Juicero: A Lesson On When To Engineer Less

Ben Einstein, a product designer and founder at Bolt, a hardware-based VC, recently got his hands on a Juicero press. This desktop juice press that only works with proprietary pouches filled with chopped fruits and vegetables is currently bandied in the tech press as evidence Silicon Valley has gone mad, there is no future in building hardware, and the Internet of Things is a pox on civilization. Hey, at least they got the last one right.

This iFixit-style tear down digs into the Juicero mixer in all its gory details. It’s beautiful, it’s a marvel of technology, and given the engineering that went into this machine, it was doomed to fail. Not because it didn’t accomplish the task at hand, but because it does so with a level of engineering overkill that’s delightful to look at but devastating to the production cost.

Maybe It Wasn’t Just the Engineering…

Type ‘Juicero’ into Google News and you’ll see  Silicon Valley was shaken to its core this month. The future of hardware startups is now in question. Juicero, the world’s first cold-pressed juicing system, is in a tailspin. This company that raised $100 Million in funding is now being compared to Theranos, Uber, Enron, and every other corporate pariah in recent memory. The creator of Juicero, once favorably compared to Steve Jobs is now doing damage control with every outlet available.

Maybe, just maybe, the idea of a machine that would squeeze pre-packaged crushed fruits and vegetables into juice isn’t a great idea. The machine originally cost $700 USD, and the juice packs were purchased through a subscription model for $35 per week. For $2500, you could have all the fresh squeezed juice you want for an entire year. Maybe consumers don’t like the idea of DRM in their food. Juicero was a fantastic success for its creator, as the $100 Million in funding goes to show. For the rest of us, we’re left with a fantastic lesson on when you should engineer less, not more.

Just Look at It!

Disregarding the madness surrounding a $700 juice machine that uses proprietary, DRM’ed juice packs, the idea that anyone would spend two thousand dollars on juice per year, and the concept that a machine that makes juice is a bellwether for the entire economy, one marvelous thing did come out of this little kerfluffle. We get to look at design choices the likes of which are never seen in a single product.

Ben’s teardown covers the huge plastic enclosure, and the eight mold revisions required to bring this product to market. If you’re not familiar with injection molding, it’s expensive, and you really, really want to minimize how many mold revisions go into your first product. The best designers can finalize a mold in two or three revisions. It took Juicero eight, probably at a cost of over a million dollars.

The teardown continues to the power supply, a massive, custom job that converts wall power to 330V DC for the motor and 3.3, 5, and 12V for the brain of the device. The ‘transmission’ for this device consists of at least five machined, hardened gears. Our ‘ol pal [AvE] will tell you machined gears are rare in the finest prosumer grade DEWALT or Milwaukee tools — they usually go with cheaper sintered metal gears. By all accounts, the Juicero is an exercise in excess.

The Juicero is a machine that is overbuilt, overengineered, and the cost to make this machine prices it out of the market of people who would actually want a desktop juicing machine. While Juicero could iterate on this design for a few more revisions, replace the expensive machined parts for stamped steel, and simplify everything, even that would be insufficient. The design of the Juicero has a fatal flaw: it compresses the entire juice pack at once. That’s sixty-four square inches of fruit and vegetable pulp that must be compressed all at once. In an article that will surely be featured in next year’s Pulitzer nominations, Bloomberg discovered you can simply squeeze these juice bags. The Juicero isn’t just over-engineered. You could squeeze this bag through rollers with a simpler mechanism to get the same amount of juice out.

There’s a saying that any idiot can build a bridge that can carry a load, but only an engineer can build a bridge that barely carries a load. This is a desirable trait; a bridge that can carry any load would also carry an astronomical cost. Designing to a price point that you’ve verified by market research is engineering, and Juicero is the most spectacular failure of this type of engineering in recent memory. Of course, it doesn’t help that Juicero has DRM’ed food, but the success of K-cups tells me that’s probably not a priority for a lot of consumers.

174 thoughts on “Juicero: A Lesson On When To Engineer Less

      1. Even if you juice the entire fruit, it’s much easier to overconsume. You can put the juice of 4 apples in glass, and drink it in a few seconds, but most people would have trouble eating 4 whole apples in one sitting.

        1. So whole fruit is better because people have no concept of what they’re eating?

          Just a thought… but it might solve more problems if, instead of restricting yourself to only eating foods which make it so you don’t have to think, you instead actually learned to think about how many calories are in something, and then actually, I dunno, controlled your diet properly.

          1. “You don’t get it.

            Eat four apples your a full.”

            No, this is the part I don’t think *you* get – why do you care about feeling full? You have a brain. Use it to figure out whether or not you need food.

        2. Yeah, drinking 4 apples worth of juice in one sitting is definitely going to put you on the diabeetus train. But one apple’s worth of juice, along with a bunch of other juiced veg makes for a tastier, balanced drink. Spoonful of sugar gets the medicine in your face or something….

        3. And … and the other big problem is that, because no one will “chew” an apple juice they will be missing all enzymes and goodies included in the saliva we ingest.
          Needless to say, no one can swallow 4 apples without chewing … and chewing a lot!!

          That is the big problem with all “juices” out there. The juice drinker need the discipline to chew well before swallowing. Ain’t easy.

