It’s easy for us industrial designers to have love/hate relationships with engineers. We want them to clean up our tooling drawings, sure, but we hate when they make changes without telling us, and we often dismiss them as not “getting” design, or UX, or aesthetics. If engineers had the final say, we tell ourselves, every housing would have sharp corners, every mechanism would be overly complicated, every operation would require three times as many steps.
But those are all broad, sweeping generalizations. There are plenty of design-minded engineers out there, and we should know–we recently met one out in the wild. Here’s our Q&A with David Kleeman, Belkin’s Director of Mechanical Engineering, on what the interplay of design and engineering was on their TrueClear Pro project (which we got a great look at here).
Listen beautiful relax classics on our Youtube channel.
Core77: Here’s a leading question: In your eyes, what does an industrial designer do versus what you engineers do?
David Kleeman: An industrial designer wastes my ti–just kidding. We’re a very design-centric company, which I really appreciate because I consider myself closer to that side than engineering in some respects. And we really do work very closely together, and very well together, I think.
[The design group’s] role is helping us aspire to that greatness that we all want to achieve. We’re all going for that kind of Apple aesthetic, seeking perfection, and they really help us define that. They’re really good with identifying the usability issues of our mechanical designs.
I really want to stress that even on the engineering side here, we are very design-focused. There’s a lot of surfacing stuff that we do here that I haven’t really seen anyone else do outside of Apple. Things like curvature continuity, C2 surfacing, that kind of thing. The engineers here are very highly skilled at that–which makes it very hard for us to hire people, because that’s not a very common thing for people out there.
In our interview with Oliver [Seil, VP of Design] on the TrueClear Pro, he mentioned that Belkin’s process usually starts with ID/UX and then “joins hands” with ME. But in this case it was reversed, and the design started with Engineering and not ID. Can you talk about that?
A lot of our products are things we’ve either made before and are updating earlier versions; or we’ve made things similar to them; or at least have some idea of how to go about making it. For instance if you’re making a wireless charger, you know what the electronics are going to be, that kind of thing. But with this particular use case–of an easy-to-use screen protector applicator–no one had any idea of how to apply this thing very precisely onto another thing. It seemed like a very significant engineering challenge, so it fell to us to come up with a mechanism.
What was your reaction to hearing that it was gonna start with you guys?
[The Design group] is really good at keeping us aware and bringing us in early to provide some quote-unquote “reality” to their concepts, but for us to actually be leading was a new thing for us. It was a big challenge, so that was stressful, but at the same time, kind of a relief–there were no size constraints, no constraints on surfacing, shapes, costs. It was totally an open book, which was pretty exciting.
Were you guys looking at the existing screen protector applicators on the market?
Yeah, that was a big part of our initial concept development, ideation sessions and brainstorming. We purchased and had all those products. We were very cognizant of patents, investigating what details of those things were patented in case we wanted to go down a similar road. But they all were really not very good. There were a lot of usability issues, the alignments were terrible. Studying these was more like “These are the things to not do.”
We also did really high-in-the-sky concepting, crazy ideas like “It goes into this black box, and all this stuff happens on the inside, and it’s under a vacuum, and there’s a laser….” then we slowly dialed it down to more realistic directions.
Of the four workable designs presented, at least one was still pretty crazy–you had one prototype that applied the screen protector to the phone using an air bladder from a blood pressure cuff that pressed it into place. I thought that was pretty cool. Can you talk about the four approaches you took?
So on a high level, the whole reason we went with four in the first place was because we didn’t know if any of them would work. We had a fair amount of confidence in all of them based on engineering intuition and that we’ve been doing this for a while, but we really didn’t know if they would work.
We quickly resolved around a way that the overlay itself would be constructed; we had as a team coalesced around this concept, and all four prototypes utilized the same overlay basis [for applying the protector], which really helped bring a cohesive nature to all of those designs.
Once we had that, it was really a matter of different ways to pull out that release liner bit, that extra plastic bit that applies the screen protector, and different ways to hold the phone in place, how to get the alignment perfect.
One of our concepts was a sort of pulling machine, that was derived from the thought that it would remove any sort of variability in how you pulled on this thing. So that a user who isn’t familiar with the machine could just kind of rip on it from the side, not really knowing what it does. So that machine had a track on a rail with gearing, completely eliminating any kind of variability. There also, to be honest, was a bit of showmanship in that–it looked cool to operate, and whether that’s necessary or not is up for debate, but we liked the whole action of it.
We had a smaller machine, the clamshell MX, that eventually ended up in the Apple Store. That was really about “How do we have a bespoke machine for a single phone that’s as small as we possibly can get it.” Get rid of all this gearing–can we figure out another way to pull this liner out, manually, while retaining the alignment?
