Posted Thu Jan 10, 2013, 1:44pm Subject: Next Mod: PID my Gaggia Classic
Now that I've set my OPV to 140psi (~9.5bar) and a Silvia V3 steam wand. Now to tackle something more meaningful, for the coffee. However, the OPV was an improvement on overall taste, but that's not really a mod.
PID doesn't seem to be anything special, but I have some questions and concerns before I put time and money into this project.
I measured the water temperature at the portafilter with a Fluke thermometer the other night. At first, I thought the water temperature was too hot, since it started readings at roughly 200F. The longer I held the probe in the stream, the lower the temperature dropped. It dropped to about 125F and I called it quits, since 125F is way too low for a good shot. I let it recover for a few seconds and tried again. Same results on the second and third try.
I suppose the logic is the boiler only holds 3.5oz (?) and when the pump is running, it introduces cold water into the boiler, reducing the temperature. At 3.5oz, it wouldn't be too difficult to persuade high temperature water to get cold.
Granted, this is a free-flow test and more cold water would influence the readings more than an actual brew test, where the water would not flow as freely and could hold the temperature better. How well does a PID stabilize the water temperature during a pull as opposed to maintaining water temp?
This leads me to an idea a fellow Geek gave me. Would it help maintain the water temp better at the brew-head if there is a preheater? The delta from say 100F water enter the boiler would be less than from a 60F source.
To get really good temperature stability, would a PID be enough or would you also need a preheated water source? Again, this is relative only to the Gaggia series since they have small boilers. I'm sure the Silvias for example have better stability out of the box.
I review much of what had been done and the thread is fairly inclusive for the direction that you are heading. Since you are recovering from the flu and will want to get plenty of rest, you might find it helpful, sleep inducing.
You flow about 50 ml for a brew in 25 seconds and that slow water entry helps delay cold water mixing. A high flow test is not very meaningful, with perhaps the exception of about 30 - 45 ml for a styrofoam cup test. That will give you a ballpark only.
The first article discusses preheat at the beginning and why it is not needed at least the way I did it. I did not want to risk more leaks unless the temperature controller could not be made to do the job. The "PID" function will not do what you are trying as it is slow and damped, the alarm will.
If you have questions and interest, then please revive that thread with your question and the information can stay in one place.
Posted Thu Jan 10, 2013, 6:29pm Subject: Re: Next Mod: PID my Gaggia Classic
My first thought on the therafilter was to use a sintered-metal material. Raw sintered beads would work best, so that they can be compressed and fused into a basket with some heat. Depending on how much the beads are packed and compressed, this could simulate an actual puck. After being packed and compressed, you would have to heat the material so they become pliable and bond together. Probably the best way to do this is by inductance heating.
Being metal, it would not be affected by the high temperatures. But with the sintered material being metal, the filter would have to be preheated before you could run a test with viable results. This is probably not a huge undertaking since I, probably like most of you, keep my PF in the group head for preheating anyway.
Posted Fri Jan 11, 2013, 12:56am Subject: Re: Next Mod: PID my Gaggia Classic
Back on topic, I just ordered a RTD sensor from Auber Instruments. It has a M4-0.7 thread which is a direct replacement for the stock thermostat. This will eliminate any permanent changes to the boiler. Shipping is EXPENSIVE!
I've decided to go the Arduino route, since I'm already familiar with programming these microcontrollers. This will also allow me to adjust the PID behavior as I identify them. I will also be building a data logger so I can log boiler temp/drift, brew on/off, recovery cycle. I'm going to keep it simple for now just to get it going and not drag it out too much. I'll fine tune it as I go. The data logger will be useful in other parts of my shop, so it will be time well spent.
I also decided to go the Arduino route because I want to add other features in the future, such as TDS control, shot-time, and maybe even energy-save mode and maintenance reminders, which would mean adding a clock.
I also have a handful of illuminated tactile pushbutton switches on order, which will replace the stock buttons, and adding a couple more functions. Functions planned: - On/Off (complete power shutdown, except arduino subcircuit) - Standby (enable energy-saving mode, probably set to 1 hour idle time and power on somewhere around 6am-ish) - Steam - Manual Pull - Programmed Pull
The beauty of both worlds of semi-automatic and lazy-automatic. Of course, these may change, but here's hoping.
Besides, most if not all readily available PID, TDS and Timer controllers are too bulky, especially when all three are added together. I already have several Arduino units and 20x4 I2C LCD displays which I plan on installing fairly flat on the face of my Classic. The electronics will be mounted in a project box mounted on the rear.
Posted Fri Jan 11, 2013, 10:37am Subject: Re: Next Mod: PID my Gaggia Classic
I've done all sorts of things with Arduinos. But a couple relevant things I haven't had the pleasure of experience with are RTD sensors and TDS sensors. So, this will be a learning experience for me as well. As long as we're all having fun, yeah?
From what I can tell, RTD sensors will be measured at roughly 1.5mV at 200F. To get any kind of precision readings, we have to multiply that to get the 1.5mV closer to 5V, where the Arduino wants to be. Hopefully, that can be fixed with a gain amplifier. I once was the type who made everything from scratch. Maybe I'm older and lazier now, I only make 50% of my projects now :D, so I've found a 100x amplifier on Sparkfun for about $15 or so (I think). 100x is still not enough gain, but we can adjust the circuit to bring to where we need it to be.
