Has anyone ever PID'd a lever machine? - Page 7

A haven dedicated to manual espresso machine aficionados.

#61: Post by happytamper »

Linz wrote:But here's the point: When the water comes out into the group it will drop to normal atmospheric pressure, and therefore the temperature must also drop to 100C (212F). The excess temperature will be used to convert a small amount of the water to steam. Changing the temperature of the boiler water will have little or no effect on the brew temperature.

I reckon Espresso physics are about as well understood as women, and I'm not sure which is more temperamental, my lady or my Pavoni.
Hi Linz,

Super observation. Never looked at it like that. Water cannot possible go through the puck higher than 100c unless the puck is airtight which I imagine it is not. But as Timo mentions, the higher the pressure in the tank the sooner the grouphead is overheated and the less the water will be cooled to a good brew temp when extracting. A simple solution to this overheating problem is to rub a cold towel on the grouphead if it seems too hot. I only have to do this when I am making espresso for a few friends consecutively. Experience will tell you when this is needed.

As far as the water temperature is concerned on a lever, it is a difficult thing to control. But it is easy to stay within a good range and with experience this range becomes smaller.

At home, I just switched from my Europiccola back to the Brass beauty (La pavoni Ambassador excellently modded by an alchemist, our Alchemist, for brew pressure reading). I had switched to my Europiccola for a while because I enjoyed that it had no gauges and needed a break from constantly adjusting my arm to a 9 bar pull, quite a feat, thought I might get bursitis. Well now that I have switched back to the gauged-out brass beauty, on the first pull, I see that I am automatically close to 9 bar. Must have built up my strength. The espresso is fullbodied (2/3 panama 1/3 Sumatra roasted on my caffe Rosto a woody black just before the oils).

Back to the temp observation, With a bit of experience, it becomes less an issue and compensating for it will become second nature. Fortunately even when the temp is off the espresso is significantly better than most. Especially with freshly roasted, properly ground beans. Absolutely as important as the temp of the water.

As far as women or people are concerned I can never compare them to any object or to each other, nor in any way exert some kind of control. Just let people do their thing and when the gods are aligned miracles will occur. Physics applies only to the physical, A minute amount of what is really happening in our world.

User avatar

#62: Post by Linz »

happyTamper - happy anniversary, I see you've been here a year. I haven't made it to 1 full day yet! I've cooled my group down by sucking cold water up into it (I posted that in another thread about grouphead cooling). I couldn't agree more that, as you imply, the stuff that is finally consumed can be divine even when the temp (or something else) is a bit off. My very first pull off the LP was pretty average for a Pavoni, but for me it was one of the best ever. Now I get a best ever most weeks. Five years ago I would never have believed coffe could taste so divine.

timo - Yep, point taken. I guess my statement should have been something like Changing the temperature of the boiler water will have little or no effect on the temperature of water entering the grouphead. I see that changing the boiler temp will ultimately affect the brew temperature because with a hotter boiler more energy is being fed into the entire system, and it has to go somewhere. On the other hand, I still think PID control is out of place on a lever machine because the boiler cannot go above the temperature dictated by the pressure. Maybe a PID using the pressure as an input would work, but trying to control a pressurised boiler by temperature is very difficult because the relationship between energy input and temperature output is non-linear and that will confuse the PID loop.

As an aside, I do wonder how much heat transfer there is through the metal where the grouphead is bolted to the boiler, and whether something like a thickish teflon gasket between the two would reduce the conduction. If that worked I guess it would take longer to warm up although that could be countered by pulling some water through first. Just a thought.
LMWDP #105

User avatar

#63: Post by timo888 »

Linz wrote:As an aside, I do wonder how much heat transfer there is through the metal where the grouphead is bolted to the boiler, and whether something like a thickish teflon gasket between the two would reduce the conduction. If that worked I guess it would take longer to warm up although that could be countered by pulling some water through first. Just a thought.
When a Cremina 67 with a directly attached grouphead is left on for a long time, the grouphead overheats and the group has to be cooled with a wet dishrag. An insulating gasket (one that held up well under wet heat) might reduce or eliminate the overheating that results from leaving the machine on. I think you're right about the warming flush.


