Jump to content

Recommended Posts

ultem_mixer.thumb.jpg.a33a3b363488cf0df7b53c9239a34665.jpg

Cracking up

Now that cafes have been starting to use our espresso machines in heavy use situations, we're now able to tell what components are up to the task.  

About six months ago, Charles Temkey came on board. He's an engineer specialized in the longevity of marine equipment. He completely took apart his DE1PRO, after making several thousand shots with it, to find what was wearing. He identified two weaknesses, which we addressed with the v1.1 launch in January.

A month ago, a customer developed a leak in the mixing chamber of their Decent espresso machine.  This is where hot and cold water comes together with temperature sensors and flow constrictors, so that we can dynamically change the espresso water temperature.

We found cracking around the inserted flow constrictors. (see photo)  

We've identified two causes of this: (1) the flow constrictors are inserted under too much pressure, putting the Ultem resin under strain, and (2) the semi-transparent Ultem we're using allows cracks to propagate.

We're currently testing a variant material called Ultem 2100F, which mixes 10% glass fibers into the resin, for hugely increased crack resistance.  It's no longer translucent.  This material is what we're going to use in the v1.3 machines, which we'll start building in November.  

Since these cracks would fall (in my opinion) under the class of "manufacturer's mistake" Decent will repair any Ultem (v1.0 and v1.1 machines) that had cracks and leaked, forever, and for free. 

Note that we've only seen cracks so far in a few machines (I think it's less than 5) but that's also probably because cafes have been cautiously (and slowly) ramping up how close they put Decent espresso machines in their main line.

We've also made a jig to insert the flow constrictors, which puts less strain on the Ultem, once the constrictor has been inserted, and started using that right away.

One reason we've been slow to put out a DE1CAFE model, is that we want to learn everything we can about our DE1PRO line, how it can fail, and how to avoid that, before we make a model that we claim is ready for any use.  

It's largely a question of getting our machines out there, in real use, getting extremely involved when there is a failure, and repeatedly addressing each area of failure.  

I suspect that the process will take about another year, and we'll continue to hugely support the cafes kind and brave enough to trust our new company with their business.

-john

  • Like 2

Share this post


Link to post
Share on other sites

John, how did the leak present itself to your customer?  What were the signs?


Strietman CT1, DE1PRO, EG-1 v2, HG-1, Kinu M47, Baratza Vario brew, Lido E, RB 58.5, Reneka Micro Sieve, VST

Share this post


Link to post
Share on other sites
7 minutes ago, Bolta said:

John, how did the leak present itself to your customer?  What were the signs?

Since this is the main mixing chamber, it means:

1) you can't make much pressure during espresso

2) water leaks out the left and right sides, as designed.  There is an air gap under the machine, and there are UL-mandated drain holes inside the chassis, to drain into that.  From standing height, this air gap isn't visible, as it's in shadows.

Photo (and arrow) below shows where water comes out, in the event of a leak.

Note that while I might complain about Intertek's waffling on Bluetooth acceptability, we did hire them as consultants all through our 3 year R&D phase, and their many safety suggestions were all implemented, and I felt they were all worthwhile.  Some amount of failure is inevitable, the trick is to cope with it gracefully.

1941400542_ScreenShot2019-07-11at5_56_02PM.thumb.jpg.82ccf49c513fa373201fa9f0cb83de96.jpg

Share this post


Link to post
Share on other sites

Just spent the entire weekend reading this entire thread from the beginning - wew - kind of sad to be done!!!

Recently purchased a Niche Zero Grinder and love it.  No longer have to argue with the wife over which coffee to buy.  She gets what she likes and I get my favorite beans.  Awesome zero retention grinder!

Having a real grinder improved our espresso enough using the Breville Barista - I’ve now got the OK to purchase a DE1Pro.  Happy days ahead!

Question:  I know the future version 1.3 will cost an additional $300, which will be available to swap out if I purchase the 1.1 now.  (Thank you!!!). Another difference is the Ultem cracking, but seems like the jig apparatus on the current build likely helps with that problem - plus that’s also been recognized as a design flaw with lifelong warranty.  (Again thanks for sharing)

Few small questions regarding the tablets.  Looks like there is a new updated tablet with updated Android OS, faster cpu, and  more memory on the near horizon.  Is the screen any different and does it ‘feel faster’ when using the app?  Ie any reason to wait?  Also, is the tablet on the DE1Pro supported by the bracket with 2 holes in the case - or is it using the magnetic holder?  Which brings me to final question.  It looks like the tablet can be optional in the future.  I haven’t seen much discussion  on this topic?  Is the thought process that most people will continue to keep their tablet as is currently?

  • Like 1

Share this post


Link to post
Share on other sites
On 15/07/2019 at 01:03, LostCreekSooner said:

Question:  I know the future version 1.3 will cost an additional $300, which will be available to swap out if I purchase the 1.1 now.  (Thank you!!!). Another difference is the Ultem cracking, but seems like the jig apparatus on the current build likely helps with that problem - plus that’s also been recognized as a design flaw with lifelong warranty.  (Again thanks for sharing)

I wouldn't call it a "design flaw", what I'd call it is "a component that will fail earlier than the other components". 