        1. Actually that one would be able to have lots of useful features if you could get product recognition right.

          A) Use of sensor suites to identify the types of materials going in would help to aid/improve recyclability by identifying the types of materials going in and aiding with autosorting and material identification. (Sends info on materials used to recycling center)

          b) Possible self cleaning function

          //Cynic mode engaging
          c) Identifying your wasteful habits to make recommendations on better products to buy (based on a self identified garbage producer profile and seeing if you match your garbage profile ((waste too much, better off buying in bulk))) It will stroke your ego when you are being more “environmental”

          //Cynic mode at full power
          d) Spy on you, based on your dirty dirty habits

      1. Sinds this juicer came to light, I’ve been trying to pitch my “IoT cloud connected phone charger concept”. It’s revolutionary! With a simple app, you can select how long you want to charge your phone for, and just leave it always connected. No longer you charge without it needing to charge.

        (So far, it’s the most ridiculous IoT application I could think of. But I’m sure someone, somewhere, will top this one)

        1. > But I’m sure someone, somewhere, will top this one.
          Hmm… what about IoT HDMI cables? You can use their free app to enable or disable them, and to turn on the connected TV (only works with Samsung TVs because it exploits a security flaw). They also have a nice feature called HDCP that can’t be disabled, but you’ll wonder why you should disable it. Finally, all of them automatically overlay ads on the video stream, and require $99/day (through the free app) to disable them.

          Who wants to buy a smartHDMI™ cable and get rid of the nonexisting problem of unplugging HDMI cables? The starter pack is just $999 and includes a cable, one-page instructions, some tape to hold it in place (required because the WiFly™ app connection could make it fly), and a code to get a 1% discount on another smartHDMI product.

        2. To be true to the Spirit of IoT you need to uses something like deep-learning image detection in “the cloud” to look at the phone and read the charge status, this send commands to switch the charging voltage on or off via a clunky relay of the gold-plated magic copper kind that the HiFi freaks use.

          It is important that the relay consumes more power than charging the phone does.

        3. I’ve always thought the most useless IoT device would be something you need to be present for most or all of the time the process takes.
          Toaster is the lead device in my mind just because you need to load and unload it, and the cycle only takes a minute and a half. What are you even going to do with all that time you save?

      2. I laughed when I saw it and referenced it in probably two school papers. The toaster. The juicero is now my wallpaper on my Windows box, because it looks so clean, and yet I know exactly what the joke means.

    1. I do some remote M2M work which is really IoT but where the functionality is pre-filtered by the fact that someone has to pay for it. So it’s not crap like a lot of IoT stuff.

      1. The IOT really was more respectable back when it was known as remote sensor networks or distributed robotics. Sadly the field grew up into your pot smoking college drop out uncle who gets drunk and looses his pants on Thursdays.

      1. +1 this post.
        Things like cars and refrigerators are often in use for at least ten years or fifteen years or more before they are taken out of service and you’re not going to see software updates for that long.

  1. I thought the point of a juicer was to use fresh vegetables, and you needed power to squeeze the good part out.

    Here you’re dealing with a less than fresh supply, and it’s processed to some extent. So why not just buy juice?

    Michael

      1. Exactly. An entire generation is being trained to make decisions based on whether their “friends/followers” on instagram or Facebook will be impressed. It is inconceivable for a Millenial to have an amazing day, and tell no-one.

  2. This product is astoundingly stoopid.
    Sure, you can make it a plane press – but how about rolling those packs for a change?
    Way less force, I guess you could make it way cheaper, lighter and simpler.
    But then you still end up with a useless product.

  3. Oh and also those things will be a goldmine for parts once the company is done with and people dump these.
    Might even be simple to mod into a letterpress printing machine.

    1. You can tell the engineering firm are not consumer product focused.
      The volume of Aluminum used is the dead giveaway…

      If they re-did the plastic line 8 times, then it was about 7 times more then these clowns should have been paid.

      From a product exposure standpoint it was a success.

      Maybe Iĺl assign a 3D printing project for the students to make a better version for $50

      1. It looks like they hired aerospace engineers to design a race car and then pivoted to making a juicer.

        Now, I’d really like to meet the VC people who thought that investing $100M into this was a good idea. How on earth did they expect to make that money back on an expensive juicer that makes expensive juice.

        1. Via the OTC securities market: They were probably long a good deal more CDS on the company debt than the original investment, these CDS would be easier to get a sucker on the other side of with bran-names like Alphabet on the investment side. A two horses kinda bet, like when one shorts Tesla ahead of earnings one will buy an equivalent amount of calls above the short, just in case the sucker takes off in the wrong direction.

  4. Not defending the ol’ Juicero by any means, but the teardown article assumes the mold was made 8 times because “Rev H” was stamped on the inside. This is a massive jump to conclusions!

    Rev H could mean anything, typically you get a preliminary quote using designation “Rev A”, then work with the molder making design changes until you are ready to order a mold. During this time, each design you send to the molder is tracked with revision numbers or letter to ensure they are working with the latest design files. It could be that they got the tool right on the first try, but they used the Rev H design files.

    1. It is also possible, that the mold is just modified for the revision. Welding some metal on to reduce the amount of plastic somewhere or grind something off to increase a wall thickness. I don’t know how difficult it is to add features like sliders later on, but at least it is possible that only one half of the mold has to be redone. So I am quite sure that not 8 complete molds have been built.

      1. Exactly! I enjoyed the legit criticisms of exorbitant cost to make this product, but you can’t cheat and start piling on false accusations.

