Has anyone ever PID'd a lever machine? - Page 3

[cannonfodder]

Here is the scoop. I filled my boiler to the top of the sight glass so I would have plenty of water. Powered it up and let it heat. After it hit the top of the first heating cycle, I purged about 8 seconds of steam to get rid of the false pressure, then left it alone for 10 min.

During that time, I used the very scientific measurement process of holding a type K thermocouple against the side of the boiler (about two inches from the base). At the end of 10 min I had a steady surface temp of 217. The boiler was cycling from about .7 to .85.

I quickly put the probe in the opening of the single spout (it is preheated and stabilized at this point) killed the power and raised the lever. After about 6 oz of trickling water the temp had hit 211F. It remained there through another 5 oz of water for a total discharge of 11oz. I held the TC, suspending in the spout so I would not get a false reading from the PF spout metal. Obviously, 211 is way to hot for espresso (it is probably several degrees hotter at the shower screen), hence the burnt shots and my desire to lower the boiler temp.

If I surf the temp in the .5 bar range the shots improve. If I pull a shot within moments of the first boiler heating cycle it is not too bad either, but far from a gosh shot. I still get bitterness. The same bean ground in the same grinder pulls a much better shot in my E61 machine. I get lots of crema and pull at the proper pressure and get extraction times that are pretty much on the mark with the lever, it is just to hot.

I believe a PID may be in my future, any suggestions?

[cpl593h]

Do you have a variac that you could try first?

[cannonfodder]

I had thought about a variable voltage adjustment device, but this machine is in my office at work. A big old variac sitting on my work table would not look to good. Being the head IT guy, my office is full of flat panels and computers, sleek black cases and high tech gizmos. A big old variac would clash, bad, but a digital display PID controller would not even be noticed.

I also have to watch what I do. I had to get permission from the corporate office risk management just to have the machine in my office (exposed pressurized boiler presents a burn and explosion hazard ). Down side, PID is expensive.

[another_jim]

Down side, PID is expensive.

Have I got a deal for you. Love controls makes a "digital temperature switch," i.e. the Brewtus style controller, for $65. These include 16 amp contact outputs, so no ssr required.

Since the boiler is so decoupled from the final temperature, a PID seems like overkill in any case.

Here's one of many models on sale for fishtanks (the water FAQ was useful in a number of ways)
http://www.automatedaquariums.com/lv_ts13010.htm

Edit:

They're available with TC input for $45 from the maker (Dwyer)

http://www.dwyer-inst.com/htdocs/temperature/SeriesTCSPrice.cfm

[j7on]

Whoa!

How can it be possible with virtually same machines(as i recall you had the Gaggia Pavoni clone) to get so different results?

Even though my grouphead is way colder than yours i experimented and actually benefitted from the wet-towel cooling, just wrapped a cold wet towel around the grouphead like 30seconds before i pulled the shot - huge difference!

These ARE hot machines and obviously we both(all) have issues with heat.


I think there could be some degrees knocked of the grouphead by adding a cork&rubber gasket between the boiler & grouphead?
That way the boiler would not be in direct contact with the grouphead and therefore not transferring as much heat.

I will give it a try a soon as i can.

[cpl593h]

I think there could be some degrees knocked of the grouphead by adding a cork&rubber gasket between the boiler & grouphead?
That way the boiler would not be in direct contact with the grouphead and therefore not transferring as much heat.

I will give it a try a soon as i can.

A couple of weeks ago I was thinking the same thing! I never got around to it because I decided to upgrade to a Cremina. I'm definitely interesting in the results you get. It makes sense - this mostly decouples the group from the actual boiler, reducing the heat transferred from the boiler to the group.

[cannonfodder]

While that would give a small amount of boiler/grouphead isolation, IMHO, the problem is not the groupheads contact with the boiler, but the fact the machine essentially runs in the steam mode, not brew mode. Regardless of how well the GH is insulated/isolated, the water feeding it is still 217F. the only way to change that is to reduce the boiler pressure. If you were an espresso purest, and never steamed milk, a quick Pstat adjustment and the issue is solved.

I believe the thermocouple temp switch Jim points out would fit me need perfectly. I have a couple of Omega Lock 1/8 rigid probe fittings in a bag. I can tap and mount that under the boiler in the base to hide it. Then mount the control unit in a nice stainless box and put it beside the machine.

I could set the temp at 205 (assuming that would provide enough boiler pressure to operate and a couple degrees of temp loss through the GH) and off I go. Then kick it up to 250 for steaming, and then back down. I wonder if I could leave it on all day, just set the temp at 190 to keep it warm. Then a quick heat to brew temp and back down again to standby.

A quick question to Jim. I see they have the TC input units, and then under them, there is a configuration key unit. Do I need both parts ($49.99x2)?

[another_jim]

A quick question to Jim. I see they have the TC input units, and then under them, there is a configuration key unit. Do I need both parts ($49.99x2)?