User avatar

#64: Post by matadero210 »

Hi everyone,

You've inspired me to PID my Pavoni. Here's how I think it should be done: use tap water pressure.

In detail: tap water comes at 1-4 bar, which is more than enough to prime the lever in the Pavoni. As part of my homebrewing (beer) apparatus, I have replaced the aerator tip of my kitchen faucet with a quick release that connects to a garden hose/hose barb. The hose goes to my charcoal filter, needed to remove chlorine. The output of the filter could then go through a check valve into the Pavoni boiler cap (drilled for hose barb). Viola--pressurized, filtered water. Next I would install a mini PID controller in the base to drive the heater and set the boiler at 95C. Lastly, wrap or insulate the group head so that it sits at the boiler temperature when water goes through it. One could add a second PID to electrically heat the group head independently of the boiler. In this case, no blank shot would be required to heat the head, and the head could provide very stable temperature to the brewing water (since the boiler will slew a bit as fresh water is taken in during the pre-infusion. We all know that there is enough metal in the head to slew the brewing water temperature many degrees, so even if the boiler were at 200 (because of 1-2 oz of cold water from the last shot), a heated/PID group would correct the temperature easily.

When the lever is raised, the line pressure forces water into the piston chamber and cold water is admitted into the boiler. The boiler temperature will fluctuate a bit. But the thermal mass of the heated head will correct small fluctuations. I do worry if the line pressure is 3-4 bar (some municipalities require 60psi) that pre-infusion might be a bit violent. Maybe a pinhole or thin tube should be used between the filter and the boiler cap to slow the flow a bit. The check valve would be useful when the boiler PID is set to 220F for steaming.

Lastly, I think a pressure meter would be nice to install in the group head to monitor the 9 bar pressure there. (I'm sure this has been done by many)

What do y'all think--its a good way to go? I think I could verify the ideas by doing the plumbing first, trying it out without the pid, and then going forward.


User avatar

#65: Post by Linz »

matadero210 wrote:You've inspired me to PID my Pavoni. Here's how I think it should be done: use tap water pressure.
<snip detail>
What do y'all think--its a good way to go? I think I could verify the ideas by doing the plumbing first, trying it out without the pid, and then going forward.
Personally I think you're on to a winner. Of course frothing becomes a problem but for making espresso the idea is excellent. There is such a thing as a pressure-reducing valve that you could use on your water inlet, thereby keeping the pressure constant. I think this is important, as it determines the time it takes for preinfusion since the pressure in the brew head while the handle is up is equal to the boiler pressure. Water at 4bar is a bit keen for preinfusion.

You could control temperature roughly using the switch if you want to try the idea without fitting the PID.

Getting the right brew temperature for the water could be a challenge. If the boiler is at the correct temp then the brew head must cool it down since it will never get to the same heat as the boiler, so your boiler may have to be a few degrees above optimum. The brew head might also take a while to get up to the right temp, but once it is you should be able to pull shots forever.

I also think your shots will be bigger. I find it hard to get more than about 40ml (~1.3oz) out of mine, but with your system there will be far less steam in the group head since the water isn't boiling, and less steam would mean more water (I think).

Keep us posted, I think this will be interesting.
LMWDP #105

User avatar

#66: Post by matadero210 »

Thanks for the encouragement, Linz.

Last night I decided to try putting a solid heater onto the group head itself. There is really no good, flat place to put it, so I thought about using heating tape. Then I would wrap the outside of the entire machine with some insulation. Thus, once the machine is warm, everything will be heated to the same point. The problem is that it would make the lovely look crappy, so I'll keep looking for a good place to mount a heater on the head.

My first, cheap thought was to put the TC on the outside at the boiler/head neck and have both boiler and head heaters connected in parallel. Now I'm wondering about using TWO PIDs (Two is better than one!), and controlling the head temperature independently. I ordered two PIDs from auberins dot com for < $90 total. At that price, I might get a third and fourth for my next project .....