Your favorite, the blood pressure cuff one, was definitely the craziest one. That one was about, “We don’t wanna do any of this pulling stuff, we don’t want to have any kind of sliding mechanism, we just wanna go with trying to mechanically recreate the ‘taco’ method that people do naturally while trying to apply a screen protector, but in a much more accurate way.” Holding everything in a fixed position also worked really well.
The fourth concept was a sliding credit card type of a thing. That one worked really well, but there were some potential patent issues with a sliding roller deal, so we didn’t really pursue that one further than our little prototypes.
Got it. And the air bladder one, it was inflated by a pump?
It had batteries in it, powering a little motor in there that would cause it to inflate.
That’s pretty cool. And once you had these four working concepts, how did you narrow it down?
Well, we had those four, we got them all the way to a testable, prototype-able condition. Then there were a few key areas we needed to look at, alignment and getting rid of air bubbles under the screen protector and all that stuff. So we did a whole testing series on all of those. They all passed–which was a surprise. To be honest we were kind of depending on some of them not passing, to make it easier to eliminate some.
But they all passed. So then it really came down to, “What would be more appropriate for a store environment?” The air bladder, for instance, was great but it took a long time. It was cool to watch, but it was a lot of hitting this button, and watching it inflate, and waiting.
That credit card thing I alluded to earlier, it involved a lot of weaving this overlay within the machine–we didn’t think it was as intuitive as some of the other ones, so we eliminated that.
Listen beautiful relax classics on our Youtube channel.
So that left two concepts. One was the larger machine with the moveable cradles and the big pulling-action deal. That seemed to be really appropriate for a large telco. And then we had this smaller one that seemed appropriate for a phone supplier that only had a few phones. That’s how we ended up with those two.
So once Engineering had done their job of producing workable mechanisms, what were the roles of Design and Engineering, from that point? Does Engineering just hand it off to Design, or is there more co-work?
So from that point, we did a beta-alpha test out in the field. ID started to get more involved and it was about making this more palatable to a consumer, aesthetically anyway. We got into the tool and manufacturing aspects of it. ID was responsible for the look and the UX, and the consumer experience, which is a huge part of this, of course.
[Pulling out my phone] I have a Belkin screen protector on my phone here, that came out of the machine in the Apple Store. What’s the tolerance of this one?
[Squinting] This one, if I look at it…should be…Yeah, it looks like it’s within 0.1 millimeters. Zero-point-one millimeters, nine out of ten times, is what we’re going for. And all of our testing has born that out.
You can measure 0.1mm with your eyes?!?
We engineers all have calibrated eyes, and if I put my fingers together like this–I’m joking. We have an optical machine that measures it. But we do have these [eyeballing] nerd contests between ourselves.
What made you pick that number?
Couple things. One, it seemed achievable. Zero-point-one millimeters is the thickness of a sheet of paper. We considered machining tolerances and figured we could get there nine times out of ten. We felt 0.1mm as acceptable for the size of the cutout around the home button and around the cameras and the speaker and all that. Prior to this machine, the tolerances were much larger because of the variability in human application. And we figured with 0.1, the consumer would be like “Okay, this fits very tightly around the home button, very tightly around the camera hole,” that kind of thing. So it was partially engineering intuition and partially aesthetics.
I’m assuming engineers have the same regrets designers do after a project, like “Oh god, why couldn’t we have focused more on such-and-such….” Any of that with this? Is there anything you would want to see added or modified?
When we did the alpha and beta testing and asked for negative feedback, we didn’t get any, which was really surprising. This is a fairly low-volume product–it’s going into stores, not directly to consumers–so a lot of it is machined to high tolerances, so there’s not a lot of tooling errors.
But from our perspective, one thing that we did struggle with, and that I would like to see improved, is the sound of that little latch.
Yeah, it’s a little bit too clacky. It’d be great if the sound of it was just a bit softer. And there is some stuff that we have fixed over time, like the opening of the flip lid–at times it’s been too hard or too slow.
Why is how fast the lid opens or closes important?
It’s purely for the consumer impression of the experience. There were early samples that opened too fast and made the whole thing jump. That didn’t make any difference to the functionality of it, but it did to the experience of it. We wanted this to be magical.
You’re talking like a designer.
Yeah, I’ve definitely drank the Kool-Aid.
Last question: With this project, the decision was made to start with Engineering, and it was obviously the right call. Would you prefer to see that approach taken more with other projects?
How much time do we have? This goes back to the whole ME vs. ID fight thing. They’ll design something impossible, then we’ll try to make it reality, then they’ll be like “That’s too big” or too whatever, and back and forth, back and forth.
Surely life would be easier if we just did what I told them to! But I definitely see the benefit of the ID-driven approach.