        They almost certainly didn’t make the mold 8 times, most of the time you only need to make one mold and if there is a big mistake or design change you will have to retool once. 3d printers exist so we don’t have to make injection molds just to test fit and function.

      2. Or just changing gate locations, adding more venting. Lot of potential minor changes that still lock up up a rev level. At the drawing level, just changing notes or adding/subtracting inspection dimensions will uprev a document

      1. Read the specs; they list two models. One is a 24V motor with only 5.32A stall current. The other is an 18V motor with 130A stall current — that’s 2.3kW right there.

        Also, those specs are for the “JQ42-57” series, where 42 would probably indicate the rotor diameter, and 57 the rotor length. The motor in the Juicero is labeled “JQ42-1210N15A” (no spaces), so the rotor is probably 121mm — over twice as long. (The teardown article calls the part number “JQ42-1210N1” and seems to treat the “5A” as a stall current, but I’m not sure that’s right.)

        So I don’t find it at all implausible that this larger motor could be wound for 330VDC and 5A (or more) stall current.

        And as for dedicated circuits, just because you’re using a motor that pulls 5A (or more) at stall, doesn’t mean you have to have a power supply capable of feeding it 5A — you can use a polyfuse or circuit breaker, of course, or simply monitor the RPM (the Juicero has an encoder on the motor), and kill power if it falls below a speed corresponding to the maximum current you are prepared to draw.

        1. You’re right that the voltage is plausible. Read my comment below. I laughed at first because 5A at 330V sounds ridiculous for something that dispenses juice. In fact, it is entirely possible that because of the wacky design they needed this much power.

      2. Why should 1,6kW require a dedicated circuit? A Normal circuit (230V/16A) delivers 3,6kW and you need just a bridge rectifier and a cap to get 330VDC from it. Even from 110/120V you only need a voltage doubler or PFC circuit like in a PC PSU.

        1. The typical circuit in the US is 120V / 15A so something that continuously draws 1.6 kW would probably need a dedicated circuit. We put our refrigerators and microwaves on dedicated circuits these days.. I was going off the article which said it draws 5A at 330V, that sounded ridiculous at first glance for a cordless drill motor, but who knows maybe it’s custom made that way to reduce size of the wiring.

    2. Actually I don’t know what I’m talking about with motors either, maybe 330V is right – it is printed on the motor. It could have been custom made for them which would explain why it doesn’t match the datasheet. A 330V motor would have lower current draw so you can use more manageable wires, etc. Who’s to say it isn’t right.

    3. The motor in the picture is a JQ42/1210N115A

      The link is to the closest match on the website: JQ42/57 series motor. Just guessing, the JQ42 is probably just frame size, and the rest of the part number most likely refers to the electrics. The data sheet doesn’t show any data for the JQ42/12 series motors, Just the JQ42/57.

    4. You’re exactly right. The designers will start with ‘A’, then test this in modeling and 3D printing, so some structural analysis and then tweak. Given the state of modeling tools these days, it’s quite possible that the molded parts worked fine the first time.

    5. The motor is actually 330Volts. The link goes to a similar low voltage version of the motor as there is no link to that exact motors spec as it’s custom built.

      I see lots of high power, phase synchronous motors of this voltage in appliances such as washing machines. That’s because our mains voltage here is 240Volts AC and when that is bridge rectified you get 338Volts DC. Then the motor phase controls just have to be a cheap TRIAC per phase. I’d imagine the design is from Europe where they have 230Volts AC which rectifies to 324 Volts.

      But it’s a stupid proposition to use this motor voltage for 120Volt regions because that takes the power supply design from being a simple bridge rectifier to being a far more complex and expensive switch-mode boost converter.

        1. 50/60 Hz voltage doublers are only cheap for low power applications. Once you need more current you have to quite dramatically increase the capacitance value and, high value, high voltage capacitors are not cheap.

      1. You would need for PFC anyway in most regions if you have smoothing caps. You don’t need perfect filtering just for a motor. I know of kitchen appliances (Handmixer, handheld blender) which use permanent magnet DC motors (probably of similar size, spec’d for 300W – for 2min) which just use a bridge rectifier for the motor. The small motors in hairdryers too, they just use part of the heating coil as voltage divider or dropper resistor and 4 diodes.

        So yes, if you want to go cheap you have to select a motor which suits your mains voltage.

      2. Why even use a DC motor as a shunt motor like in a drill could do the job with no rectifier?
        As for control two relays it doesn’t need to be super accurate heck it would be possible to make the thing work well enough without any kind of controller at a all.

        1. I don’t know what you mean by ‘shunt motor’ but if you mean ‘shaded pole’ then they aren’t suitable because they are not very efficient. A shaded pole is fine for low power applications like a 50Watt pedestal fan where the 15 watts of heat can be blown away but at 1500Watts you would need to get rid of 450Watts of heat.

          The power supply has two opto-couples. They aren’t for the low voltage side as that’s a basic transformer. So they must be for motor control, suggesting that the voltage to the motor fluctuates under control from the micro. One opto is probably on-off and the other PWM.

        2. ah, well in that case [Nitori]’s has a very good idea. Just feed mains voltage to a universal motor and use a TRIAC to phase regulate the speed. You just need a Faraday cage to shield the RFI but it looks like the engineers would have no problem engineering a Faraday cage!

  5. “but only an engineer can build a bridge that barely carries a load”

    Not a very good engineer if they build a bridge with no safety factor though I get what you were getting at.

