Page 28 of 28 FirstFirst ... 18262728
Results 1,351 to 1,384 of 1384
Like Tree1061Likes

Thread: Decent Espresso Machines (DE1) - Any thoughts?

  1. #1351
    338
    338 is online now
    Senior Member
    Join Date
    Mar 2017
    Location
    Sydney
    Posts
    385
    GC, just noticed your spelling and assumed it is to match previous posters or the machines terminology. Just out of interest, does the software on your Decent allow for correct spelling of metric units, as suggested by the International System of Units (SI) or only as spelt by the only country in the world which wont adopt the recommended spelling? Assume in software it would be easy to have imperial, correct metric and USA metric.

    Either way wont affect taste in the cup
    RavenMad likes this.

  2. #1352
    gc
    gc is offline
    Senior Member
    Join Date
    Dec 2008
    Location
    Kuranda, Australia
    Posts
    174
    Quote Originally Posted by 338 View Post
    GC, just noticed your spelling and assumed it is to match previous posters or the machines terminology. Just out of interest, does the software on your Decent allow for correct spelling of metric units, as suggested by the International System of Units (SI) or only as spelt by the only country in the world which wont adopt the recommended spelling? Assume in software it would be easy to have imperial, correct metric and USA metric.

    Either way wont affect taste in the cup
    Ha! I actually changed my spelling to match post.

    In the app it just uses the safer "mm" and "ml". I guess DE have a big US market to think about (or hope to!)
    338 likes this.

  3. #1353
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by gc View Post
    +1 for milliliters....altho....at what depth do you run out of water? Or can you specify the 'danger' level in milliliters as well?
    3mm is the minimum we'll allow, before we ask you to refill the tank. That represents about 70ml of water, which is juuuuuuust enough to pull an espresso.

    We really don't want to run out of water during the shot, not because it's dangerous, but because it's a waste of good beans.

    Note that the app lets you set the refill point you want, so if you make 200ml Americanos, I'd recommend you set ~230ml that as the minimum the machine will allow you to start a drink with, so that you don't have to watch your espresso get cold as you refill the tank later on.

    -john
    gc likes this.

  4. #1354
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by 338 View Post
    GC, just noticed your spelling and assumed it is to match previous posters or the machines terminology. Just out of interest, does the software on your Decent allow for correct spelling of metric units, as suggested by the International System of Units (SI) or only as spelt by the only country in the world which wont adopt the recommended spelling? Assume in software it would be easy to have imperial, correct metric and USA metric. Either way wont affect taste in the cup
    There was a big discussion about a year ago, about how totally useless the US "ounces" measure is for coffee. Everyone agreed, and so American users of the DE1+ have "fahrenheit" for temperature but grams, and milliliters for weight and volume.

    Note that everything on the tablet is translated, including "mL", so (for instance) Korean users gets their own translation.

    -john
    338 likes this.

  5. #1355
    Senior Member
    Join Date
    Oct 2011
    Posts
    1,353
    Perhaps a % scale would be useful? Common in industrial equipment.

    Even better, a function which predicts how many shots (milk steaming included) based on historical usage patterns...

  6. #1356
    Member
    Join Date
    Jun 2018
    Posts
    38
    Quote Originally Posted by decentespresso View Post
    So today, I brought out the postal scale and measured the water level at each 100g increase in weight.
    <snip>
    My measurements are accurate more or less to a millimeter, as surface tension on my metal ruler was an issue.

    > ml = mm * 22 + (mm ^ 1.52)

    A side effect of my doing this was that I've now verified that the water tank really can take 2 liters of water. Our SolidWorks model said it could, but that's very much not reality (ie, ceramic shrinkage needs to be taken into account).
    A couple of notes: You are unlikely to end up with an exponent of 1.52 in a 3 dimensional object with substantially planar surfaces, so your measurements are likely to be out (as you note above).

    If you have a solidworks model of the void space in the tank it will be easy to work out a better formula for height vs volume by taking slices of the model at 1 mm increments. I use Autocad rather than solidworks so I can't tell you the exact procedure for your software but the general principle is that you can successively slice the object and query the object parameters which will give you an exact volume.

    Since shrinkage is linear it is a trivial procedure to convert from the model to the real world.
    Last edited by Lyrebird; 2 Weeks Ago at 12:26 PM.

  7. #1357
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by MrJack View Post
    Perhaps a % scale would be useful? Common in industrial equipment.
    Even better, a function which predicts how many shots (milk steaming included) based on historical usage patterns...
    "% full" to me is laziness on the part of the manufacturer.

    What I feel I want to know is "how many drinks before I need to refill?"

