Posted Sun May 27, 2012, 8:09am Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
Thanks ... that would be interesting to use a dual PID ... especially if one were to add a second boiler for preheat - maybe have a preheat boiler set at 195F and the main boiler at 215F.
=====
I've had several different ideas - so it is probably pretty confusing trying to keep them all straight.
One idea was to have the machine idling at a lower temp (the temp that the boiler ends up at the end of a perfect shot) and then something to kick it up to the temp required to start another shot without the long wait.
I found (on my very old Gaggia with the larger boiler) this works out to be about 210F at the end of the shot or idling -- then I kick the heater for 10 seconds or so and then stop the heater -- over the next few seconds I watch the boiler temp climb and as it hits 217-218 I start my shot and the boiler continues to climb to 222F and then starts to drop. At the end of the shot, the boiler is back to 210F and I hit the steam switch again for about 10 seconds and start the next shot as the boiler temp crosses 218 again.
That works for that larger boiler because I don't need to heat the boiler during the shot - the temp doesn't drop (as much) because the boiler is larger. =======
The other idea was sort of the opposite of that - and might not work at all. I was thinking of just setting the PID to a much hotter temp to idle at 225F or so --- it would be too hot and if you tried to pull a shot from that idling temp you would get a flash of steam and a bitter brew. But my thought was to always flush for a set count (2-3 seconds - you'd have to experiment) so the shot starts at the right temp but the boiler metal is already hot to counteract the incoming water on the smaller boilers. And my hope was that the PID would be cycling the heat on and off trying to hold the boiler temp at 225 during the shot (this would also help counteract the cool water being pumped into the boiler during the pull).
On the Gaggia Classic, with the smaller boiler than on my older Gaggia, I have to keep cycling the heat during the shot or else I get a pretty large drop off in temp by the end of the shot. I might start a shot at 200 and by the end of the shot it's down to 192 if I don't do anything to counteract it.
========
If I understood what D4F is doing - I believe when he starts his shot, his alarm function turns on the heat right as the pull starts ... I believe it turns the heat on full power (not cycling the way the PID function does) and then shuts it off at his alarm set point - and then the boiler overshoots that temp as is normal with these boilers. So my prediction is that if I put my thermofilter on that machine and watched the temps I would see the shot start at 200F and immediately start to fall (even though the heater was turned on at the start of the pull, it takes about 10 seconds for that heat to show up) so 10 seconds into the shot I would expect to see brew temps around 194 and then start to climb rapidly and by the end of the shot be way too hot. That's just a guess based on my understanding of how D4F's alarm function was described and my experience manually controlling the boiler - but I may have totally misunderstood D4F's setup. My apologies if I got it all wrong.
What I would try instead would be to disconnect the alarm function and simply set the PID 10-15 degrees (F) too hot - but do a flush before the shot starts to get the brew temp where you want it to start - and then let the PID try to get the temp back up to the too hot temp (it should start doing this as you are flushing so the heater is activated before you start the shot) and then start your shot. There would be a lot of experimenting to find the PID set temp and the length of time to flush and then the length of time to wait after the flush before you start the shot. But by juggling those three things, I think I could find a combination that would give you the correct starting temp and would hold the temp pretty close for the duration of the shot.
The whole intent of that would be to find a routine where you can do back to back shots with proper starting temps and less temp change during the shot.
Posted Sun May 27, 2012, 10:02am Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
From what I've gathered from this thread, the use of the full-on heating is to provide a pulse of heat input to the boiler so that it reduces the time lag of the boiler getting back up to temperature. I like the idea of this because it's exactly what a modern control system would do.
Using the alarm circuit or a second PID channel might not make the most sense. The cool thing about a relay (or an SSR I believe) is that you can use multiple different circuits to drive the "coil" of the relay to "close the contacts" of the relay. An SSR allows for faster switching which is why it works for the PID. So, it's possible to create a controller just for the pulse that can be used in conjunction with the PID.
If one can figure out how much heat energy is required to heat a boiler's worth of water from its entering temperature to the desired temperature AND can figure out how long it takes for that heat to transfer from the Gaggia heating elements into the water then you can use it to create a heat pulse that will reduce the lag of the water coming up to proper temperature.
Posted Sun May 27, 2012, 12:33pm Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
D4F had the math worked out early on in this thread (or was that another thread?) and it matched pretty closely what I got from experiment.
But I think the way to do this would be to use a computer to monitor everything and work out all the variables - you'd want to monitor the temp of the water inside the boiler - and you'd need a way to monitor the temp of the water coming out of the group and probably need to measure the room temp and the incoming water temp too - and the temp of the heaters plus the temp of the metal casing of the boiler - and the rate of flow of incoming water (and this would change at the beginning of the shot as the coffee puck gets soaked until the pressure is built up - at least I suspect the flow would be a moving target until that point). Then the software could adjust the power to the heaters while monitoring all those changing temps and flow rates. Once you got the algorithms all worked out it could be simplified into a small controller or a glorified PID with the anticipation and time lag worked out so you tell the PID to start the shot and it waits until all the temps are ready to start the shot ... and maybe signals you a countdown timer to start of the shot in case you want to wait to prepare your basket until just in time to lock/load and start the pour.
I am able to do this manually - but it is really tricky to get right and all I have to do to screw it up is leave the heater on for a couple of seconds too long - or wait a couple of seconds too long before I start the shot -- if I'm off by just a little I could easily see 215F to the coffee puck - it really changes quickly with those powerful heaters and tiny boiler in the Gaggia.
So I have to wonder ... by the time you got all that worked out, it'd probably be cheaper and easier to just go HX and have the bonus of not waiting to steam milk. When I see that you can buy a new N.S. Oscar for about $1000 I start to wonder why I'm fussing about trying to get an SBDU to hold steady temp (because it's really fun and challenging I guess).
D4F Senior Member Joined: 15 Mar 2012 Posts: 1,222 Location: USA Expertise: I like coffee
Espresso: Gaggia Classic PID Grinder: Preciso
Posted Sun May 27, 2012, 1:56pm Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
A couple of interesting posts. The math showing enough calories of heat in starts on page 1. The use of a single SSR with either function is in the SkeneDesign diagram also page 1. The PID function primarily controls the SSR to stabilize the boiler. The alarm on and off are set but not energized; that circuit has no electricity until I flip the brew switch. I actually can hear the alarm circuit in the PID make a soft click as the machine passes those limits, such as when using steam. The brew switch turns on the Gaggia normally, and the same switch energizes an old cell phone charger for 6v DC and that goes to the PID alarm circuit to control high and low setting or alarm limits and then overrides the PID function and control the SSR. The alarm circuit drive the heater until the sensor hit the set set and then off until it falls below the low limit and then back on... Similar to the PID circuit but not damped. The PID circuit is too slow and damped to help, so I do not think a second PID would help.
Now about second shots. If you have several PFs loaded, or baskets, and can pull the second shot in 15 seconds, my system may not work. But wait, there is more :) If I decide to use a 3 second brew hit to pre-infuse, then wait a few seconds and brew, then I have stretched the time of heating and absorption of heat. Picking a high alarm set temp will cause more % of time of "on" to arrive at the higher set, so temp controls time to a degree. I have no need for 5 -6 doubles, but if I pulled them quickly, the on time could easily approximate the manual method and should be able to give similar result. Also, a long continuous pull should work about the same with either method as both cycle on and off multiple times. With the PID/alarm method, a Polder thermometer placed similar to on the manual method averages the temperatures and does not oscillate, If over 30 seconds I get the same amount of on and off then the average boiler, and therefore water, temperatures should be similar.
I wonder how the methods are similar and different using say a double shot pull, a set time to ready shot two and then a second pull. What happens when the shot time is 1 - 2 minutes or a reasonable time to ready a second shot with use of the same PF and basket. The 25 - 30 seconds of pull and heater on and off then have time to absorb and technique has to account for that.
I am trying to make a simple electronic fix for the Gaggia that may take some time to learn and adjust, but so did pulling a decent shot. If one were to hook it all up to computers and data logging, then the whole process could be learned more quickly.
To keep it simple, I wonder if a pre-infusion system, not available from Auber for the generic PID could activate the alarm/heater function which would start with the pre-infusion and continue to cycle through the rest phase and brew. More time to add and absorb heat.
If I started again, I would like a 1/16 PID such as SYL-2362 with PID boiler set temperature, 1 alarm for brew as I do, and then I would "alarm" the steam temp to and not have to try to keep it at about 280F by flipping the switch. Of course, that assumes that I am understanding correctly that it can have SSR(PID) and 2 alarm outputs. The 1/32 only can have SSR(PID) and 1 alarm. Better yet if it could come with pre-infusion, one box and system to do it all.
D4F Senior Member Joined: 15 Mar 2012 Posts: 1,222 Location: USA Expertise: I like coffee
Espresso: Gaggia Classic PID Grinder: Preciso
Posted Tue Jun 12, 2012, 10:50am Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
Andy, I have again viewed the videos with the Classic and large boiler model. I am always trying to improve and stabilize the Classic with the PID and Alarm function of the PID. I finally noticed a starting temperature and stable temperature of 235F. I assume that you just raised it to that temp before the video, or you would have steam. Obviously that is the outer boiler temperature and not the water, but if held for long without incoming cool water, then the boiler water would steam.
I would like to see a video with temperatures of you pulling a shot, then a second shot after say about 1 minute or your desired time. I am trying to get an idea from the Polder thermometer where the boiler temperature needs to be and the time sequence needed to be ready for a second shot. Pulling almost continuous shots leads to a near steady state of heat and incoming water. That is less complicated than on and off with a minute to dump and load, what many of us do when we do not have multiple PF's and baskets.
Even without the PF temperature measure, a similarly placed Poulter and similar temperatures on a similar Classic, should give similar brew temperatures. That would let others with manual temperature adjustment, or PID alarm adjustment, come close to duplicating your results without the equipment (PF temperature).
Posted Tue Jun 12, 2012, 4:02pm Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
Remember I have my TC mounted in a different place than you do. Yes 235F sounds crazy hot but I do not get steam spurting from the group at that temp - it actually does put out about 200-203F at the brew head at that temp on the boiler. But it does this when it has been less than 2-3 minutes since it flushed 1-2oz shot. If I let it sit at 235F for 10 minutes, yes it would be very much too hot.
Doing the temp test with a styrofoam cup is one thing - you get a more instant read than the ThermoFilter (the thermofilter has a lot of mass that needs to get up to heat over the first 7 seconds of the simulated shot). But the Styrofoam cup test doesn't give you as accurate a read of the temp change over the course of a shot since you get 1-2oz in a styrofoam cup in just a few seconds.
So ... the video you are asking for ... you want a "simulated shot" using the thermofilter? Or you mean you want a video of me pulling a real shot with a bottomless PF? (I have no way to show you the real temps at the brewhead if I do that). And do you want it from the Classic?
If I do the Classic ... it would look similar to the video I posted of the Carezza I have for sale. In the Carezza video I am using the steam switch during the shot ... if I were doing it on the Classic I would be flipping my toggle which is hidden away on the back of the unit so you wouldn't see what I'm doing. But the Carezza video was shot less than 2 minutes after pulling shots ... the temps are different (lower) on the Carezza boiler than the Classic to get the same shot temps - I suspect because the Carezza has no metal to the cabinet and heats up really fast and doesn't radiate any warmth from the boiler to the cabinet. But in the Carezza video, I'm pulling huge shot volume because I don't have the pressure profiler on the pump to slow it down and my shot simulator in the THermoFilter flows too fast (about 3oz or more in 25 seconds) so I'm using more of the heater during the shot.
Anyway ... let me know exactly what you want (Classic?, Me cycling heater? or just letting it fall naturally? 2oz in 25 seconds? etc. ... and then pull a second shot after say 45 seconds recovery? And you want to see the boiler temps in the video? or the thermofilter temps? or both?) and I'll make a video of it.
Re: the temp of the boiler metal vs temp of the water ... If I were to steam some milk and then refill the boiler, the boiler temp would be reading about 245-250F and I would need to wait for it to cool down to about 212F before I could get below 208F water at the brew head. ... But if I had just pulled a shot one minute ago and the boiler was now reading 220F I would probably only see 189F at the brew head. So there is a huge difference between cooling down from a totally hot boiler and heating up from a cold boiler and from sitting idling steady at a set temp. But if I know I want to pull a second shot right away ... I'd try to let the boiler cycle up to 230F and watch it fall back down to 225 and then heat it back up to 230 again and it would fall a little slower back to 225 then maybe one more cycle from 230 and fall to 225 and then probably be able to hit the heat for 4 seconds and start a shot and get 200F at the brew head (and all of that would take about 45 seconds) and cycling the heat through the shot several times would hold it pretty steady for a 2oz shot.
Posted Tue Jun 12, 2012, 6:10pm Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
Something else to think about ... I have always wondered why in the world Gaggia made the upper half of the boiler shaped the way they did ... nearly square with three of the walls being flat and one wall all rounded. There may be something to the shape and to where the water enters the boiler that causes the water to swirl in some particular pattern - or maybe not - I just don't know.
Anyway ... when you are trying to get your brain wrapped around my method of starting with a hotter boiler and cooler water inside ... think about the water coming in at the bottom of the boiler and gradually working its way up to the top of the boiler where it is then able to enter the stand pipe and head to the brew head.
So you always have hotter water at the top of the boiler and cooler water coming in at the bottom.
For me, it makes more sense to have a very hot boiler but to have the water inside a bit cooler so that it will just barely be getting hot enough as it gets to the top of the stand pipe with the heaters actively working during the pull (part time) keeping the metal of the boiler hot enough to warm up the incoming water just the right amount by the time it swirls and gets to the top of the boiler. Anyway ... that is how it seems to be working for me with what I measure in the thermofilter and the way I imagine what is happening inside the boiler.
If you imagine the same scenario with a boiler all stabilized at 215F ... you somehow have to get the water inside that is already at the perfect temp to go into the brewhead by pumping cold water into the boiler without having the shot temp plummet --- it just sounds impossible because you would have to start out at 215F and then very quickly get the metal up to 235F at the same rate as the water coming in is cooling the water inside. So you are dealing with rapidly increasing metal temp at the exact same time as rapidly decreasing liquid temps inside and trying to get the timing and curves to exactly balance each other out.
But in my method ... you start out with the water already cooled a bit and the metal already heated up enough (at about the same as it will be during the shot) ... so I am not trying to quickly change the temp of two opposing thermal masses ... I'm trying to hold the metal temp pretty steady while the water is entering at a pretty steady rate of flow and it is just so much easier to start in the ballpark and then try and hold it in equalibrium rather than trying to take two thermal inertias moving rapidly in opposite directions and make them strike each other exactly centered at the exact right moment to balance. (I'm not sure if "thermal inertia" is a proper term ... but I think of it that way)
D4F Senior Member Joined: 15 Mar 2012 Posts: 1,222 Location: USA Expertise: I like coffee
Espresso: Gaggia Classic PID Grinder: Preciso
Posted Tue Jun 12, 2012, 9:49pm Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
Yes, I realize that your Polder and my PID sensor are in different places. I may mount a Polder thermometer similarly to try to simulate what you are doing.
I figured that the 235F was only for a short time. We may differ there as the PID will give me a stable temperature to start, different than the bimetal. I would start with a boiler of about 215F stabilized for about 20 minutes with PF in place. I then load the PF, tamp, and lock. Not much cooling time. I pull a shot and then when done reload, say 60 seconds – 120 seconds, and then pull again. During the 1 minute, or 2 minutes, the machine cools as would yours if you were busy and not flipping the switch at that time, then pull again. I am interested on what happens and how you get shot temperature for the second shot. Is there enough retained heat to start the shot, or too much. Do you hit the steam switch before the second pull? I am trying to find a way to get a second pull with a 1-2 minute prep interval between. I would like to see the boiler and thermocouple temperatures and know the method that you use to get them there. You may need to vary the routine as I am trying to get out of a steady input and output. I would like to see a way that the average user, basically one PF, can use your method, or a PID and alarm, to get a second pull and keep the temperature up, as well as blunting the intrashot fall.
What temperature does your machine set at prior to the first rise to 235F? I am up to 215F baseline. I may have to adjust the temperature up to 235F with the steam switch to the pull, or some determined seconds of heat input. I am hoping that if I can duplicate the boiler temperatures with a similarly placed Polder that I can get a similar in the cup result. I am learning by watching the Polder. The question remains whether the PID alarm will go on and off at a desired rate to keep the Polder temperature similar to yours. Once I see what happens as you do pulls similar to what I do, then I can try to match boiler temperatures hoping that the cup temperatures will match. Hopefully if I can get similar Polder temperatures, you may have enough interest to try a PID and I can share setup and you can monitor thermocouple temperatures. I can easily have enough heat; I could set the alarm temperature to 300F and keep in on during the pull. Obviously too much heat, but noting that the PID alarm temperature indirectly controls heat cycles.
Totally aware that the technique is different as you warm up versus cooling down from steaming.
I am aware of boiler shape and the central tube. Brew water comes from the top and may not mix much up to a certain volume, but as we know, there is intrashot cooling unless you help the Gaggia. That is why I am trying to use the alarm to function to add heat, and you do manually. Since the alarm is a higher set, and not controlled by P, I, and D functions, it is full on until the temperature is reached and then cycles on and off. I am not opposed to hitting the steam switch preemptively and starting the heating before the entry of cool water. Cool water does not start my function anyway, the higher set temperature turns it on with the flip of the brew switch.
I am aware of some of the steady states that you use, cool water in and heat in. That will get less steady when you rest between shots and that is what I am trying to find out. How, if any, will you need to adjust?
Posted Wed Jun 13, 2012, 12:42am Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
D4F Said:
I am interested on what happens and how you get shot temperature for the second shot. Is there enough retained heat to start the shot, or too much? Do you hit the steam switch before the second pull?
OK ... have just pulled a shot - or flushed 1oz or so into my cup - my boiler temp (as read on the Poulder thermometer I have mounted) is down somewhere around 212F and falling due to the flush and if I leave it alone, the T-Stat is going to cycle the heater on (which will take it on a rollercoaster ride around 240F). So I will hit the steam switch for about 10 seconds and off --- that will have the boiler coasting up to about 230F and then coasting back down fairly rapidly ... about the time it has coasted back down to 225F I will hit the steam switch again for maybe 3-4 seconds and it will coast back up to 230F again. Now it coasts much slower on the way down and about the time it is hitting 225F I will hit the heater switch (this time my added switch - not the steam switch - because I plan to leave the heat on for 2 seconds after the shot starts and the steam switch prevents the 3-way from opening) so I hit the heat for 4 seconds while I drop my prepared basket in the PF and lock it in place and then start the shot a the end of the 4 seconds of heater and leave the heater on for another 2 seconds as the shot is starting then flip the heat off for 3-4 seconds. Meanwhile the shot has started to pour and the boiler has not really coasted up much (because cold water was being pumped in and counteracted the heat I put in. At this point it really depends on if I'm pulling a 20-30ml shot or a 60ml shot for how often I will cycle the heat on during the shot -- and also whether I want the heat to really stay up there to the very end or I may want the temp to drop towards the end (lever machines do so maybe it is desirable?) ... but I usually pull slow flowing 30ml shots and probably hit the steam for 1 second and then off for 4 seconds and then back on. If I'm pulling a faster flowing shot it might be on for 2 seconds and off for 3.
The video you saw ... with the really high heat ... that was pulling shots (testing pulling simulated shots through the Thermofilter) so there was very little time for the boiler to recover (I may have even had the boiler up close to 250F at times the way I was pumping cold water in with little recovery time) ... so I think that will be misleading as I never pull my actual espresso shots quite like that. I am much more likely to try and settle my temp in around 225F over the course of a minute or two -- and I either let the "Gaggia T-Stat roller coaster ride" do it's thing (swinging between 200F and 240F about) and catch it somewhere in the middle of that (say around 212-214F for this example) and kick the heat a little to try and settle it in around 225F.
But the way I used to do it before I had the Themometer on the boiler --- I would have let it do its cycle and watch the light and the instant the ready light came on (heater just turned off) I would start the pull. And that is the same routine many people without PIDs use to temp surf their Gaggia Classic. Now with the thermometer on there I can see that the ready light comes on about the time the thermometer is reading 225F and if I don't start a shot right then the thermometer continues to climb to about 240F - but if I do start the shot just as the ready light comes on the boiler temp doesn't climb much due to the cool water coming in. Anyway ... it's fairly similar to what I'm doing - trying to start the shot around 225F with the heater having been on for a few seconds.
I'll try to make you another video and see how much time I need to leave between shots to get two shots with steady 200F through the entire pull.
D4F Senior Member Joined: 15 Mar 2012 Posts: 1,222 Location: USA Expertise: I like coffee
Espresso: Gaggia Classic PID Grinder: Preciso
Posted Wed Jun 13, 2012, 9:39am Subject: Re: Gaggia SBDU Preheat PID Temperature Recovery and Stability
I think that I follow your sequence, but a picture (video) is worth a few thousnd words of description.
If I give a pre-brew hit of the steam switch, the PID alarm function then does about what you describe. I will have to mount the Polder to see if I can get similar temperatures. I need to mount if in a totally removable fashion. My new machine is only a few months old, still guaranteed and I am trying to have every mod removable. So, where is the thermistor bead on the boiler?
I think that from your description you understand what I am trying to find. Use of added heat, with intermittent pauses of 1 - 2 minutes for reloading, and keeping brew temperatures up intrashot, and the boiler ready for the next shot. You are most likely to be able to do it, but it will be interesting to see a sequence that does all that. It stops all of the steady processes, but that should just cause a change in sequence, and amount of "on and off" time to accomodate.
You may find in the process how long that you can go without added heat and still be ready for the next pull, or if about 1 minute between pulls that you need X seconds of pre-pull heat and if 2 minutes of time, then Y seconds of pre-pull heat.
I like your work and it is great for we GaggiaGeeks. Though there few other posts, people seem to view/read.
I can't help but wonder if non-Gaggia SBDU readers are doing some of this also. I like to think that the Gaggia high wattage and small boiler is an advantage to some of what we are doing, but much of the technique of holding temperature is similar across SBDU brands.
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 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.
= No New Posts since last visit 