    How long until the DRM is broken? Keureg 2.0 took a few months and a piece of tape.

        1. Okay, good point. But it checks an online database to the QR code.
          If feel like some people here already know how to spoof a server on their home network.

          Piece of junk IoT.

    1. My master’s thesis was on safety for low-load structures. It will come as a surprise to many, but the risk factors for low load designs is, under certain conditions, essentially incalculable. Imagine a perfectly balanced thing. Since it is balanced, there is no resistance required to keep it balanced. However, almost any force can knock it out of balance. What resistance must be designed in this case? That leads to an unknown risk.

      So, yeah, only an engineer can design such a thing _correctly_ (assuming they actually know about these things).

  6. I think different: the creators have the right product for the WRONG CUSTOMER. A $700 juice machine is perfectly OK, but not for home use. Just think about all the software that could be written for a piece of engineering like this, to make it process a bunch of different drinks.

    1. Indeed, the pre-measured bags would make it ideal for some kind of franchise operation, like some kind of juice bar on a mobile cart like you’d see on the sidewalk or in the corridor of a mall. For a private person, you could easily buy a quality juicer off of amazon for less than $100 and fruit for far less than $35 a week (Heck, you could set up some dash buttons to buy certain ratios of fruit for that).

      Really, it’d make a lot more sense to sell it as a sort of kit with a food services cart, a mini-fridge, PoS terminal, and large supply of packets for a few thousand dollars so someone could charge pedestrians commuting to work in San Francisco or New York. The whole thing would be provided free, or at a steep discount, with the initial loss made up by way of franchise agreements and licensed consumables (Hence the DRM).

      Its almost like no one at this company bothered with even a basic business class or even asked someone with a business education.

        1. Uhhh, sorry reported you without noticing. Stupid touchscreen.
          Anyway I fully agree, one of these in every hotel room and a “fresh” pack of juice as a welcome gift. Because VIP paying expensive rooms deserve the superior refreshing feel of expensive juice! (not sarcasm, it might actually work wonders in reviews/opinions of the hotel etc).

      1. After looking at the teardown, I agree that the machine looks like it was designed for some sort of commercial operation. The folded sheet metal frame and machined aluminum parts look like something you’d see in a low-volume machine tool operating on an assembly line. Except that those would have been designed by somebody who saw no point in expensive injection molding tooling, and would most likely be covered with sheet metal. Likewise, the pre-packaged fruit makes zero sense for a home customer (they’d want to get the fruit fresh from a farmer’s market and throw together custom blends) but would make perfect sense if you were having some high school kid stamp out a couple dozen in an hour.

      2. “Its almost like no one at this company bothered with even a basic business class or even asked someone with a business education.”

        Well it’s a startup, so business and economy have not the same meaning that brick and mortar economy.
        In this sense, 100M$ VC raise is a very good indicator of their skills in this domain.

        1. A quote attributed to a successful entrepreneur (Richard Branson) is “Complexity is your enemy, Any fool can make things complicated, it is hard to make things simple”.

        2. It’s almost like nobody at the company had ever drunk juice before. It’s amazing how hype can separate millions of dollars from idiots, with no connection to reality. I dunno if they did any market research along the lines of “would you pay $700 for a juicer and you have to buy the fruits from us, sterilised and not fresh?”. Cos I don’t imagine they’d get a single “yes”.

          Certainly much of the value of the stock market is in hype, belief, faith. The belief that you can sell your share for more than you paid for it. Regardless of what the share IS, what the company does, that doesn’t matter. Just “can I sell it for more”. And that optimism, or feeling, is what the whole world’s economy floats on. Lack of faith leads to horrendous crashes, we’ve known that since the Tulip Bubble.

          So, erm, “People do stupid insane thing en masse, world doesn’t make any sense!” is something we should be used to, really. Certainly if this were a financial site, nobody would give a shit about the gears, or even the fact the thing’s a complete white elephant to start with. Doesn’t matter, for investment.

      3. The problem with that is: how many of those could you convince VCs you could sell? A couple of thousand? To secure $120,000,000.00 in VC capital, you have to convince them that every household in America (and eventually the world!) will want one.

        Remember, Silicon Valley entrepreneurs aren’t selling products to end-consumers. They’re selling visions to venture capital firms with far too much money.

  7. There are a lot of press juicers that are starting from minced raw vegetables that you could buy in a brick-and-mortar shop for 150 € – and you could also use it to make sorbets.
    http://www.panasonic.com/uk/consumer/home-appliances/juicers/mj-l500.html
    There are also other cheaper or more costly models, of course. In this case the business model is clear: sell a kitchen appliance applying a mark up. The owner will buy the fruit at the mall.
    The problem is that one has to clean the machine after, but if one has a dishwasher isn’t a big problem, there are also some auto-cleaning machines

  8. They need the very expensive machined gears because of the drive/press system they chose which seems to me to be the stupidest option they could have chosen.

    What they lacked is clearly an experienced system/design architect. Probably because they were stuck in the startup “we need young innovative people” mindset that is so patently bullshit. Here’s a hint, being ignorant of the state of the art (And thus the way things have been done that DON’T work) is NOT a good way to design a product.

