCBs), the springs and more. One of the reasons we selected a turn-key partner is to make sure we have a smooth, integrated supply chain towards the assembly of Slide.
Concretely, their DFM work involves many small changes to our models to make sure they’re producible: ready for tooling. Tooling is the process of creating steel molds used for producing plastic parts through plastic mold injection. These changes are necessary to make sure the plastic parts coming out of production are durable and all fit together. This tooling process is due to start next month (May), so more about that in the next update.
In terms of electronics & software, we’ve starting working on automated in-factory testing & flashing of the electronics boards (PCBs). The goal is to make sure all Slide PCBs are electrically tested before leaving the factory, and that firmware is flashed securely. Related to this, we’re finalizing the basic OTA (over the air) infrastructure for updating Slide’s firmware after you installed the product at home. We’ll touch on the wider topic of security and our cloud system in one of the next updates, as well.
Looking forward
In the next three weeks, we’ll continue to work with our production partner as they prepare for tooling. We’re on schedule to start mass-producing and shipping Slides this summer from July onwards. With Slide finalized, we’ve already started planning ahead. First and foremost, our focus is on getting Slides into all of your hands, ASAP.
But we also want to share a few thoughts on the next step of our journey, after fulfilling all your orders: launching Slide in retail. We’ll first do that in the Netherlands in Q3 2018, but expand internationally as well.
Slide remains a one-of-its-kind product: a simple, single-box system to automate existing curtains. Compared to other products, the patented tech of Slide makes it the only universal system that doesn’t have to be built-to-order. Because of that, we’ve gotten massive interest from many smart home players and retail channels that can’t wait to start selling Slide. These include DIY stores, consumer electronics stores and online platforms that currently can’t offer a universal smart curtain solution to their customers.
From this position of high demand, we’re now lining up a mix of retail channels that we think fit well with the Slide audience. We’re doing this together with selected distributors and retailers. The redesigned Slide will retail in the Netherlands for €199. This finalized version of Slide is in many ways very different, and much better, than the prototype concept of Slide that many of you backed in an early stage.
We’re giving all our backers an upgrade to that retail version, even as some of you backed at a much lower price point. For us, this is a pivotal part of the journey and obvious thing to do. We also hope and believe it will, at least in part, make up for the long wait. :-)
Update interval from 3 to 4 weeks
For those who frequently read our updates, you will realize there’s a bit less content now than before. That all has to do with the fact that we’ve handed a very large part of Slide over to our production partner. In the months ahead, we’re in a production phase where the steps remaining are fewer, and less iterative, like tooling, assembly and shipments.
Because of this, we’re changing our update frequency from once every three weeks to once every four weeks, from now until the end of fulfillment. Once we start shipping, we’ll also reach out to backers individually with their tracking information as batches get shipped.
We know our updates are long, and we even pride ourselves on it. We spend a lot of time writing them, because we think it’s very important to keep you in the loop about the good and the bad. And because we want to include you in the journey of creating consumer hardware and everything it entails. But we want to keep them relevant, too. Doing updates once every four weeks makes sure each update is informative, without over-doing it.
--
That’s all for now! Our next update is coming out on May 28.
All the best from Amsterdam,
Thijs & Kaj
I totally love the "pull the curtains to close"-feature which is new.
These guys really know what they are talking about. They also demonstrated how strong the wire that pulls the curtains is, no worries about this product at all. I cannot wait until they can start sending out the Slide!
Can anyone who had the beta unit or knows a bit more advise me on something
It says it works on curtains up to 6m, I actually have two small Windows with double curtains well under this - could this be rigged with a single device to control both of these at the same time?
They are pole and ring in curtain design
Thanks!
Thanks
Very interesting to see that, but again not quote what I'm wondering (tho I think i might be able to infer from the video my answer)
I have essentially two sets of those curtains side by side (so 4 pieces of fabric) - all identical. Wondering if (with additional clips) the curtains can be rigged from one Slide unit. I'm guessing not since given the size of that clip it looks like it does more than just grip the wire?