As far as I can read it, without the key, you cannot change from the type you ordered to another (say K to J). With the key you can. It sounds to me like the key is for plants where lots are used, and they want to store spares usable all over.

See the other thread I posted on this to let everyone know about it (I think this gadget is too bang for the buck to hide). Turns out Barry is using a few for odd and end jobs, so you can shoot him an email for nitty-gritty questions. He likes the unit, even as low cost display, but cautions it's not quite 1/32 din but an odd size. Sean says Dwyer is a good ompany, and he's the real uber-controls maven among us. I have one on order, and will use it on the Peppi and report soon.

[cannonfodder]

I just put in my order.

[cannonfodder]

My temp controller arrived yesterday so I spent the day running around picking up parts today.

I removed the base off the boiler, drilled/taped and mounted my K TC. I have the unit bench rigged on the machine right now and running. I think I am going to like this.

The unit is wired with the NC (normally closed) side of the relay for heating. I can change my boiler temp at will which is very nice. I can set the controller for 170F and leave it there all day, then a quick bump up to brew temp, another quick bump to steam then back down to standby. There is a 1 degree deadband. I set it for 200f, it heats to 201 and shuts off, and there is some carry over after the element shuts down but you can adjust for that. It cools to 199, click, heat again, 201, click, and it is off.

I can tell all of you that 205F does not provide enough pressure to operate the machine. There is just enough pressure to make the gauge twitch and the steam wand gurgle. 225 gives me .5 bar and sufficient pressure to work the machine. Before the controller, I would try to temp surf with the power switch to hold .5 bar, and boy was it a hassle.

I have to mount the controller in a project box, I have already cut the box for the unit. I have the TC wired and the element wired. I need to wire the controller to the power switch on the machine, this is very important and I will go into that later. I am going to drill a hole in the rear center of the base, put in a grommet, run my wires in some split loom tubing into the back of the project box. I am thinking about attaching the controller box to the base of the machine for ease of transport. That way I do not have a controller flopping in the breeze when I move the machine around.

I did take some photos and would gladly post them, but I have exceeded my bandwidth allotment from photobucket for the month. With all my HB postings and blog traffic it went quick this month.

I will probably do some more work on it tomorrow but Thanksgiving cooking is going to take most of my time. Have a nice Thanksgiving everyone.

[akin]

Hi David,

all you need to do now is to add a bike pump

Akin

[j7on]

Do you have a variac that you could try first?

Would this really work?

I have been thinking about PID'ing my machine too(after i realized that the thing controlling the pavoni is a pressurestat not a thermostat, otherwise it would have been easy to drill a new thermostat pot on the side of the thing...)

My wall-outlet is inside my kitchen cabinet anyway so it would not look horrible, then if i was to drill in a probe-thermometer inside the boiler i could monitor the temperature too.

The PID project i am sure we all enjoyed following was great, but it's just that big box on the side...not good for me..

By prying out the pavoni emblem from the front and installing the thermometer there it would look custom made and with just a variac unseen inside a cabinet would be "hassle-free" + variacs are cheap.



But would it be possible to adjust the p-stat with a variac or does the machine just stop working when voltage drops?

[cannonfodder]

Here is the problem, these types of machines require at least .5 bar pressure to operate, that is around 228F water temp, much higher than the accepted norm of 205 for brew water. Even after I added the digital controller my water temp at the grouphead is still to hot. It is better and I do notice a difference, but was it worth all of the work, still undecided.

I let my machine preheat for about a half hour at 190F, then I ramped the temp up to 230F (brew temp), took a type K thermocouple and held it suspended in the opening of my single spout PF. I raised the lever and watched the temp. After about 8oz of water it stabilized at 213F. I have to assume that if I was pulling a normal shot the temp at the shower screen would be lower than that due to group head temperature, but after slowly streaming out 8oz of water it was hot.

Now for the sake of fun, I let it cool, repeated the process but ran the temp up to steam pressure, 242F (about 1 bar) and took the water temp using the same method. That is about where the machine ran at using the OEM pressurestat. Guess what the water temp was, 213F again.

So in the end, the water I got from the machine at the lower setting, and the water that would have been dispensed at the normal pressure level appear to be the same. I have to emphasize, appear. While I do have access to a couple thermocouples, I have no desire to spend hundreds of dollars on testing equipment. I have become somewhat less enchanted with all of the tech tricks and gizmos over the past month and gone back to concentrating on what is most important. My barista skills and what ends up in the cup.

So take it with a grain of salt. Can you improve on the stability of these styles of machines, yes, but is it worth all of the work, undecided. You also loose some of the old world classic design by putting a digital controller or big old variac on the counter beside the machine.

[j7on]

I also thought of adding a manual pump inlet between the pressuregauge(so i can still have function of it) and that glasstube(where you see the waterlevel).

Then use the variac + temp probe to get water temperature right, add the needed pressure to get water into grouphead with the pump = perfetto!?