    1. Having looked at the actual teardown now I will add:
      Holy hell they did NOT bother looking at easy off the shelf solutions. That thing is constructed more complicated than many of the systems I work with in my day job (And that’s state of the art semiconductor photo-lithography equipment where cost is usually the last consideration). That door locking mechanism alone (O.O)

      Very obviously they never bothered doing anything with off the shelf components even when they would quite possibly have been cheaper off.

      Once Juicero falls over I hope some of the engineers involved will give us more insight into the process that led to this abomination of a machine.

      1. Throw lots of money at something and your team feels the need to spend it on pimping out everything and it starts to become a company wide mindset. Remember the dot com era? Seems fairly clear what lead to this outcome.

        However, I am not as quick to criticize the overall outcome. Even if it is vastly overkill and the cost to produce it will limit the market penetration, the hardware engineering and attention to detail does seem at first glance to be fairly glorious.

        Also, shoving off the shelf parts into things isn’t always as easy or cost saving as you might think and what if something gets discontinued or modified which causes the whole supply chain to break because they didn’t consider your process when making that change.

        1. It’s not glorious – it’s stupid and the concept is even, Clearly thought up by some cool kids who hired a equally young and stupid engineer who didn’t know anything to concoct a monstrosity.

          If this is the sort of garbage along with IOT and Social Media is all Silicon Valley can come with now, they ought to close up shop.

          1. Please don’t misunderstand me. The idea of what this does and the market placement is utterly incompetent. I just have a lot of respect for extremely well engineered hardware. Not a big market for things like specialized space vehicles (at least not historically) and even though they were generally built by the lowest bidder, they were still extremely elegant machines. That’s what I think can be glorious.

      2. I have seen $10000+ precision instruments made far more cheaply than this. A flat press is a really bad way to do things unless you have to, and there has to be a better way to use mechanical advantage

  9. I think we aren’t looking broadly enough at it…
    The mechanism and electronics were probably designed for a new CIA weapon, and they are just re-utilizing some of its components to pocket more cash.

  10. What surprises me is that no one is really talking about what, IMHO,the products greatest failing is. Basically it’s just a press right, it doesn’t add any value other than a little muscle. Their biggest mistake was having the juice packs be “complete”, i.e. they should have made the packs concentrate and have the machine mix the concentrate with water. At least this way it’s measuring, squeezing, mixing, dispensing. Even though lots of articles have compared it to Keurig, another good comparison is sodastream. Both machines do more than dispense and actually have far more value add and this is where juicero really stumbled.

    1. How do you “concentrate” fruits and vegetables? You could dry them, but they wouldn’t be much use for juicing after.

      The idea of selling encapsulated, presumably sterilised, not fresh, vegetables by mail, is a stupid one. There are many others though.

      1. How do you think they make the concentrated juice they have at the store?
        Granted, it would still be an overly expensive machine for something you could easily make with your hands, but at least there would be some thin veneer of utility.

        1. We’re talking about concentrating actual fruit ‘n’ veg. Not just their juice, the fruits / veg themselves. This device squeezes fruit / veg to make juice, so that’s what it requires as input. You can’t concentrate solid fruit ‘n’ veg in any way that leaves them fit for juicing, far as I know. Since “concentrating” basically means removing water.

  11. People talk all about this product’s failings, but it’s interesting to look at other similar products that are being sold. Consider, for instance, the “Samson WELLPRESS Manual Juice Press”, which is essentially a car jack in a frame that squeezes two pans together. It goes for $400+. Or look at the “Tribest Green Star Elite”, which is a masticating juicer that goes for $600+. Or the “Kuvings Whole Slow Juicer” which goes for $400+.

    You can look at Juicero and say they’re crazy. But apparently there’s a whole crazy market already out there, and they’re just the latest entry.

    1. Usually they are a little more useful. I have an expensive Omega Juicer (I want to say it was $300-350 but I’m sure we had a coupon and maybe got away with it for 200). You can make frozen yogurt too. It’s heavily over-engineered and I think it has a 10 year (or maybe lifetime?) warranty. But there’s no DRM and it uses industry-standard off-the-shelf-fruit.

      Not saying you can’t get away with a cheaper juicer, but it was kind of a splurge purchase for the wife, and it does get good use.

  12. The whole bloody concept is seriously flawed, not just the engineering. People who want juice, want it fresh because vitamins and stuff. Well, at least it made some greedy people a little less rich.

  13. I would argue that the juice pack able to be squeezed by hand was a design concession. The machine probably couldn’t generate sufficient force over the 64^2in to adequately squeeze unprocessed fruits and vegetables. That’s 125psi, based on previous claims that the machine generated 4 tons of force.
    I suspect they found out early on that the machine needed some help, so they simply cube all the produce into small, easily squeezable bits.

    I think that a small air compressor that pumps up a bladder would have been a smarter way to apply force. That or they could also sell a CO2 canister subscription and the whole thing would be powered by CO2 gas pressure. There are a bunch of ways to skin this cat, perhaps their biggest achievement was creating a fresh juice delivery infrastructure, not a machine that presses the packs to extract the juice.

  14. Don’t forget that brand name juicers go for 300-500$ as it is. They use their warranties and support costs as a reasoning. It’s all BS. And this thing is WAYYYY over engineered. I was thinking one large plastic roller moving over the bags surface like squeezing toothpaste would do the trick at a MUCH lower BOM cost.

    1. Agreed. I feel like a symbolic human juicer would be hilarious. Just shove bits of a human in there and it squeezes out money lol. Another part of me is infuriated by this and the IoT endeavor peddling unnecessary garbage to folks when there are so many people that have so little.

