This is a real stab in the dark: I am looking for Help! I want to install a small size PID to fit to a Sama Club lever machine. I want it to be a specialist PID just for this machine... I am doing this purely for fun...

the main difference to most PID's I have looked at is- I want a PID with two probes- one connected to an SSR to control my boiler (inline with a P-stat) and the other is to read (only) temperature at the group head.

I plan to drill a small hole in the neck- actually to drill out the screw that is already there. The group is an Elektra group but has been modified for a thermosyphon connection. The neck seems to be the ideal place to measure the group temp as it is the point of the thermosyphon connection in these type of group heads- and the screw hole is already there asking for a probe...

I am hoping the PID can display the boiler temperature and the Group temp simultaneously (I think most PIDS display the set temp- and the actual temp- I am hoping they can be programmed to take on different functions?). It will be interesting to watch the group heat as the thermosyphon kicks in- and then the rapid changes when flushing and making a shot).

I need it to run on 220 volts. I need the boiler probe to have a tip that will be immersed in the water- the tip for the group head will be touching brass only- not water.

If there is someone who understand the requirements- has a soldering iron- and would like a challenge: I would be happy to pay/barter/trade for a customized PID for my Sama. I actually need to completely re-wire the entire machine. I plan to add a lamp to illuminate the water level indicator. I have already stripped and hammer-toned the cabinet and chassis- and completely restored the boiler. The PF has been stripped naked...

The machine itself is almost fully restored- the group is tested and ready: I am just waiting on a few parts

(anyone have a surplus P-stat- or PF 220v element?, or rocker switch to fit a hole: 20mm x 35mm??)

I have a PID system on my Silvia that is as you describe. The system I built consists of four assemblies:

• Main board with AVR microcontroller and 4-channel thermocouple sensing subsystem (approx 3" x 2" x 1/2"). I am using two channels at present, boiler and grouphead.
• LCD display assembly (16 x 2 character LCD, I2C interface PCB, 4 pushbuttons, 3 LED's). Size is approx. 3" x 2" x 1".
• Garden variety, zero cross switching 25A SSR ("hockey puck" form factor, approx. 2.5" x 1.5" x 1")
• Switch mode AC to DC converter (i.e., 9V wall wart PSU)

If you can fit those components in/around your machine and don't need me to build enclosures and the like, or source the thermocouples, then I could make a duplicate at a pretty small cost. No UL, CE, or other listings

Alternatively, I have some 1/8 DIN Watlow 988 controllers gathering dust (not really, they are safely stored) that have two input channels. These are the size of a small brick (55mm wide x 100mm tall x 120mm deep), but can do what you want.

Jim

sorrentinacoffee wrote:the main difference to most PID's I have looked at is- I want a PID with two probes- one connected to an SSR to control my boiler (inline with a P-stat) and the other is to read (only) temperature at the group head.

If you plan to have a PID controller involved, why do you want to continue to use a P-stat?

On the PID controller, if you are unable to find one which monitors (and displays) two control loops, then you can always wire up a switch that will momentarily switch out the sensor on the boiler and switch in the sensor on the group head. I had thought about doing that on my PID La Pavoni, but didn't want to have the additional wiring or components involved (in order to keep the installation cleaner).

This would allow you to switch over and monitor the temperature of the group head during extraction.

Ray

Note that adding a switch to a thermocouple wire will skew the readings.
Thermocouples are about the junctions; if you add a regular electrical switch, you'll have additional junctions at the switch.

Thanks everyone:

Jim G: your set up sounds about perfect for my project- and your little blurb about your interests seems to indicate you are exactly the guy I was looking for! If I am going to install the pid into the cabinet of the machine- then I only have limited space. There is a good deal of room under the chassis- I was thinking that the SSR would go there. I was looking at the standard box shaped PIDS- the smallest I could find were around 45mmx45mm at the face- and 75mm deep. This is about the perfect size and there is room for it to go into the front panel of the machine where the power switch cutout is. Only negative is that it is right behind the steam wand... I am assuming your proposed set up is quite different? It may be even better as the LED display does not sound very deep... I would liek to keep that part as small as possible as it will be mounted beside the boiler- and getting close to the P-stat. When you say: "Main board with AVR microcontroller and 4-channel thermocouple sensing subsystem (approx 3" x 2" x 1/2") I am using two channels at present, boiler and grouphead."- - this sounds like the brain? could this part go under the chassis as well? Is it possible you could send a picture? There is a lot of room there- the only limitation is the height- which is a bit over an inch.



RayJohns: following on from Allon's comment at the end of your Pavoni thread I am thinking of keeping the P-stat as a fail safe- having it inline with the SSR but set to say 1.4/1.5 BAR. I would never allow the PID temp setting to reach those levels so the P-stat would just sit there in case of a PID failure or similar. Make sense?

I believe the safest way to PID a previously pressurestat controlled machine is to place the SSR in series with the pressurestat and set the pressurestat higher than you would nominally operate the machine, but still in the safe range. Then, if the PID program goes nuts or the SSR fails on, the machine reverts to pressure control, albeit at a higher temperature than your normal operation. There would still be the additional overtemp safety and overpressure safety.

(allon)

The Sama machine does not have an overtemp cut off (unless there is one built into the element??)- or a low water level cut off.... I am thinking the probe in the boiler would get hot enough if the water ran out (for whatever reason) to trigger the SSR and turn off the element? The old element I have is worn but still working- I can get another from PV- but I wonder if a different type of element would be more suitable?

allon wrote:Note that adding a switch to a thermocouple wire will skew the readings.
Thermocouples are about the junctions; if you add a regular electrical switch, you'll have additional junctions at the switch.