TDS sensors are the same, however, from what I understand and an oversimplified statement, a TDS sensor is nothing more than 2 wires at a set distance, connected to an opamp (basically an amplifier) that is driven by an oscillator (to generate an AC signal). Then the output of the opamp will be given as a voltage. So, the more TDS in the water, the higher the conductivity, the higher the output voltage, and the code within the Arduino will scale that and display it on the LCD. Anyone who can elaborate, please do. I would like to make sure I get this circuit and algorithm correct. It's important. There's GOOD coffee at stake!
I have 2 handheld TDS monitors from Amazon ($15) on order. 1 for calibration, and 1 to tear apart, to see if there's anything I can use. A commercial TDS sensor is roughly $60... and the TDS controller roughly $160-200.
A little off topic, one fun thing I've done with Arduinos, and still working on it, is retrofitting my 100W CO2 laser to also drive my 80A plasma cutter. It was crazy because the first thing to do was to find the laser trigger, and when I found it, the signal was inverted. As far as I know, no SSR take inverted signals, so that's where the Arduino came in. I had to read a toggle switch to determine if we are in laser mode or plasma mode. If laser mode, I had to carry a raw trigger value as well as an inverted trigger value. The raw trigger was a pass-through to the power supply. The inverted value + time delay (500mS, so the signal was stretched) to drive an air solenoid for the air assist. Without the stretch, the SSR would kill the solenoid in less than a week! It wouldn't be able to keep up with the laser either. This is an added feature that the tool didn't come with. I always had to turn on the air manually and it was a pain when I walked away from a job. The air would still be on after the job had finished.
In plasma mode, the raw trigger would just get inverted to drive a SSR that triggered the plasma cutter. Air is controlled by the plasma unit so no special programming was required there. Everything works GREAT! But when I hooked it up... I got HUGE EMF noise! It was strong enough to reset the laser controller and reset the machine every time the trigger fired. I've since shielded all the plasma cables, grounded everything and I've eliminated MOST of the noise, but not all yet. Once that's squared, it was a SWEET Arduino project. It took about a week to program and debug. Hopefully, this one won't take too long either.
Of course, I had to build a random trigger simulator to test the circuit off the tool.
tracerbullet Senior Member Joined: 13 Feb 2012 Posts: 168 Location: Saint Paul Expertise: I love coffee
Posted Fri Jan 11, 2013, 12:51pm Subject: Re: Next Mod: PID my Gaggia Classic
You mentioned pre-heating the water, and it can definitely be done. Something I did, and many others have as well, is wrap copper tubing around the boiler to obtain some pre-heating of that water before it actually goes into the boiler. I have a post about it here on CG, but I hesitate to link to it because if I did it again I'd do it differently.
However, there are OPV's that attach to the outputs of the pumps directly, instead of the boiler itself. From there you can adapt to a copper tube that's wrapped around the boiler, and then into a fitting into the boiler where the former OPV was located. Some of the reviews I've read made temperature comparisons before & after and showed improvement.
Might be something to consider, and search around on if your'e bored.
Posted Fri Jan 11, 2013, 1:33pm Subject: Re: Next Mod: PID my Gaggia Classic
Thanks Tracerbullet. I have seen your post, thanks to D4F, very good writeup and photos. Aden mentioned it to me and reported his implementation worked well, so it's something I'm considering. Maybe I missed it, but how is the water within the tubing getting heated? Is it through physical contact with the boiler or through radiant heat from the boiler? The first time I heard of it, I thought the copper tubing was used as an inductance heater! That would suck SERIOUS power and would probably melt the boiler itself. I guess I was wrong.
After putting what was taught in barista school aside, "must be pulled at 198'ishF", it seems for the most part, people are heating the water to the upper limit and ending the shot at the lower limit, so the resulting average shot temp is somewhat in the mid-range of the brew temp. Would that be correct... or almost correct?
I'm taking all this in, but I'm going to dumb it down and start with just an Arduino-based PID to start with. Then with my data logger, I will compare preheat and non-preheat reports and go with what gives more stable results.
In any form of barista training, I'm sure you're always spoiled with commercial equipment. When you're released into the world, and all you can really afford to practice on is a consumer or prosumer machine which you need to hack to get to commercial standards. Life is funny.
Posted Fri Jan 11, 2013, 5:40pm Subject: Re: Next Mod: PID my Gaggia Classic
Just as I suspected. I have a few Chinese K-type thermocouples (why does Chinese always come up?) laying around and directly hooked it up to my multimeter. I lost my nice BK Precision meter, so I had to get a cheap $20 Homedepot stand-in until I can afford a new BK or Fluke. :(
What did I find? 185F water, straight from my commercial BUNN tea brewer (and I have measured this temp with my Fluke before at 185F) shows 1.8mV on the meter. It actually stabilized at 1.9mV just after I took the picture. According to Omron (I think), they have a chart that states 1.5mV at 200F. This just means there needs be a calibration value within the program. This could also be an error offset from a $20 multimeter, also. I'll have to get my Fluke Thermo and stick it in the water to verify water temp, maybe starting at boiling water and logging manually until the data logger is done.
Symbols: = New Posts since your last visit = No New Posts since last visit = 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.