  15. Ben is a genious. he just reivented this http://image.shutterstock.com/display_pic_with_logo/5056/5056,1127867000,1/stock-photo–th-century-washing-machine-583964.jpg and was able to collect 100 M$ with a destinated-to-fail-since-the beginning product. “Idiots and their money are doomed to be separated soon”, and Ben just prove us that Law once more.
    out there there are billions of idiots and means to be an idiont don’t take advantage from them.

  16. My guess is that an earlier version of the business model had actual cut-up fruit in the pouches and needed the extreme/even pressure to turn it into juice, so they designed a few-hundred-dollar machine to do it.

    Then either some fruits proved too resilient or too expensive, and they went with juice in the pouches, leaving the original ultra-strong mechanism in place because they already built the darn things.

  17. I heard about this on Marketplace on NPR that the bag can be manually massaged and give the same results, sans device. Fred Waring had it right, cause you could make some great cocktails in one.

  18. no no no. It is not a failure of engineering. it is a failure of design and product development, who clearly told the engineering firm who probably got bonuses for this, hey guys, spend as much money as possible, we need to justify this things expensive premium experience. They got what they asked for. in eight mold revisions or less.

    1. It frustrates me how the teardown refers to the machine as “beautiful engineering” – engineering is about balancing many requirements and constraints, but this machine is a horrific overkill monster. It’s not that hard to make a nice geared press mechanism. Just because it has expensive machined parts doesn’t make it beautiful engineering.

      But on the other hand, you have a good point – maybe they actually did do a good job working within their constraints, becuase actually they had a high cost target. You want a $700 dollar machine? You get a $700 machine!

  19. While I agree the Juicero is way overkill, I don’t think we see enough products that are well-engineered.
    We seem to live in a throw-away soceity as when something breaks and is out of warranty we just get rid of it and add to landfill. I get that corporations like to squeeze every cent out of BOM to increase margins, but there should be a better balance between quality vs cost when designing a product. The Juicero is a bad example of this, but its a step in the right direction. The custom PSU, locking door, and custom molding probably could have been cut back a bit. The core-mechanical parts would be more justifiable as they have the most wear and tear.
    A $299 juice press would be a bit more appealing than a $399 juice press if knowing the core press is a robust device. And of course as product popularity grows, 3rd party manufactuers would catch on and start producing refillable or compatible juice pack to avoid $35pw subscriptions.

    1. That’s because of a false economy.

      When you buy that new laptop or PC you *don’t* have to pay for the life of the South African teenager who is going to die from toxic chemical exposure when he tries to extract back the rare and costly materials used in it’s manufacture.

      You also *don’t* have to pay for his very young sister who will suffer a life long medical condition due to secondary exposure and be a costly burden on her village rather than a contributing member.

      If you had to pay for these things then the product would be much much more expensive and you would then be more inclined to purchase something that will last longer.

      These changes *will* happen because what we are doing is just not sustainable. We are consuming earths resources at 1.6 Earth Planet resources per year.

      This adjustment is going to happen at a time when we loose up to 40% of employment due to jobs being replaced with technology. People are discussing how we need a better wealth distribution model to cope with ever increasing employment but that is not the biggest issue.

      The biggest issue is that humans need work for reasons other than wealth. Work is a rock in out psychology that provides us an identity and a sense of self-worth. Removing this psychology rock will result in aggravation and escalating violence. Human nature has some very ugly side that have been hidden by the status quo we have had for a long time.

    2. The last well engineered, robust, wear and tear resistant, long-life product of good to excellent price point that could be found in near every home is the cast-iron skillet. And it remains such a value to this day that many have one or more in their kitchen at this time. Any glitz or gadget is not really well engineered, it is well marketed.

      1. Multipurpose too, legend has it that Great Grandma was making breakfast at the logging camp where she was cook, and a bear stuck it’s head in the cookhouse door, whereupon she smacked it upside the head with the hot cast iron frying pan, eggs and all, it took off and never came back.

      2. “We must not forget that the wheel is reinvented so often because it is a very good idea; I’ve learned to worry more about the soundness of ideas that were invented only once.”
        — David L. Parnas (Why Software Jewels are Rare, IEEE Computer, 2/96).

  20. This thing is not over engineered, it’s foolishly designed. The relevant quote is from Neville Shute, “an engineer is a man who can do for ten shillings what any fool can do for a pound”

    In my opinion it’s evidence of how far removed the rich can be from normal life; if you’re a multi-millionaire you might believe it’s reasonable to spend $2500 a year on fresh juice, because that’s what you already do.

    1. Except you don’t really know, do you? This thing goes for $700, right?

      Maybe another team building a high end, DRM’d, subscription-based, juicer aimed at the luxury market would be selling it for twice that.

      1. all I’m saying is its a cost-inefficient design. The asking price is a marketing decision, just baiting the hook to get you buying the DRM bags. I recently bought a new inkjet printer for $A17, just to get the (full sized) ink cartridge that retail for $40 each.

        1. LOL @ inkjet

          I have –
          1 x Canon BLP3000 mono LASER printer
          1 x HP LaserJet 1300 mono LASER printer
          2 x Xerox CP105 b color LASER printer
          1 x Xerox MF205fw color LASER multifunction
          0 x anything that uses ink as it’s far too expensive per page.
          These are just my home printers.