After looking at the cap, I'm thinking about removing the over-pressure spout and mounting the TC and water inlet there rather than messing up the cap. Another approach would be to machine a 1cm extension flange between the boiler and head. That joint already comes apart and presents a nice opportunity to insert a flange with water inlet and 1 or 2 thermo-couples. Best of all, I might not need to permanently mess up the lovely. Any design must be careful to inlet the cold water as far from the thermal action as possible. Since the head has a 12cm tube for drawing water, the flange might be a decent place for drawing water. I'm not sure if there's a good place to mount the boiler TC under the machine (and it might be too close to the heater coils), but that's another option to think about.

Linz, thank's for the idea about the pressure-reducing valve. Are they available with any precision?

1. TC for boiler / water temp. Locate inside boiler from flange using epoxy-filled stainless tube. OR locate outside boiler with thermal grease.
2. TC for group / metal temp (on outside). Locate outside head (dry).
3. PID both heaters in boiler in parallel. Use existing switch to override for steam.
4. Use new, 1cm flange between boiler and head to mount boiler TC and water inlet. OR mount inlet and TC through lid or steam-release valve.
5. Water inlet with pressure reducing valve plumbed through charcoal filter to sink faucet adaptor. Locate in lid, steam-release valve, or in new flange.

comments or ideas welcome!


User avatar

#67: Post by matadero210 »

I have the parts coming. I spoke with a talented machinist (hi Lars-Erik!) who gave me the key idea to make this work. Use a pipe "T", a reducer, and a compression fitting to insert a thermocouple probe (tube) into the plenum with the incoming water. This T would then be teflon-taped into a threaded hole in the boiler: one tapped hole for incoming water and thermometer. I've bought the parts and will try this on my Europiccola. He also suggested doing the same with a cartridge heater/TC unit in a 60ml boiler (12" length of 1" copper pipe). This 'double' sized boiler would heat from tap water to 95C in about 1 minute, and then dispense into the group head. The heater would be adjacent to the group head and would mostly preheat it. I would like to try this second project by using a Pavoni group head. Anyone have one sans boiler?


ps: [added on 13 Apr 07] Done. see Positive pressure, PID Pavoni: world domination begins

User avatar

#68: Post by RayJohns »

I was just reading through this thread, since I'm also working on a PID upgrade for my La Pavoni. I wanted to comment on the temperature drop between the boiler and the group head, when pulling a shot vs. pulling a blank shot.

If you pull a blank shot through the group head, the temperature of the water which exits remains relatively constant (i.e. the same as the temperature of the water in the boiler itself). However, if you introduce ground coffee (in a portafilter basket) and then pull a shot, the temperature in the resulting espresso will be noticeably lower than the original water temperature in the boiler. Why?

If you think about what is actually occurring inside the group head, I think you can begin to get an idea as to why you see more of a temperature change. Work is required to help extract the espresso from the coffee grinds and this means energy is consumed (in the form of heat). Also, the coffee grinds themselves soak up heat (i.e. they go in cool and come out hot).

In addition, all this consumption of heat & energy is occurring in a highly compressed environment. In the boiler, the pressure is 1 bar approximately. However, inside the group head (during extraction), the pressure [of any air/steam] is more like 8 or 9 bar. As such, any heat that is consumed inside the group head (while it's under the pressure of extraction), may tend to have a large net impact on the final uncompressed temperature. When the water exits and everything expands again, there may be a bit of an auto-refrigeration effect taking place (similar to inside of your refrigerator, where you compress a gas, cool it and then allow it to expand again). While it may not be as great as what is seen in refrigeration, I suspect it is nevertheless still having some impact here. That's my story and I'm sticking to it :-)

One thing is for sure though: the coffee grinds in the basket go in at room temperature and - after extraction - come out very hot :) This alone probably draws enough heat & energy out of the system to manifest itself as a temperature drop between the water in the boiler and the water in the final cup of espresso.