    Which is, of course, your second point.

    About 70ml is needed to make an espresso, hence that's the lowest water level we'll allow, since we don't want you to run out mid-shot.

    However, there's a BIG variable, which is water-for-Americanos, and to a lesser extent "are you steaming milk?".

    An average water consumption number based on historical patterns would work if you always make the same drink. If you sometimes need another 200ml for the occasional Americano, you can tell the DE1+ to require a refill at 270ml.

    This would be feature overkill, but what would be helpful is a chart of historical water consumption per drink, so you can make an informed decision as to what your water requirements are.

  8. #1358
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by Lyrebird View Post
    A couple of notes: You are unlikely to end up with an exponent of 1.52 in a 3 dimensional object with substantially planar surfaces, so your measurements are likely to be out (as you note above).

    If you have a solidworks model of the void space in the tank it will be easy to work out a better formula for height vs volume by taking slices of the model at 1 mm increments. I use Autocad rather than solidworks so I can't tell you the exact procedure for your software but the general principle is that you can successively slice the object and query the object parameters which will give you an exact volume.

    Since shrinkage is linear it is a trivial procedure to convert from the model to the real world.
    The Solidworks solution is a nice one, but I still will need a function fitted to the curve to give me continuous measurements. However, I will get an intern to do the modeling numbers for me, since as you say, they'll be more accurate (no surface tension on the ruler messing up my measurements).

    Currently, the middle of the water tank is where my estimated milliliters error is greatest (8.11% is the worst), because the the water tank suddenly gets a few centimeters wider. I adjusted the function so that the error was least at the two end-points (full, and empty)

    Below is a chart showing the inacuracy of my math function in converting mm to ml. The X axis label is in 100ml. It's quite a bit better at measurements when "almost full" and it is at "almost empty"

    currrent.jpg

    I fiddled with the function a bit, and found that I could trade accuracy-close-to-empty for accuracy-close-to-full. Here's what this function looks like:=B2*24 +(B2^1.3)

    Since I'm likely to be judged on how closely the full and empty numbers approach reality, this isn't likely a good solution either.

    fiddled.jpg

  9. #1359
    Member
    Join Date
    Jun 2018
    Posts
    38
    It is evident from the figures that you have a biphasic function with the step as the inflexion point plus you have some measurement errors which are worst at the lower end (look at the zigzags in your error graph). The solidworks trick will get rid of the measurement errors and since you are programming this it's not hard to do a biphasic function to take into account the step.

    It can be done with if statements:

    If h<=step: vol = f1(h), if h > step, vol = f2(h).

    Assuming step is at about 40 mm,

    f1 might be vol = 22 * h + h^1.45

    f2 might be vol = 1100 + 26.5* (h-40) + (h-40)^1.7

    The distribution of the errors with that function makes it obvious that they are consequences of the limitations on measurement (for instance changing the first datapoint to 100ml / 4.2 mm reduces the error substantially)

    If you must have a continuous single function you can often fake a biphasic function by inserting a trigonometric function.
    Last edited by Lyrebird; 2 Weeks Ago at 03:04 PM.

  10. #1360
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by Lyrebird View Post
    It is evident from the figures that you have a biphasic function ... since you are programming this it's not hard to do a biphasic function to take into account the step.
    Sigh, now why didn't this solution occur to me?

    Yes, I think you're 100% right. Two functions.

    Many thanks.

    -john

  11. #1361
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by Lyrebird View Post
    It is evident from the figures that you have a biphasic function with the step as the inflexion point plus you have some measurement errors which are worst at the lower end (look at the zigzags in your error graph). The solidworks trick will get rid of the measurement errors and since you are programming this it's not hard to do a biphasic function to take into account the step.
    Here's a photo of the tank.

    IMG_8980.jpg

    There are two challenges with the water level mm->ml function:
    1) the hand-hold and widening shape of the tank about 4.5cm in height
    2) the draft angle to the ceramic, required so that it can come out of a mould.

    #1 is pretty well solved by your suggestion of "different functions for different heights" and this is what fit my data best.

    if {$mm < 18} {
    set exp 1.35
    } elseif {$mm < 48} {
    set exp 1.4
    } else {
    set exp 1.45
    }
    set ml [expr {((( $mm * 23) + ([::tcl::mathfunc:ow $mm $exp])))}]


    #2 is hard to solve currently, since my function is working off the false idea that the water vessel has vertical sides, when actually they're sloping. That's why I have that last exponent change.

    However...