          I haven’t used or bought an ink based printer for over a two decades. Two decades ago, a mono LASER was $300 to $500 and still cheaper per page than ink printers. Now it costs $20 for a new el-cheapo mono LASER and $150 for a new el-cheapo color LASER. Sure the quality is crap but they’re still less per page than ink.

          Much older mono’s are far more reliable and cheaper to run and the cartridges are dirt cheap. Keep in mind that the cartridges that you buy with the printer (starter cartridges) have only 25-50% of the capacity of a separately purchased cartridge.

          All 3 of my color LASER printers use the same cartridge so soon I will be reversing the DRM on the cartridges (4 wire I2C or SPI??) and buying toner separately.

          All of these printers were given to me or bought second-hand. Most of them were faulty when I got them. Usually it’s a $3.60 paper separator and a $4.80 fuser film sleeve and there good for years and then after that they are usually repairable in most cases.

          I haven’t reversed one of these cartrige DRMs yet but unless it’s encrypted, it’s can’t be that hard. It’s probably just a SPI EEPROM or FLASH.

  21. This is the sort of product that I’d expect to see advertised in Wired magazine. Meanwhile back at the local fruit shop we picked up an entire box of lady finger bananas for $5 and the bags of apples were almost as good a deal. Fresh too, not seconds or whatever. The way I see it you need the break from whatever else you do, that you get from the time it takes to prepare food, so both the fresh food and the preparation as good for you. Whereas this machine is just bad for everything, including the environment.

  22. If Apple designed a juicer it would be like this. Apple wastes money having gorgeous, fully machine cut screws made to attach parts buried deep inside their laptops where nobody will ever see them until the day the laptops are broken down for recycling. Or they’ll never be seen at all if someone just tosses their PowerBook or MacBook in the trash when the battery dies and inflates. Cheap screws with stamped heads and roll formed threads would work just as well and get driven in by the millions by robots every day.

    Today I installed a new battery and trackpad into a Unibody MacBook. Milling the entire upper body of the thing from a single slab of aluminum is gross conspicuous consumption, despite aluminum being one of the most common elements and that the shavings would get recycled.

    It had a dent in the edge by the hard drive. There are plastics that would not only not have dented they wouldn’t break at the level of force that made that dent – and they’d cost far less to produce that casing part, plus would be lighter.

    A while ago I completely tore down an aluminum PowerBook G4 that was totally dead. The MacBook used the same design concepts. *Everything* connects to the mainboard with a *cable*. Even if the connector on the component butts right up against an edge of the board, there’s a cable to connect it. Super precision screws everywhere, though not as many as in the older PPC model due to the one piece billet frame where the older model had a die cast (magnesium?) frame screwed to posts (welded somehow?) on the stamped aluminum shell.

    Apple computers from 1977 until the Death of Beige were mostly elegant examples of engineering and design made to minimize how many parts were used and make them economical to produce. (While still suckering people into paying far more than the production cost.) Many of their computers were made to be very easy to open and take apart for repairs and upgrades. One example, the desktop G3. Pull the cover off and the innards tipped up to the sides, with automatic kickstands so they wouldn’t fall back and mash your fingers. IIRC the entire computer could be taken apart without tools.

    But lately it’s like Apple is living in Frederik Pohl’s “Midas World” where to consume the rampant production of *everything* produced by machines, the things humans design for the machines to produce are insanely complicated, using as many parts as possible to do simple functions.

    Perhaps Apple has been inspired by the height of mechanical cash registers and tabulating machines where the manufacturer’s ads would boast of how many parts were in them, and put glass panels on their sides to showcase all the gears. Except Apple has forgotten the glass panels so we can’t see all the pretty parts.

  23. To better understand the design behind this juicer, it might be helpful to start with the design of an existing cold press juicer: the Norwalk Juicer, a product that retails for $1000 and up. Here’s a short video showing how it is used:
    https://www.youtube.com/watch?v=87qDxfAEIkI
    For the Norwalk, you start with raw veggies and go through a 2-step process of first grinding them into little bits, then using a press to squeeze all the juice out of the bits. The Juicero follows the same process, except that they do the first part themselves, and just give you the bags to squeeze out at home. This saves you all the prep work and cleaning.

    The Juicero is obviously aimed at people who’d like to use a Norwalk but are too lazy. This is for people who believe the spiel about cold-pressed juices being more nutritious than juices created by other means. Of course, such folks have not factored in how much nutrition is lost by the delay between grinding up the veggies, storing them in bags that are shipped to you, then storing them in your fridge until you are ready to squeeze them.

    What I’d like to see is a study that really tests how juices made in various ways compare nutritionally, quantitatively. Is cold-pressed really any better than juice from a masticating juicer? If so, by how much? And how does either compare to juice made in a blender, filtered and not? I have a feeling it’s splitting hairs. In any case, such a study probably wouldn’t affect the folks who want “the best possible juice”, and will spend silly amounts of money to get it.

    1. Decades ago we had already centrifugal juicers, they ground the apples, carrots, etc and in the same step extracted the juice be centrifugal force. I can not see how pressing with plates should be any better.

    2. I have a “masticating” type juicer that attaches to the Kitchenaid powerhead (a device that is like the Demel of kitchen appliances.

      Cuts, grinds and squeezes fruit and veg in one pass at a pretty low RPM. especially compared to “juiceman” type centrifugal systems.

