Our Valued Sponsor
OpinionsConsumer ReviewsGuides and How TosCoffeeGeek ReviewsResourcesForums
Espresso: Espresso Mods and Restorations
Custom Thermocouple for Gaggia Classic
Rocket R58 Double Boiler
Rocket Espresso R58 Double Boiler -  Everything you need for the perfect shot!
www.seattlecoffeegear.com
 
Not Logged in: Log In to Postlog in
New Topics updated topics   New Posts new posts   Unanswered Posts new unanswered  
Search Discussion Board search   Discussion Board FAQ faq   Signup sign up  
Discussions > Espresso > Espresso Mods > Custom...  
view previous topic | view next topic | view all topics
showing page 5 of 6 first page | last page previous page | next page
Author Messages
jonr
Senior Member


Joined: 25 Jun 2013
Posts: 298
Location: Americas
Expertise: I like coffee
Posted Thu Aug 8, 2013, 1:36pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

Thanks.  I hadn't thought about a SD card for logging - I've always sent things to the serial port and then captured them on the PC.  What do you see for typical offset between steady state external and internal temps?   Any data as to how much the offset changes with ambient temp?   Is internal temp and brew head water temp pretty much the same (say initially, before incoming cool water has had much effect)?  Anything you have is appreciated.

I work with engine controllers and there is constant debate about accurately modeling systems vs just measuring.  And open vs closed loop.  And eliminating a sensor and deriving the value with some fancy software (when you build a million engines, that $2 part is real money).
back to top
 View Profile Link to this post
daduck748
Senior Member


Joined: 13 Dec 2012
Posts: 157
Location: California, USA
Expertise: Just starting

Espresso: Gaggia Classic (2012)
Grinder: Breville BMF600XL
Posted Thu Aug 8, 2013, 4:31pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

I don't have any qualitative data for you there, but from what I've noticed, a steady state SEEMS to average about a 5F lower temp than internal temp.  I can't even go get you numbers now because now with the pre-heater coil installed, the external sensor body is now sandwiched between two coils and now I'm reading about a 10-15F drop from before the mod.

Simply put, since the sensor is at the top of the water level, it is the closest you can practically get to the top of the down-tube (beginning shower water temp) and the channel to the shower is relatively short, it is very close.  There are factors at play, like as the cool water enters the boiler, it will also cool the bottom end of the down-tube, so as the hotter water at the top run through the tube, is would cool the hotter water coming down the tube some.  So the water is constantly heating and cooling through out the pull sequence, but it hardly takes any time, so the changes in temp are probably negligible.

I think you can only get closer to actual brew temps if the sensor is either at the bottom end of the down-tube or in the shower itself.  The problem with getting a sensor in those locations is that there is no water in those areas most of the time, so the sensor would be measuring ambient air temps (in the tube), or shower head temp; not water temp.  So, a sensor just below the water-line seems optimal.  Please correct me if I've overlooked something.  If you've checked out the performance charts at the beginning of thread, my sensors outperforms some of the common TC's out there, but even with my solid-brass pill-bodies, it still take a fraction of a second to detect actual temps.  It also only takes a few moments for water to run through the water-ways anyway, so can anyone see the difference?  Probably not.  Can you see the difference in a data log?  Sure you can.  It'll just take someone to put some time aside to do it.  Someone probably already has.  Not me.

This is a bit off topic, but since you've mentioned some aspects of engineering, here's my thought on open/closed loop systems.  As an engineer (and heck, as a father as well), everything has its time and place.  When it comes to espresso machines, I think most people here will agree that a closed loop system is the way to go, basically to control and contain temp drift.  If you rely on an open system to maintain water temps, even if your algorithm is SUPER!, I warranty you will eventually see drift.  Before I developed my PIDuino algorithm, I had a tried an open system... one that dripped milliseconds (about 50ms and about a 500ms wait, so what?  about a 10% duty cycle) of power to the heaters.  By trial and error, it held my temps to 212F on the nose and it wouldn't budge!  I had my espresso machine set to 212F way back when.  But after about 3 days (with pulling about 4-5 shots daily), my once-stable 212F temp has drifted about 2-3F.  I hadn't done anything.  It was probably due to a change in atmospheric temps or the neighbor's dog pooping on my lawn of something.  Who knows.  What would happen in the dead of summer when indoor temps could hit 80-85F (that's just how hot my house gets right now... at night).  So, a closed system would work well for maintaining temps.  It doesn't work too well during a pull.