    Quote Originally Posted by Lyrebird View Post
    The distribution of the errors with that function makes it obvious that they are consequences of the limitations on measurement (for instance changing the first datapoint to 100ml / 4.2 mm reduces the error substantially)
    You're right.

    This is a bigger problem, since fitting a function to points that are mis-measured, isn't a great use of my time. So... this set of maths will do for now, until I get some better measurements using CAD.

    This 2nd pass does greatly reduce the errors a lot, however.

    screen 2018-07-06 at 1.40.39 PM.jpg

  12. #1362
    Member
    Join Date
    Jun 2018
    Posts
    38
    The sloping sides are no problem, mathematically they can be modelled as a truncated pyramid.

    I'm pretty sure you can make SolidWorks export to .dwg format. If you do that and send it to me I can probably work out an exact function for you, including the handholds.
    Lav likes this.

  13. #1363
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by Lyrebird View Post
    The sloping sides are no problem, mathematically they can be modelled as a truncated pyramid.
    You can, because ahem you didn't fail maths in Uni.

    Quote Originally Posted by Lyrebird View Post
    I'm pretty sure you can make SolidWorks export to .dwg format. If you do that and send it to me I can probably work out an exact function for you, including the handholds.
    Wow, now there's an offer I can't refuse! Nagging an engineer now.

    -john
    gc, MrJack and Lav like this.

  14. #1364
    Member
    Join Date
    Jun 2018
    Posts
    38
    I use a fair bit of maths in one of my day jobs which is designing and building bicycles. Google my username followed by "cycles".

    I'd post a link but that might be construed as a non sponsor link.

  15. #1365
    Senior Member Barry O'Speedwagon's Avatar
    Join Date
    Dec 2011
    Location
    PRL
    Posts
    2,380
    Quote Originally Posted by Lyrebird View Post
    The sloping sides are no problem, mathematically they can be modelled as a truncated pyramid.

    .
    I presume the volume is just that of 'big pyramid' - 'small pyramid'?

  16. #1366
    Member
    Join Date
    Jun 2018
    Posts
    38
    Yep.

    If the base is square and the release angles are equal the height of the big pyramid is width / (2 . tan(release angle)).

    It's more complicated if the base isn't square and / or the release angles aren't equal but I can run you through the maths if you like.

  17. #1367
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by Lyrebird View Post
    I use a fair bit of maths in one of my day jobs which is designing and building bicycles. Google my username followed by "cycles".I'd post a link but that might be construed as a non sponsor link.
    Here's a link to the DWG file:
    https://decentespresso.com/p/watertank.dwg

    screen 2018-07-06 at 5.42.22 PM.jpg

  18. #1368
    Member
    Join Date
    Jun 2018
    Posts
    38
    Thanks, but the file opens as a 2D object in read only so I cannot do anything with it.

    I do not know whether the problem lies with AutoCAD or the original file.

  19. #1369
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618

    Progress on plumbing

    plumbing.jpg

    We're not yet shipping the plumbing kits for our DE1PRO customers, because I haven't yet been satisfied with our designs. Here's an update.


    DRAINING THE MUCK
    For the past two years, we've been drilling a hole in our ceramic drip trays and attaching a fitting to drain the mucky water out. This has worked "fairly well". By that, I mean that it works, but the fitting tends to leak. The basic problem is that the fitting is above the bottom the of the drip tray, so there's always "standing water", which eventually seeps through the fitting.

    Someone pointed me to this "drip pan" design, which seems to solve this problem. https://www.chriscoffee.com/Drain-Kit-Alex-p/adrkt.htm

    The top two photos show our design to try this idea out. We'll still drill a hole into our ceramic tray, but there will be no fitting. Instead, the mucky water drains into a drip pan, which has a tube fitted to it. I do still have a concern about leaking on the new fitting leaking, but sealing metal-against-metal is a lot easier than sealing metal-against-ceramic.


    PLUMBING KIT TO GO
    A not insignificant number of DE1PRO users want to travel, and if they're cafes doing catering, they want to be able to "plumb in" when they get there. So... we're going to repurpose the Tamping Kit suitcases that we have way too many of and use that to ship the plumbing kits to customers.

    That way, you'll have 3 suitcases you can bring everything in:
    - your DE1PRO suitcase
    - your Barista Kit suitcase
    - your Plumbing Kit suitcase

    The Tamping Kit https://decentespresso.com/tamper has been a bit of a dud because most people opt to pay a bit more and get the way-more-functional Barista Kit https://decentespresso.com/barista_kit ... So I have a warehouse full of tamping kit suitcases that I don't know what to do with. They'll now get repurposed as "Plumbing Kit To Go"

    One note on the photo: we're including tubing made from surgical silicone. Clear tubing for the "clean water in" path, and black tubing for the "dirty water out". That way, while setting up your DE1PRO at a gig, you won't accidentally hook them up backward and dirty your clean water tank. But I've never made that mistake, of course. Cough.