Eventually, everything breaks at some point.  For the ultem-made mixin chamber, it appears that if you combine "too tight a flow constrictor" with "commercial use" with "cold water clean daily", then you can cause this material to break before the other components have started to fail.

Now that we've extended the lifetime of 3 components that would fail first, we're now waiting to find the next component on the list, that will fail before the others.  

On 15/07/2019 at 01:03, LostCreekSooner said:

Few small questions regarding the tablets.  Looks like there is a new updated tablet with updated Android OS, faster cpu, and  more memory on the near horizon.  Is the screen any different and does it ‘feel faster’ when using the app?  Ie any reason to wait?  Also, is the tablet on the DE1Pro supported by the bracket with 2 holes in the case - or is it using the magnetic holder?  Which brings me to final question.  It looks like the tablet can be optional in the future.  I haven’t seen much discussion  on this topic?  Is the thought process that most people will continue to keep their tablet as is currently?

Screen is identical, same resolution and brightness. App speed feels identical.  However, more memory and faster CPU was needed to run a newer Android version (8.1 vs 5.1) at the same perceived speed, so that's why the newer tablet is beefier.   However, this also means that the newer tablet, being more modern, is able to run non-coffee-related software much better than the current tablet.  If that's what you want to do with it (up to you), then the newer tablet is better for you.

The DE1PRO is staying with the magnetic Steelie stand.

A tablet will be optional for daily use (making drinks), but essential for creating profiles, seeing the charts, etc...  The "no tablet" crowd is very likely to only be cafes, who don't want their employees to fiddle, and they want to lock down the recipe used.

-john

 

  • Like 1

Share this post


Link to post
Share on other sites
On 11/07/2019 at 17:42, decent_espresso said:

Cracking up

Now that cafes have been starting to use our espresso machines in heavy use situations, we're now able to tell what components are up to the task.  

About six months ago, Charles Temkey came on board. He's an engineer specialized in the longevity of marine equipment. He completely took apart his DE1PRO, after making several thousand shots with it, to find what was wearing. He identified two weaknesses, which we addressed with the v1.1 launch in January.

A month ago, a customer developed a leak in the mixing chamber of their Decent espresso machine.  This is where hot and cold water comes together with temperature sensors and flow constrictors, so that we can dynamically change the espresso water temperature.

We found cracking around the inserted flow constrictors. (see photo)  

We've identified two causes of this: (1) the flow constrictors are inserted under too much pressure, putting the Ultem resin under strain, and (2) the semi-transparent Ultem we're using allows cracks to propagate.

We're currently testing a variant material called Ultem 2100F, which mixes 10% glass fibers into the resin, for hugely increased crack resistance.  It's no longer translucent.  This material is what we're going to use in the v1.3 machines, which we'll start building in November.  

Since these cracks would fall (in my opinion) under the class of "manufacturer's mistake" Decent will repair any Ultem (v1.0 and v1.1 machines) that had cracks and leaked, forever, and for free. 

Note that we've only seen cracks so far in a few machines (I think it's less than 5) but that's also probably because cafes have been cautiously (and slowly) ramping up how close they put Decent espresso machines in their main line.

We've also made a jig to insert the flow constrictors, which puts less strain on the Ultem, once the constrictor has been inserted, and started using that right away.

One reason we've been slow to put out a DE1CAFE model, is that we want to learn everything we can about our DE1PRO line, how it can fail, and how to avoid that, before we make a model that we claim is ready for any use.  

It's largely a question of getting our machines out there, in real use, getting extremely involved when there is a failure, and repeatedly addressing each area of failure.  

I suspect that the process will take about another year, and we'll continue to hugely support the cafes kind and brave enough to trust our new company with their business.

-john

Hi John,

Would buying a reinforced grade of plastic (PEI is used I assume because of temperature limits) be any use to prevent propagation of cracks? I just ask as we have had a similar work-related issue with nylon's cracking *specifically high glass-filled content - it becomes very brittle*, and one of the solutions was to use a rubberised grade. It has exactly the same capabilities, but the rubber in it prevented crack propagation.

Furthermore - why PEI? Would something like a PA66 not also be suitable for similar temp. ranges?

Sorry, I know they're nerdy questions - just querying the choice.

Cheers.

Share this post


Link to post
Share on other sites
38 minutes ago, Cooffe said:

Hi John,

Would buying a reinforced grade of plastic (PEI is used I assume because of temperature limits) be any use to prevent propagation of cracks? I just ask as we have had a similar work-related issue with nylon's cracking *specifically high glass-filled content - it becomes very brittle*, and one of the solutions was to use a rubberised grade. It has exactly the same capabilities, but the rubber in it prevented crack propagation.

Furthermore - why PEI? Would something like a PA66 not also be suitable for similar temp. ranges?

Looking at the PA66 you recommend, I don't see it it being used in any food situations:

http://www.performance-plastics.co.uk/product/classification/nylon-66-pa66/

As to why PEI/Ultem, we're using it because it's extremely unlikely to taint water flavor (it's commonly used medically), very high temperature, can handle high pressures, and is fairly tough.