I am just sceptical of the variac, anybody even have a clue if you can use it to control the p-stat?(=controlling water temp in boiler)

[cpl593h]

I am just sceptical of the variac, anybody even have a clue if you can use it to control the p-stat?(=controlling water temp in boiler)

My only experience with the Variac is on a Europiccola two-switch, which has no pressurestat. I didn't take that into account when recommending it.

[Borys Ta Polon]

Hi fellow lever people!
First of all, Let me start by stating for the record that I love my Astoria lever machine. However, I have been entertaining the thought of replacing my pressure stat with a PID controller. I have already won an ebay auction for a cold fusion unit last night.

Some days I can achieve god-shots all day long. Other days a new storm front rolls in and the espresso is a bit sour. At its best a pressure stat achieves +or- 3 degrees. This equals a variation of 6 degrees. To further complicate matters barometric pressure will affect the performance of the pressurestat. I am hoping the +or- .1 variation of the PID will let me achieve god-shots more consistently.

Any ideas or info?

Thanks - Borys Ta Polon

[HB]

Borys Ta Polon,

Hope you don't mind, I merged your post with this related thread.

[cannonfodder]

The thermostat I installed was a project of diminished returns. While I can play and adjust the temperature on the fly, it honestly did not yield that much of an improvement in the cup. It is just the nature of this type of machine. A PID in a gravity fed lever would probably yield better results. Because the machine requires a certain amount of pressure to even function (about .5 bar minimum) your boiler water is over the optimal brew temperature to start with. I believe .5 bar is around 228F on my machine. The group then acts as a heat synch to pull heat out of the boiler water to cool it to brew temperatures. That is why you only get about three shots out of one of these style machines before it overheats and starts burning your shots.

One of the reasons I wanted to add a thermostat was so I could leave the machine turned on all day at work in 'hot standby'. I can set the temperature to 180F and leave it on all day. A quick bump up to 230F pull a shot and drop back down. I can also bump the pressure up more than factory for steaming.

With your PID, you will probably have to add a SSR (Solid State Relay) unless you have one with an integrated relay. The thermostat I used has an internal relay capable of handling my boiler amperage. You will have to drill into the boiler and tap a fitting. Then install a compression fitting to run your thermocouple probe through into the boiler. Then it is a simple matter of disconnecting the pressure stat, hooking those two wires to the relay, relay to PID, PID to thermocouple probe and power.

When I set up my machine, I left the inner workings as-is, so if it flopped, or I grew tired of the setup, I could revert back to the factory configuration by simply unplugging two wires and reattaching the Pstat, which I may do one day.

Here are a couple of my blog posts showing the process and my final thoughts..
Digital Temperature Controller on a Lever Espresso Machine

Caveat To Lever machine Temperature Controller

[Linz]

As I understand physics (and I do understand physics pretty well), there is a direct relationship between pressure and temperature. At any given pressure, the temperature at boiling will be constant and will never rise above that temperature. That's why many manufacturers use a pressurestat instead of a thermostat - the control of temperature using a pressurestat is much more accurate. It's also why PID controllers don't do much for lever machines (at least those ones that run the boiler at pressure). You need (as has already been mentioned) at least 0.5 bar (and preferably 0.8 bar) to force water into the group.

At 0.8 bar (typical tank pressure) water boils at about 115 degrees C (I think that's about 240 F). Which is HOT. The machine relies on the pressure drop into the group head, and the cooling effect of the head itself, to drop the water to the correct brewing temperature.

I've also tried measuring the brew water temperature without coffee in the basket, and I get about 100C (212F) but I can't believe this is a correct reading. I think that with coffee in the basket, the group head under the piston fills with water, and the ten or so seconds of preinfusion contact with the cooler metal will drop the temperature to a good brewing temp. Then you pull the shot.

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.

Wikipedia has a temperature/pressure graph http://en.wikipedia.org/wiki/Image:Water_vapor_pressure_graph.jpg. To use it, you have to convert pressure in bars to torr by multiplying by 750. Also remember that 0.8 bar on a gauge is 1.8 bar total because we already have approximately 1 bar of atmospheric pressure.
0.8 bar = 1350 Torr.

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.

[timo888]

... 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. ...

(emphasis mine)

Welcome aboard, Linz.

The temperature of the water leaving the boiler drops to 212F when it hits atmospheric pressure. But it is not yet "brew water" until it passes through the coffee; until then, it is premature to speak of "brew temperature". I am not just quibbling: when the boiler water temperature is lowered, the water flowing out of the boiler and into the conduit (before it reaches atmospheric pressure) imparts less heat to the metal, placing less demand upon the group as a heat sink. Also, a lower setting would cause the heating element to come on less often and for shorter durations, which would also result in a lower group temperature, because less heat is conducted out to a directly attached group or convected out to a thermosyphon group. When the cylinder fills with water, the group is better able to draw more heat out of the water during the preinfusion. At this point we can say, changing the temperature of the boiler water will have an effect on the brew temperature.

Regards
Timo