    -john

  20. #1370
    Member
    Join Date
    Jun 2018
    Posts
    38
    An alternative would be to remould* the drip tray so it incorporates a short tubular extension in the base.

    Use a precision diamond drill when drilling the drain hole and make a drain fitting incorporating a radial O ring seal that inserts from underneath.

    No muss, no fuss.



    * Yes I know a new mould is megabucks. Maybe worth considering if you have to remould for other reasons.

  21. #1371
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by Lyrebird View Post
    An alternative would be to remould* the drip tray so it incorporates a short tubular extension in the base.
    We have gone down that path as well, and modeled it up. Here's a rendering of it.

    There are very real limits to what shapes fired ceramic will take, so that's why this design is so simple.
    preview-full-image.pngpreview-full-Another idea.jpg


    Quote Originally Posted by Lyrebird View Post
    Use a precision diamond drill when drilling the drain hole and make a drain fitting incorporating a radial O ring seal that inserts from underneath.
    That's exactly what we've been doing here for the past 2 years. We have a drill press and diamond drill tips. We've made about 30 of these. Way back in this forum, there will be discussions here about how to drill cleanly through very hard porcelain, a problem with (cooperatively) did get solved.

    This approach works well, except that it does drip slightly from the remaining sludge that collects around the drain hole. By "drips slightly" I really mean slightly: perhaps 10 drops per day.

    Quote Originally Posted by Lyrebird View Post
    Yes I know a new mould is megabucks. Maybe worth considering if you have to remould for other reasons.
    Actually, a ceramic mould is surprisingly inexpensive (under USD$500) but the minimum order quantity of 1000 is an issue, if the design doesn't work.

    We might still try this design, with the built in drain.

  22. #1372
    Senior Member
    Join Date
    Aug 2013
    Location
    Rockingham W.A.
    Posts
    1,073
    Quote Originally Posted by decentespresso View Post
    We have gone down that path as well, and modeled it up. Here's a rendering of it.

    There are very real limits to what shapes fired ceramic will take, so that's why this design is so simple.
    preview-full-image.pngpreview-full-Another idea.jpg



    That's exactly what we've been doing here for the past 2 years. We have a drill press and diamond drill tips. We've made about 30 of these. Way back in this forum, there will be discussions here about how to drill cleanly through very hard porcelain, a problem with (cooperatively) did get solved.

    This approach works well, except that it does drip slightly from the remaining sludge that collects around the drain hole. By "drips slightly" I really mean slightly: perhaps 10 drops per day.


    Actually, a ceramic mould is surprisingly inexpensive (under USD$500) but the minimum order quantity of 1000 is an issue, if the design doesn't work.

    We might still try this design, with the built in drain.
    G'day John

    As the plumbed in drip tray is completely different from the standard "non plumbed version", I would suggest that you angle the sides a lot more to reduce the actual total water capacity. There is no point having anything beyond a funnel shape (or as close as you can get to it) together with a few locating lugs (3? 4?) to hold it in place. Removing / smoothing any corners will also avoid trapping excess water. The greater the internal slope, the less water retention within the drip tray. The two reasons a) less capacity to store water and b) the greater the slope the faster the water will move down to the drainhole (which should be as large in diameter as possible - think "funnel" again). Recessing the drainhole can also help reduce any pooling issues.

    Hope this helps

    TampIt
    PS: Increasing the slope just before the drainhole also helps.
    Last edited by TampIt; 1 Week Ago at 02:45 AM.
    RavenMad and level3ninja like this.

  23. #1373
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618

    Decent Espresso CAD files released

    screen 2018-07-09 at 12.25.05 PM.jpg

    As promised two years ago, here are downloadable CAD files for our DE1+ and DE1PRO espresso machines.
    https://decentespresso.com/de1plus#cad

    What can you do with these files?
    - you might want to "hot rod" or "accessorize" your Decent Espresso machine
    - you’re planning a cafe or kitchen, and want to decide where to put the machines
    - you’d like to have the parts models so that you can CNC your own replacement parts
    -to help you do your own future repairs

    You can work with the model using the free web-based OnShape CAD software: https://cad.onshape.com/documents/85...a4f513a8a6cbe1

    The STEP file loads well into most CAD programs, including https://www.freecadweb.org/ and is probably better for most serious use.

    -john
    Andy and magnafunk like this.