      Adjustable for pulp output too. Gets a LOT of juice out of most things I feed in-makes almost dry pulp from oranges, for example.

      If I include the powerhead unit AND the juicer assembly at new retail I’m about where the juicero starts (nowhere near the price of the Norwalk new), and I have a device that can be, and is, used for other tasks.

      The ONLY thing the juicero offers is the one thing you mention in your third paragraph: convenience. It takes a few minutes to take down and clean my juicer assembly. I suspect the takedown on the Norwalk is worse.

      Realistically not sure the “convenience” of specific, expensive and non-transparent (transparency being both visual and ‘is it really organic/sourced sustainably and/or responsibly?’ regardless of what’s on the label)

      I almost always *know* where the oranges and apples come from as I picked em myself when possible. Or get em from the farmers’ market from some people I trust.

      Hell, imagine you’ve got a couple of those $35 a bag juice packs and find they’ve gone south. Toss em out and go “oh well”? A bag of fruit, if it goes bad, is a LOT less.

    3. The only advantage I can see with a press type juice is it might get a little more from difficult to juice fruits and vegetables like apples or carrots but the end product would still be the same though maybe with a little less pulp.

  24. can’t say Juicero has designed the press right, but have you ever roll-pressed a bag with soft bits (fruit flesh) PLUS hard bits (fruit seeds) in it?
    Either the liquid will flow to the back side of the bag OR you also crush the seeds too (depending whether the rolls do seal the front/full side from the back/”empty” side of the bag).
    Crushed/ground seeds may add a bitter or otherwise undesired note to the flavour.
    The industrial presses for grape/orange/applejuice and winery I know of (but I’m not a reference) avoid crushing the seeds.
    Juice centrifuges I saw also leave the seeds intact.

    Maybe the QR-code on the bag controls the press only downto a rest distance (headroom?) depending on the bag content, say seed-size.

    Removing seeds and stuff at the filling of the bags could be too tedious to make the the bag content suitable for a roller press… who knows.

    My 2cts and bowel-feeling do not trust juice made of wholly ground fruits… Fruit peels too anyone?

  25. the article is meant as a warning, but sadly there will be those who will, like the machine’s creator/pitchman, who take it as a blueprint to success.

    Because a few people made out like bandits, the model of “overprice and justify with over-engineering and claims of convenience” will continue as long as there is enough “disposable income” to be harvested.

    This device is pretty much EVERY single technological ill of modern society. Limited use (unitasker) very expensive, DRM dependent, locked to a single manufacturer, and clogged with internet connectivity. When there are better options available.

    It’s like HP and Apple got together and designed a sink faucet.

  26. “overengineered” is an ambiguous term broadly applied to any design aspect which non-engineers dislike.

    These non-engineering types typically like to consider themselves technically above-average yet never actually been in the position to design a product for a paying customer or really contribute to society in any technical way.

    I look at this project and see a design-by-committee project farmed out to engineers who although might be good at their job, lack the communication channel to create a decent product.

    This lack of communication is likely due to different languages or the committee being assholes. Revision H on the mold tool is a good indicator.

  27. While this Juicer is a bad example, because it is useless, this is still an example why the US doesn’t export expensive machines like Japan and Germany do: when you see an expensive, well built machine, you shit on it. Americans are all over cheap prototypes, shipping quickly. That’s why US companies basically own the internet. Doesn’t work so well for cars or machines-building-machines.

  28. Someone needs to cut one of those bags open. I suspect there’s little, if any “juicing” going on. The hand squeezing looks like what’s really in the bags is already juice and the whole thing is an expensive scam.

    The expensive CNC machined metal parts should have been adapted for production as die cast aluminum. If some of those parts couldn’t withstand the tensile stress, make them aluminum forgings, finished where needed by minimal machining. The gearing should be adapted to use stock gears. The door locking mechanism is just plain silly overcomplicated. Have then never heard of a lever and a Hall effect sensor? Close and latch the lever and a magnet in the door trips the sensor, enabling the press.

    It’s like a machine from “The Midas Plague”, built complicated with far more parts than it needs, just to absorb some of the prodigious industrial output of the robots and automated factories.

    Next step, adapt the design to use diced fresh fruits and vegetables, and design and sell a gizmo that dices the input to the perfect size.

  29. The old saying about the three dangerous things in IT (a hardware person with a software patch, a software person with a soldering iron, and a user with a bright idea) can add “a foodie building hardware”. Since nearly every bit of nutrition “facts” are food industry or commodity group sponsor-bias/P-hacking BS, the mechanical result is predictable. The profits are real though.

    From the Bloomberg article: “There are 400 custom parts in here,” Evans told Recode. “There’s a scanner; there’s a microprocessor; there’s a wireless chip, wireless antenna.”

    Probably four miles to the gallon, 893 horsepower, reverb radio, six-way power mud-flaps and a furry dash too.

  30. Well actually it’s not over engineered as such, it was just based on stupid ideas. The idea obviously came from marketing. No sane engineer would ever have thought about that, after all it takes effort and makes the product worse.
    The design, pressing evenly on the whole surface probably also came from marketing, they probably wanted something like a gentle squeeze, and the outside of the device probably was already designed before consulting with the first engineer.
    Also any engineer would have added a display and at least some buttons, so you wouldn’t have to use your mobile phone. It also wouldn’t have been connected to anything but power. There’s little to no benefit in having it connected to the Internet.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.