True. However, a TC junction has to be made of two pretty dissimilar alloys. Is adding a switch (or even the junctions between the TC wires and the PID controller itself) going to have a huge impact relative to the actual TC junction?

Ray

sorrentinacoffee wrote:RayJohns: following on from Allon's comment at the end of your Pavoni thread I am thinking of keeping the P-stat as a fail safe- having it inline with the SSR but set to say 1.4/1.5 BAR. I would never allow the PID temp setting to reach those levels so the P-stat would just sit there in case of a PID failure or similar. Make sense?

Your idea of including the P-stat isn't a bad one. However, if the machine already has a pressure relief valve, then I think the P-stat might be somewhat redundant.

Myself, I think there is something to be said for keeping everything as simple as possible. The fewer things involved that can fail, the better. Of course, it's your project, so you are the one who has to decide which direction to go.

With my setup, I tried to keep things as simple and straight forward as possible. I want the machine to work well, be simple and last as long as possible. I was also dealing with rather limited space.

As you may know (from reading my original thread), I opted to leave out the over-heating fuse that normally is attached to the bottom of the boiler (to protect the machine in the event of running out of water). Including it would have involved some ugly wiring and increased the potential for wires touching and shorting out under the boiler. Also, since I keep very close tabs on the machine, when it's in use, I think the likelihood of the boiler running out of water is rather rare.

What are your plans as far as steaming? Does this machine use one boiler for both brewing and steaming, or is it a heat exchanger or what?

Ray

Exactly Peacecup- actually I have found that the Sama (Elektra) and PV groups seem to cool themselves reasonably quickly.Having a direct LED readout from the neck of the group will make taking that flying f!@# just that bit easier... . Will be immediately possible to see exactly how the group warms up as the thermosyphon kicks in- during and after shots- flushing, and with changes to the boiler pressure/temp from the PID.

Ray:

I agree about simplicity- and the Sama machine is fantastic in that regard. It is a perfect, simple system with very serviceable parts. And unlike your Pavoni the (spring) group is connected to the largish 3 liter boiler by a thermosyphon connection. So it actually does not need a PID anywhere near as much as a Pavoni-0 the overheating issues are almost negligible. As PC noted- the group head is quite stable in operation and you can already leave the machine on for hours at a time without it overheating... generally these machine are set between 0.9 and maybe 1.3 BAR. The steam comes direct from the same boiler.

So why even add a PID? As I see it the extra electronic will be the first things to go almost undoubtedly- the same is likey true for your Pavoni even after your bespoke work. Hopefully they will last for many years though.

The only reason is because I can (with help)- and I think it will be a very interesting experiment. The Sama was virtually ruined when I got it- rusty, with IKG of scale in the boiler- destines for the trash heap. I intend to restore it beyond the original factory condition as much as possible- and to add some bling too- and along with the ability to read the group head temp it will be very informative to see how fine adjustment of the temp/pressure affects the espresso making process.

I think it will offer better performance- the ability to extend the deadband for the element recycling- I think a range anywhere between 1 and 1.2 BAR would be possible good espresso ranges and with a PID I could experiment with longer element cycling times. I also plan to insulate the boiler.

I think keeping the P-stat inline is a good idea- as an additional safety feature. It should last well- not getting used at all in theory...

here are a few picture of the machines journey so far:

in parts: all electronic completely shot. pressure gauge shot. The boiler was full of scale and the element cemented in place. One of the thermosyphon tube was completely blocked... the group head removed (perfect! Solid brass- but filthy)



The boiler was soaked in multiple acid baths at an espresso machine workshop (strong acid) for two weeks, then boiled in citric acid- then soaked for a week in a bucket in the sun in a special acid solution all my own: including the juice of 30 lemons and half a can of citric acid, some CLR, vinegar, etc. At the end one weld at the sight glass failed and a steel nut welded to the base of the boiler dissolved...



Had the boiler repaired by a radiator repair guy- he welded a new nut onto the boiler and filled those divots. He also repaired a failed weld on one of the sight glass pipes. Then I sandblasted the chassis and cabinet


starting to look good...



first coat of hammertone:



hmm, patchy. After wet and dry sandpaper- and several more coats- and then a polish- looks great:

sorrentinacoffee wrote:...I am assuming your proposed set up is quite different? It may be even better as the LED display does not sound very deep... I would liek to keep that part as small as possible as it will be mounted beside the boiler- and getting close to the P-stat. When you say: "Main board with AVR microcontroller and 4-channel thermocouple sensing subsystem (approx 3" x 2" x 1/2") I am using two channels at present, boiler and grouphead."- - this sounds like the brain? could this part go under the chassis as well? Is it possible you could send a picture? There is a lot of room there- the only limitation is the height- which is a bit over an inch.

Yes, main board = brain. It is pretty flat. The tallest things on it are the USB connector and the power jack along the left side. The green terminals on the top are for the thermocouples. The OT1 and OT2 pin headers at the bottom are to drive one or more SSRs. The 4-pin I2C header at the right is for a remote LCD.



The LCD panel can be mounted anywhere convenient. It connects with a 4-wire cable. The extra long button stems work well if you panel mount the display. There are sockets for 3 LED's, but I don't use them for espresso PID.



Just to show how things go together, I have it running from a 9V battery. The top line of the display shows elapsed time (I have mine set up to turn off the heater after 8 hours) and current temperature:setpoint. The second line shows boiler temperature, group temperature, and current heater power level.



If you have room, the LCD can also be stacked on the main board. This eliminates the cable. The whole assembly could be panel mounted, if you have enough depth. FWIW, for my installation, I separated the display from the main board because it fit better.



(These photos are of a couple of prototypes I built along the way, but essentially the same as where I ended up).

Jim