If PA66 has a food-grade variant, I'd very much like to know about it.

As to Ultem, once our designs stabilize, we'll be investing in molds, rather than CNC carving out of blocks, and then the cost comes way down, as we will then be wasting far less material.

Share this post


Link to post
Share on other sites
8 minutes ago, decent_espresso said:

Looking at the PA66 you recommend, I don't see it it being used in any food situations:

http://www.performance-plastics.co.uk/product/classification/nylon-66-pa66/

As to why PEI/Ultem, we're using it because it's extremely unlikely to taint water flavor (it's commonly used medically), very high temperature, can handle high pressures, and is fairly tough.

If PA66 has a food-grade variant, I'd very much like to know about it.

 As to Ultem, once our designs stabilize, we'll be investing in molds, rather than CNC carving out of blocks, and then the cost comes way down, as we will then be wasting far less material.

Hi John,

Have done a bit of looking into this. As I understand most Nylons are food safe, but the colourant used isn't, so as long as it is in its "Natural" colour (not black/white etc.) it should be OK. I know that the PEI you use is quite expensive (we've used it for certain applications within my workplace), however very geometrically stable under temperature, also it is temperature resistant. I'll PM you the grade of PA66 that I am thinking of. The PA66 may be more cost efficient when bought in bulk. I'd also argue that when tooling the part, the toolmakers will be more familiar with PA66 and be able to better suit the mould for shrinkage/warpage.

I can also see that the part your referencing looks to be fairly thick (>3mm), it may be that when you move this from CNC to an injection moulding, you'll need to work on the productionisation of the part (however I'm sure you know this) to prevent sink marks/warpage. Not sure on the quantities, but you can get Aluminium moulds that are fairly cheap, and coat them, and will last for 200k+ shots, which might be a better alternative to steel if you don't need a high clamp force (which is possible with the PA66).

Cheers,

Share this post


Link to post
Share on other sites
1 hour ago, Cooffe said:

Have done a bit of looking into this. As I understand most Nylons are food safe, but the colourant used isn't, so as long as it is in its "Natural" colour (not black/white etc.) it should be OK. I know that the PEI you use is quite expensive (we've used it for certain applications within my workplace), however very geometrically stable under temperature, also it is temperature resistant. I'll PM you the grade of PA66 that I am thinking of. The PA66 may be more cost efficient when bought in bulk. I'd also argue that when tooling the part, the toolmakers will be more familiar with PA66 and be able to better suit the mould for shrinkage/warpage.

I can also see that the part your referencing looks to be fairly thick (>3mm), it may be that when you move this from CNC to an injection moulding, you'll need to work on the productionisation of the part (however I'm sure you know this) to prevent sink marks/warpage. Not sure on the quantities, but you can get Aluminium moulds that are fairly cheap, and coat them, and will last for 200k+ shots, which might be a better alternative to steel if you don't need a high clamp force (which is possible with the PA66).

This is all incredibly helpful, thank you.  Looking forward to your PM with more details.

-john

 

Share this post


Link to post
Share on other sites

Opaque Mixing

We've been testing the new glass-fiber mixing chambers this week.  This new material will prevent cracks from propagating. So far, we can't find any performance difference, which is good.  You can see the current material, as it's amber-colored and fairly transparent.
In the photos, you can see us testing externally, separately, under a 19bar testing jig, and finally integrated into a machine.

-john

new_ultem.jpg

  • Like 1

Share this post


Link to post
Share on other sites

Drawings for DIY IKEA Coffee Cart

Here are the drawings (and renders) for our coffee cart, using IKEA's "Rimforsa" https://www.ikea.com/us/en/catalog/products/60399412/ table.

We've made two versions so far: for one or two espresso machines.  

The bamboo top is thick enough to hide much of the countersinking (of the machine and of the pitcher rinser). However, even though it looks hollow, there are vertical ribs every 20cm, which greatly complicate cutting and placement.  Our drawings show the ribs, to help you measure more exactly.

I'm also looking at the "Bror" cart from IKEA, as it comes with wheels.  It's smaller, but simpler and cheaper too.  Going to IKEA later today to check it out, and my jigsaw should arrive from Amazon this weekend.  I'm going to try to make my own cart, and hopefully get the OK to run a coffee cart at our local sunday farmer's market.

-john

WechatIMG356.png

WechatIMG355 1.png

Screen Shot 2019-07-19 at 11.29.04 AM.jpg

WechatIMG346.png

WechatIMG347.png

  • Like 1

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • About:

    Coffee Forums UK is the UK's premier coffee forum Started in June 2008 by Glenn Watson, we now have more than 22000 mainly UK based members, and welcome more than 3000 members and visitors from around the world each day! With strategic investment and digital expertise from the Jackson Lockhart team (Tait Pollack and Adam Bateman), we are taking Coffee Forums UK to the next level, and are delighted to share the journey with you.

    New Members:

    We are often referred to as the friendliest forum on the web and we look forward to welcoming you onboard.

    Terms of Use

    Advertising

    Coffee Forums Media Kit

    Buy Advertising Space

    Donate

    Get Your Supporter Badge (per year)

×
×
  • Create New...