During a pull, you take the PID function off-line and then what?  You have to go with SOME theoretical method.  Perhaps an algorithm where you convert Joules/Time and convert that to watts.  That's cool.  Like I said, that sounds interesting and I'll have to try your 80% one of these days (just lazy to plug my laptop into my espresso machine.  People look at me sideways when I say stuff like that).  But no matter what your method, there is some portion of guesstimation in there, which basically comes from trial and error.  Very scientific, huh?

As far as modeling engines: I'm mostly in the semiconductor industry, working with PHD's all day long and these guys... whoa... over my head most of the time.  Some (not all) of these guys are actual scientists and they work on a very sophisticated modeling software (which remains unnamed for proprietorship).  These simulations take DAYS to complete, running on a matrix system, meaning probably about 32 CPUs crunching numbers at the same time.  In the end, there's always some variable that caused it not to turn out the way it's supposed to.  Once they get run the sim successfully, off to the real world it goes!  These systems live (for the rest of their lives) inside cleanrooms around the world.  Cleanrooms, like you see on TV - clean, white, spotless.  REALITY check!  Cleanrooms are not like that.  I've seen SUPER clean cleanrooms and... not so much cleanrooms.  Supposedly, they are ALL SUPER clean, spotless and controlled (temperature, humidity, etc., the whole nine yards), but the reality is, even every "controlled" cleanroom has variants that cause that very expensive simulation that ran multiple times for x-number of days to fail.  How do you fix it?  Take your current situation and add your guesstimated value.  LOL... again, very scientific... but that's reality.

Granted an additional $2 part multiplied by millions of engines is a HUGE savings.  And spending a few hours (even 80-hours) on software development to emulate what what a physical sensor would do would save millions in the long run.  However, depending on how critical the application, spending the extra $2 for accurate data might be worth it, especially when your returns will pay for that $2 sensor many times over.

I hope that helps.  It's only my perspective so take it for what it's worth.

 
Gaggia Classic
Breville BMF600XL
Ducati Hypersport 748r
back to top
 View Profile Link to this post
jonr
Senior Member


Joined: 25 Jun 2013
Posts: 298
Location: Americas
Expertise: I like coffee
Posted Thu Aug 8, 2013, 8:18pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

> a closed system would work well for maintaining temps.  It doesn't work too well during a pull.

I agree.  It's also a common engine control strategy.  Use closed loop whenever you can, go open loop (or a hybrid mode) when there are too many delays or rapid changes for closed loop to work properly.

I'm waiting for parts so I can measure and log basket temperature in parallel with boiler wall temp.  I'm not expecting data that is usable for closed loop, but I expect that it will be useful for tuning(manually or automatically) open loop values based on previous shots.

20F-30F difference between boiler wall temp and basket temp seems to be a common value.

> It was probably due to a change in atmospheric temps

I agree.  I get a display of % power needed to maintain the boiler temp (while closed loop) and it varies considerably with the surrounding air temp.  3-7% is a typical value.
back to top
 View Profile Link to this post
D4F
Senior Member


Joined: 15 Mar 2012
Posts: 2,021
Location: USA
Expertise: I like coffee

Espresso: Gaggia Classic PID
Grinder: Baratza Forte-AP
Posted Thu Aug 8, 2013, 8:21pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

Wow, you actually read my posts and quoted me :)  Thank you, I will take that as a compliment.  You are starting to sound practical, not just technical.  I particularly enjoyed your comment
"coming from a father."

daduck748 Said:

As an engineer (and heck, as a father as well), everything has its time and place.

Posted August 8, 2013 link

I enjoy seeing your work.

A comment on ambient temperatures.  If the PID or other device can maintain the boiler sensor temperature and the ambients move up or down then the % power or duty cycle should shift to cover that.  I depend on my house thermostat to do that and not have to manually manage all of the time.  Then the microambinet temperature inside the cabinet should not change much.  In the end it is at the puck and then in the cup.  You are not controlling for a computer readout, but your ability to taste.

 
D4F also at
http://www.gaggiausersgroup.com/
back to top
 View Profile Link to this post
jonr
Senior Member


Joined: 25 Jun 2013
Posts: 298
Location: Americas
Expertise: I like coffee
Posted Thu Aug 8, 2013, 8:35pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

> Then the microambinet temperature inside the cabinet should not change much.

I think that how much and how much effect this has on basket temp is an open question.  Think house thermostat mounted on an exterior wall (they don't control indoor temps very accurately as the seasons change).
back to top
 View Profile Link to this post
D4F
Senior Member


Joined: 15 Mar 2012
Posts: 2,021
Location: USA
Expertise: I like coffee

Espresso: Gaggia Classic PID
Grinder: Baratza Forte-AP
Posted Fri Aug 9, 2013, 10:14am
Subject: Re: Custom Thermocouple for Gaggia Classic
 

I forgot to mention an OPV solution.  Remove the OPV and make a plate of say 1/4" T6, stainless, or brass with threaded holes at each end to face off the boiler entry of the OPV.  Sealed with the O ring and then a central "T pipe" to exit the OPV through the O ring threaded, pressed, or CNC made on the part to accommodate your fitting to the copper tube.  The "brother" counterpart of that piece to replace the OPV entry at the boiler, similar to the piece on non-OPV machines.  Flex between pump and OPV and then OPV to Copper and Copper to new boiler fitting with or without flex as you need.  I see how well you make parts so that should be a simple solution for you and another jewel of production.

 
D4F also at
http://www.gaggiausersgroup.com/
back to top
 View Profile Link to this post
daduck748
Senior Member


Joined: 13 Dec 2012
Posts: 157
Location: California, USA
Expertise: Just starting

Espresso: Gaggia Classic (2012)
Grinder: Breville BMF600XL
Posted Wed Sep 18, 2013, 2:57pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

Sorry for the long delay.  You've probably forgotten about this by now.  Work has been keeping me pinned lately and I haven't had too much time to log in... anywhere for that matter.

As far as the OPV, I hear what you're saying and I've designed a new water pump with built-in OPV.  I started making one (without OPV), then realized I ounted the out-port in the wrong direction so I scrapped it and redid the design to implement OPV, right at the output.  Ont he boiler side, I'll be building a block-off plate with only a single 1/8 NPT port for incoming water.  This should fix it.

I'll keep you guys posted when I get a working unit made.

 
Gaggia Classic
Breville BMF600XL
Ducati Hypersport 748r
back to top
 View Profile Link to this post
D4F
Senior Member


Joined: 15 Mar 2012
Posts: 2,021
Location: USA
Expertise: I like coffee

Espresso: Gaggia Classic PID
Grinder: Baratza Forte-AP
Posted Wed Sep 18, 2013, 3:08pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

You are not forgotten :)  Family and work get in the way of play, and that is good.  Picture is worth a thousand words, show your work.

 
D4F also at
http://www.gaggiausersgroup.com/
back to top
 View Profile Link to this post
jonr
Senior Member


Joined: 25 Jun 2013
Posts: 298
Location: Americas
Expertise: I like coffee
Posted Wed Sep 18, 2013, 5:15pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

I'd love to have an OPV that could be adjusted on-the-fly by a microcontroller.  Response time doesn't have to be super fast - say 1 second.
back to top
 View Profile Link to this post
daduck748
Senior Member


Joined: 13 Dec 2012
Posts: 157
Location: California, USA
Expertise: Just starting

Espresso: Gaggia Classic (2012)
Grinder: Breville BMF600XL
Posted Wed Sep 18, 2013, 6:16pm
Subject: Re: Custom Thermocouple for Gaggia Classic
 

D4F Said:

You are not forgotten :)  Family and work get in the way of play, and that is good.  Picture is worth a thousand words, show your work.

Posted September 18, 2013 link

Well, at this point, there's really nothing to see.  First I realized I placed the mount holes in the wrong position (I think I was thinking about it upside down and not right-side up, which made me put the pump 'on the wrong side'), and I hadn't clamped it well and it jump out of the vise as I was machining the corners... so it's a tosser.  It was just still sitting on my desk so here it is.

What about it?  Well, it's a motor with an externally wound inductance core, with a spinning pump rotor.  That's all it really is.  The magic happens in the pump block.  Well, in this picture, it's just 5 holes; 4 for mounting the motor and one in the center for the water inlet.  This inlet is a standard 1/8 NPT thread for a 1/8 NPT tuve quick disconnect.  The tube goes right into the water tank.  

The circle marking on the left side of the block (bottom with the picture) is the position for the pressurized outlet.  This is also "supposed" to be 1/8 NPT to mate with a 1/4" compression fitting so my pre-heater would slide right into it.  That's how the thing was supposed to work, and it would have.  I almost fixed it by rotating the part around until I had a decent orientation to work with... then messed it up.

Well, at least you can see how it was supposed to work and I was REALLY close to getting it done.

The new design obviously makes the cap much larger as I've integrated the OPV into it as well.  There would be a third port for the overflow, dumps right back into the tank, and an adjuster port.  This design would allow you to adjust the OPV on the fly since the adjuster screw is not inline with the overflow.  It just isn't controlled via microcontroller, like jonr would like.  :D

Believe me, I'm a technology guy and I love automation.  Implementing a microcontrolled OPV would make the system much more complex in both hardware and software, for a feature you may use once every six months or once a year.  When I adjusted my OPV on the Gaggia Classic, I set it to about 140psi, probably like most here.  I've checked it about six months after that it is was still around 140psi.  I also asked the same (sort of) question in Barista training and everyone looked at me funny.  The instructor said, "Well, you could adjust the pressure by this screw... down here, behind this panel..."  and of course I asked why the manufacturers made it so hard to access.  Again with the funny look on their faces and said, "Well, it's basically a 'set it and forget it' adjustment".

Now to implement a microcontroller controlled OPV, you'd have to basically design a miniature power screwdriver that can turn the adjustment screw.  You're best using a stepper motor or a servo (toy servos may not have enough torque) so you'd need like a NEMA17 servo or stepper to make this happen.  Of course, to drive the servo or stepper, you'd need a driver, and of course a power supply for the motor.  If using a servo, you need an extra GPIO in PWM mode, or two GPIO for the stepper for Enable and Direction.

Once you have the mechanics down, you probably need a digital pressure gauge (very expensive) or fashion some sort of resistive actuator to sort of get a variable resistance to read via analog input on the Arduino (you're using Arduino, right?).

Then you would need the software to drive all this.  You not only need a PID routines to maintain water temp, but you need a second PID routine to manage the OPV pressure.  You could probably share CPU-time with the main Arduino, but that would start to slow down your main PID.  So, I would guess you'd want a second Arduino to control the OPV to offload the temp-PID.  You'd also need to implement a menu item for the OPV in your menu system.

That's how I would implement a microcontrolled OPV.  It might be a little much for me to do, so I'll leave the implementation to you.  :)

daduck748: IMG_3671.JPG
(Click for larger image)

 
Gaggia Classic
Breville BMF600XL
Ducati Hypersport 748r
back to top
 View Profile Link to this post
showing page 5 of 6 first page | last page previous page | next page
view previous topic | view next topic | view all topics
Discussions > Espresso > Espresso Mods > Custom...  
New Topics updated topics   New Posts new posts   Unanswered Posts new unanswered     Search Discussion Board search   Discussion Board FAQ faq   Signup sign up  
Not Logged in: Log In to Postlog in
Discussions Quick Jump:
Symbols: New Posts= New Posts since your last visit      No New Posts= No New Posts since last visit     Go to most recent post= Newest post
Forum Rules:
No profanity, illegal acts or personal attacks will be tolerated in these discussion boards.
No commercial posting of any nature will be tolerated; only private sales by private individuals, in the "Buy and Sell" forum.
No SEO style postings will be tolerated. SEO related posts will result in immediate ban from CoffeeGeek.
No cross posting allowed - do not post your topic to more than one forum, nor repost a topic to the same forum.
Who Can Read The Forum? Anyone can read posts in these discussion boards.
Who Can Post New Topics? Any registered CoffeeGeek member can post new topics.
Who Can Post Replies? Any registered CoffeeGeek member can post replies.
Can Photos be posted? Anyone can post photos in their new topics or replies.
Who can change or delete posts? Any CoffeeGeek member can edit their own posts. Only moderators can delete posts.
Probationary Period: If you are a new signup for CoffeeGeek, you cannot promote, endorse, criticise or otherwise post an unsolicited endorsement for any company, product or service in your first five postings.
Learn @seattlecoffeegear
Learn all about coffee, watch videos, read how-to articles.
www.seattlecoffeegear.com
Home | Opinions | Consumer Reviews | Guides & How Tos | CoffeeGeek Reviews | Resources | Forums | Contact Us
CoffeeGeek.com, CoffeeGeek, and Coffee Geek, along with all associated content & images are copyright ©2000-2014 by Mark Prince, all rights reserved, unless otherwise indicated. Content, code, and images may not be reused without permission. Usage of this website signifies agreement with our Terms and Conditions. (0.271121025085)
Privacy Policy | Copyright Info | Terms and Conditions | CoffeeGeek Advertisers | RSS | Find us on Google+