  24. #1374
    CoffeeSnobs Owner Andy's Avatar
    Join Date
    Mar 2004
    Location
    Internet
    Posts
    15,897
    Blog Entries
    1
    Quote Originally Posted by decentespresso View Post
    What can you do with these files?

    Who's going to be the first person to 3D print a miniature Decent Espresso machine?
    Javaphile, Lav and 338 like this.

  25. #1375
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by Andy View Post
    Who's going to be the first person to 3D print a miniature Decent Espresso machine?
    Hey Andy, have you seen this?

    DcQzgeWW0AIF4e9.jpg

  26. #1376
    Super Moderator Javaphile's Avatar
    Join Date
    Dec 2004
    Location
    Earth!
    Posts
    14,653
    3D printed with fully movable/removable parts!


    Java "Do I smell a 3D printer at work?!?" phile
    Toys! I must have new toys!!!

  27. #1377
    Senior Member
    Join Date
    Oct 2011
    Posts
    1,353
    Quote Originally Posted by Javaphile View Post
    3D printed with fully movable/removable parts!


    Java "Do I smell a 3D printer at work?!?" phile
    Looks like LEGO?

  28. #1378
    Senior Member JMcCee's Avatar
    Join Date
    Feb 2017
    Posts
    191
    The espresso has hints of polymer and nylon
    Javaphile likes this.

  29. #1379
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618

    A few 110V DE1+ machines, shipping today, at a $500 discount.

    A few 110V DE1+ machines, shipping today, at a $500 discount.

    I have a half dozen DE1+ machines in stock, that I will sell at the early-adopter price of $1999 ($500 off the current price).

    Here's the catch:

    1) This offer is only for people who have helped us along this journey by posting to the forums. You must have previously posted something before we started shipping (February 2018) to our discussions on Home Barista, Coffee Forums UK, Coffee Snobs, our Instagram, or on the Facebook group.

    2) these machines have a slight cosmetic defect, of a rounded scratch around the front panel mounting holes. Photo attached. You'll need to be OK with that.

    ---

    If you'd like to take me up on this, send me a personal message and include a link to one of your pre-March postings. I'll follow up with you directly afterward.

    Sorry.... no 220V machines available under this offer at the moment (stay tuned).

    -john

    IMG_9014.jpg
    IMG_9015.jpg

  30. #1380
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618

    Special offer sold out

    IMG_9031.jpg

    All the remaining DE1+ machines I had under the special offer are now sold out, and shipping today. 9 people took advantage of this, and I look forward to hearing from in the coming days, as you start pulling (hopefully decent) shots.

    -john



  31. #1381
    Junior Member
    Join Date
    Jul 2018
    Posts
    2
    But John, the photo only shows 7 machines?

    This was a great offer and I was pleased to be able to take advantage. Many thanks!

  32. #1382
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618
    Quote Originally Posted by tegwj View Post
    But John, the photo only shows 7 machines?
    Two machines under this offer shipped yesterday, the other 7 going out today.

    We're making 25 "perfect mirror panel" machines at 220V at the moment. With about 100 people remaining on the pre-buyer list, we should be wrapped up at the end of August, and then shipping v1.1 DE1PRO machines to those who are buying from us at the moment. We're able to ship about 60 machines a month at the moment.

    -john

  33. #1383
    Junior Member
    Join Date
    Jul 2018
    Posts
    2
    Quote Originally Posted by decentespresso View Post
    Two machines under this offer shipped yesterday, the other 7 going out today.
    Brilliant. I don't know if you have a clean way to do it but if there is any geographical correlation with customers you can share (e.g. is anyone else in the Washington, DC area like me?) it would be good to know.

    I shall refer to the imperfection on my mirror panel as the "20% Character Mark" and treat it with appropriate reverence.

    If anyone is interested in a Vibiemme Domobar Double Boiler v4.0 in Northern Virginia, please let me know!

  34. #1384
    Site Sponsor
    Join Date
    Aug 2016
    Location
    Hong Kong
    Posts
    618

    Chinese promotional video from our CNC vendor

    This company makes the brass parts that go into the group head of our espresso machine. Every time you see brass in this video (and that’s often) those are Decent Espresso parts. An interesting peek inside high-end Chinese manufacturing.

    Some amusing moments:
    - the ping pong tables and treadmill
    - all the girls on the right side of the table, all the boys on the left side.
    - nice matching shirts
    - really, really nice CNC and testing equipment
    - amazingly good English text-to-speech narrator (with a slightly Nordic accent?)



    -john

Page 28 of 28 FirstFirst ... 18262728

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •