www.olympia-express.ch: espresso, the chemistry of love

PID thoughts

Postby mitch236 on Sun Aug 08, 2010 5:47 pm

Ok, I'm sure this has been discussed before but I couldn't find it. I just installed a PID device into my La Marzocco Linea single group and I realized something. The PID is great at maintaining brew temps in the boiler but there are two main problems. One is by the time the water gets to the PF some heat has been lost. I need to rent (or buy) a Scace device to figure out how much but there is another problem. The second problem is the water that is replacing the brew water is cold. How cold is a function of the ambient temp but it is at least 100f less than what you want. So I came up with a solution. I am going to find a flexible heat tolerant tubing and wrap my steam boiler and put it in line with my water feed so that the water entering my machine is preheated. I have no idea how much heat I can transfer but that steam boiler is mighty hot!!

I'm sure this has been thought of before so I'm curious about how it turned out for others who have tried.
User avatar
mitch236
 
Posts: 627
Joined: Jul 21, 2010
Location: Florida

Postby malachi on Sun Aug 08, 2010 6:13 pm

1 - LM has a saturated group. Depending on your specific Linea (different generations have had slightly different designs in terms of grouphead) the offset is slightly different but the saturated design means the offset is always reasonably small. You will, of course, need to ID what the offset is. In general, you'll find that (once in steady use post idle) an est 2s flush tends to be sufficient to manage.

2 - intake water preheat like you describe has been part of the Kees machines for a while - and Schomer modified his machines in the same manner. I believe both used copper for this purpose. The challenge, of course, is that when not in active use - water will still be within the copper tubing and will as a result now be hotter than brew temp. I've talked to both Terry Z and Kees in the past about calculating the optimal volume and thus the amount of "copper coil" you should use - but honestly don't remember much about how it's done. The alternative, of course, is to use a heat exchanger set-up in the steam boiler for preheat. Now... the reality is that intake water preheating is unlikely to be really important for home use. It's intended to address issues with very busy cafe use where frequent autofilling results in boiler and brew temp dropping outside of range. While understanding offset between boiler temp and brew temp is, IMHO, vital for home (as well as cafe) use -- I think intake preheating is (at best) a tiny optimization for home use.
"Taste is the only morality." -- John Ruskin
malachi
 
Posts: 2593
Joined: May 05, 2005
Location: sfca

Postby mitch236 on Sun Aug 08, 2010 6:23 pm

I can see where a busy cafe would have much more severe issues but since installing the PID, I notice a significant drop in temps just after one 1.5oz shot. Maybe about 5-7f. It seems temp consistency is of upmost importance, it would seem a 5f variance is alot. Perhaps I'm wrong. But I can't imagine wrapping the steam boiler would account for much more than a 3-4 second purge before pulling a shot but maybe I'm wrong there too. So much to learn!!!
User avatar
mitch236
 
Posts: 627
Joined: Jul 21, 2010
Location: Florida

Postby Peppersass on Sun Aug 08, 2010 6:51 pm

malachi wrote:The alternative, of course, is to use a heat exchanger set-up in the steam boiler for preheat.

That's how the GS/3 does it -- the intake water runs through a tube in the steam boiler before it gets to the brew boiler.
Dick Green
User avatar
Peppersass
 
Posts: 661
Joined: Jul 20, 2009
Location: New Hampshire

Postby HB on Sun Aug 08, 2010 8:50 pm

mitch236 wrote:...since installing the PID, I notice a significant drop in temps just after one 1.5oz shot. Maybe about 5-7f.

I haven't seen the La Marzocco PID kit, however I assume probe placement is such that incoming cool water will be detected rapidly; that's a good thing. If you place a thermocouple over the lip of the basket and across the puck (or use a thermofilter), you'll find the actual brew temperature in the grouphead changes very little. If you don't want to shell out for a thermofilter, you can build your own, rent one, or simply lay 36 gauge teflon-wrapped thermocouple wire across the puck.
Dan Kehn
User avatar
HB
 
Posts: 12672
Joined: Apr 29, 2005
Location: Cary, NC

Postby miKe mcKoffee on Sun Aug 08, 2010 9:07 pm

mitch236 wrote:I can see where a busy cafe would have much more severe issues but since installing the PID, I notice a significant drop in temps just after one 1.5oz shot. Maybe about 5-7f. It seems temp consistency is of upmost importance, it would seem a 5f variance is alot. Perhaps I'm wrong. But I can't imagine wrapping the steam boiler would account for much more than a 3-4 second purge before pulling a shot but maybe I'm wrong there too. So much to learn!!!

Don't know whether you have an older EE or AV Linea but both have the issue of low temp from idle, AV much more severe than EE. It's not actually the group that's under temp but the external plumbing the water flows through leaving then re-entering the group during a shot. (Hence the newer Piero groups which eliminate this issue) The AV will yield more under temp from idle shot because of the volumetric flowmeter mass which must be brought up to temp. I can only imagine the brew boiler temperature dive on a 1 group flushing must cause. On a 3 group Linea with it's large 5L brew boiler doesn't phase it (hardly) at all. But on a 4 group with it's two 3.4L brew boilers (one for each pair of two groups) it's a real issue. (have 1 3g, 2 4g's) So yeah can easily see a 1g's 1.8L brew boiler temp would majorly dive. In production pulling shots on the 4g's you have to alternate group boilers or wait for temp to come back up after pulling the shot and post shot group flush. Which is a PIA when often want the two brew boilers at different temps! I too have thought about pre-heating the incoming water via tube wrapping the steam boiler on my 4g's but haven't found/made the time yet. Modding with Piero groups would be more elegant, but at about $600 per group (including labor to further mod since mine are older style bolt on groups and Piero mod designed for bolt ons) for my 2 4AVs that's close to $5k I just don't have.

From idle flushing with pin holed back flush blank to minimize brew boiler temp dive does work better than full flow flushing.
Mike McGinness, Head Bean (Owner/Roast Master)
http://www.CompassCoffeeRoasting.com
miKe mcKoffee
 
Posts: 1356
Joined: Jun 03, 2005
Location: Vancouver, WA, USA

Postby Arpi on Sun Aug 08, 2010 9:54 pm

mitch236 wrote:So I came up with a solution. I am going to find a flexible heat tolerant tubing and wrap my steam boiler and put it in line with my water feed so that the water entering my machine is preheated. I have no idea how much heat I can transfer but that steam boiler is mighty hot!!

I'm sure this has been thought of before so I'm curious about how it turned out for others who have tried.


That is also what the expobar Brewtus (aka expobar minore) uses circa 2004

http://www.home-barista.com/expobar-brewtus-review.html

"The Expobar Brewtus takes the current prosumer dual boiler machine design one step further by introducing an important element: It pre-heats the water via the steam boiler's heat exchanger before it enters the brew boiler, as shown in the diagram to the right. This feature is essential to Brewtus' jump in temperature stability and control.

The pre-heated water has another side benefit: It is practically impossible to outrun the machine. A typical prosumer HX will give up the ghost after 8-9 shots in a row if done in rapid succession. Most machines need time to recover. Incoming cold water from the water tank reduces the water temperature in the boiler, which affects the boiler's ability to quickly recover. Pre-heating the water before it enters the brew boiler tackles that problem. The Brewtus is capable of delivering 30 shots in 30 minutes without blinking, thanks to the routing of the incoming water from the pump to a heat exchanger passing through the steam boiler. The near brew temperature water then continues to the brew boiler.

Brewtus' innovation isn't really new since a similar feature is available in top commercial dual boiler machines like the Synesso Cyncra and the La Marzocco GB5, and has been a common commercial modification for more than a decade. Expobar's innovation was the application of an existing proven technology and implementing it cleverly to fit a specific need."

Image

Cheers
User avatar
Arpi
 
Posts: 865
Joined: Jan 25, 2009
Location: Baltimore

Postby Rostik_KIEV on Sun Aug 29, 2010 5:00 pm

Friends!
In drawing water circulates counter-clockwise (during brew). Why not on the contrary?
Why water arrives in group from a thermosiphon, instead of on the other hand?
User avatar
Rostik_KIEV
 
Posts: 69
Joined: Dec 30, 2009
Location: Kyiv, Ukraine

Postby cafeIKE on Sun Aug 29, 2010 5:59 pm

The drawing is idle state.
When brewing, water flows from the top and bottom of the boiler. That's why there is an offset from the boiler set point to brew temperature. The water in the bottom of the boiler is much cooler than the set point, which is normally read at near the brew boiler top.
User avatar
cafeIKE
 
Posts: 2905
Joined: Jun 27, 2006
Location: Woodland Hills, CA

Postby mitch236 on Sun Aug 29, 2010 6:23 pm

cafeIKE wrote:The drawing is idle state.
When brewing, water flows from the top and bottom of the boiler. That's why there is an offset from the boiler set point to brew temperature. The water in the bottom of the boiler is much cooler than the set point, which is normally read at near the brew boiler top.


I always thought the drop in temps between measured boiler temps and measured pf (Scace) temps was because of heat loss running through exposed plumbing. That's why the Piero cap mod results in less loss because it reduces the length of exposed travel.

On my PID probe, it appears to measure the bottom of the boiler.

The next experiment I am going to try is to install a hot water heater after the pump and directly before the boiler. That's not the most cost efficient way but I can wire the unit to only turn on when the espresso machine is on. I'm not sure if this will affect the taste of my shots but it is an interesting experiment in creating the ultimate machine.
User avatar
mitch236
 
Posts: 627
Joined: Jul 21, 2010
Location: Florida

Next

Return to Tips and Techniques