Got up this morning, fired up Clarabelle, read a little email while she warmed up, pulled my first shot of the day and went to get dressed. Came back a few minutes later to the recognizeable smell of hot electronic components and Clarabelle at a full boil. Went to drop the thermostat and it wouldn't go off. Tried the on/off switch and it didn't do anything. So I figure there's most likely a problem with the microswitch in the thermostat assembly. Does anybody here have experience working on the old Caravel thermostats and have any hints about what I should be looking for when I take things apart later tonight? I've been intending to PID her for a little tighter control over temp since the dead zone on the thermostat is about 20°F (11°C), but I'd rather not have to do it just now.

There is some info in this thread:

http://www.home-barista.com/lever...avel-t2370-40.html

I had seen that, but was hoping that somebody might have more info/experience messing with it. I'll be sitting at the table tonight with Clarabelle in pieces in front of me, hoping I can get her back into some semblance of working order before I need coffee tomorrow morning.

GVDub wrote:Tried the on/off switch and it didn't do anything. So I figure there's most likely a problem with the microswitch in the thermostat assembly. (...)

It is really strange that the on/off switch didn't work... in my model it was able to override the thermostat (now the pid).

Good luck getting it fixed. Although there may be a 'silver lining' if it pushes you to get the pid installed

Regards, Henry

hbuchtel wrote:It is really strange that the on/off switch didn't work... in my model it was able to override the thermostat (now the pid).

Good luck getting it fixed. Although there may be a 'silver lining' if it pushes you to get the pid installed

Regards, Henry

My microswitch was toast - burnt toast. When I pulled off the bottom and looked, one of the wires detached and the switch crumbled into a gray powder before my eyes. I intend to drop an email to Doug to see if perchance he has parts, but otherwise, it's going to be PID time, which is an expense I was hoping to put off another month. Ursula may come home from work for a bit.

GVDub wrote: Ursula may come home from work for a bit.

I knew having an extra women would be a good thing.

I just hope this doesn't end up being like my friend who used to drive a Rover TC2000. He had to keep 4 of them just to have one on the road running at any given time.

Out of curiosity, were you running ClaraB on 110 or 220? Could be some date here to glean from your misfortune. You could likely bypass the entire thermostat and just wire direct to the element with the pilot light circuit intact. I think there are a lot of Caravelers,, or are they Carvaliers?, who just go to boil and then turn off the machine to let the temp drift down to a good 94 or so for the shot. The thermostat has such a wide range to begin with that it seems more or less useless for any actual temp control, it just keeps the kettle hot enough.....actually the thermostat just confuses the issue for me, I prefer to go to boil, and turn off the machine to drift down to what seems like the right temp by timing of the dose fixings etc, but then again I don't demand great great great on a 3 shot succession but OK, good, better, getting there, OK, had better, WOW, is fine with us. As long as there is one WOW in the day we are more or less happy.
And even a WOW every other day puts us ahead of the curve. But by the way, not a lot of spare parts for those old babes.

Running on 220, with a 1000 watt step-up xformer. I've tied the two wires that were connected to the microswitch together, so it's now just on when I flip the switch on the xformer. Didn't figure there were a lot of spare parts, but it never hurts to ask.

I've been getting consistent wows with Clarabelle, even with the normal temp fluctuations. Adding a PID for temp control will make it even better, and possible to start getting tighter on the shots once I get used to it (that blend works best at 201.2°F? No problem, just tweak the set point). I've got most of the hardware I need (project box, SSR, panel mount power sockets, bits of wire), and I just need to decide on which controller I want to get and find the right immersible probe. If it works well for Clarabelle, I'll do one for Ursula, too.

GVDub wrote:I just need to decide on which controller I want to get and find the right immersible probe. If it works well for Clarabelle, I'll do one for Ursula, too.

This is the RTD I am using for my La Peppina. The metal sheath is 2" long and it is designed for immersion:

Auberins Platinum RTD. PT100 (6 ft. cable)
Item# PT100S $15.65 USD 1 $15.65 USD

That's exactly the one I've been looking at. Is the cable immersible too, so that it can just be dropped in the kettle? I'm also looking at the 1/16 DIN dual display temp controller they have. I suppose if I ordered today, I could have this together next week, huh?

Yes the cable is immersible too. I dropped mine in my Peppina kettle. It also survives mild descales. Don't forget a driploop and fix legs under the project box.

It's a Hammond project box with ribs on the plastic body and aluminum faceplates front and rear, so it's already fairly water safe, but I'll stick some feet on it nonetheless. I may even put a tilt bar on it for ease of reading on a table top. It's roughly a 1/3 rack width, so there should be enough space on it to mount the PID, SSR, and sensor connection to the front panel and have a power switch and the in and out power connections on the back. As long as I'm careful with the nibbler, I think I can do a pretty clean looking box.

Because of the problem Bluecold had with water I also mounted my PID toward the top of the box instead of the center, this gives me at least an inch of open space under the controller. I do not know about any other controller, but the Fuji comes with a rubber seal that mounts between the faceplate and the outside of the project box to waterproof that spot. I mounted my heatsink fins so that they stick out of the box. To do this I cut the box (mine is all plastic) with a hacksaw so that the slit was just smaller than the width of the fin center and then filed it until it just barely fit. I should pop some sealer around, but have not done so. The holes for the power cords and RTD still need sealing, I have an RTD plug and socket which I should put on someday and that would take care of that spot.

Someone also suggested drilling a drain hole in the bottom of the case. I think that is a good idea, but I have not added feet yet so I will wait until I do that.

This thread has lots of good information in it, and the last two posts have pictures of my finished box.

One thing I was going to ask: Is there a screw which connects to the tank that might be a good place to connect a TC instead of immersing? When I bought the RTD for my Peppina I also ordered a TC with a washer end in the hopes that I could avoid dropping the probe through the lid. When I get my hands on my Caravel I wanted to try using the washer type TC.

Take Care,

One of the things I like most about the Caravel is that the kettle isn't really connected to anything. Wiring a probe to it would defeat that, which is why I think I'll be sticking with that immersible probe. I'm getting the socket mount for the RTD with the probe and using gasketed IEC connectors on the box so that all cables that hook into the box can be disconnected and it should be somewhat more water resistant.

Since the Caravel only uses 3 amps at 220v, the 25 amp SSR I picked up shouldn't need a heatsink, which will make putting things together a little easier. I've just wondering why some enterprising company has made a PID with an SSR built in. After all, they make ones with mechanical relays built in, and the mechanicals have a much shorter MTBF.

GVDub wrote:Since the Caravel only uses 3 amps at 220v, the 25 amp SSR I picked up shouldn't need a heatsink

I am not sure about the relationship between the wattage through the SSR and heat generated. From what I read it is the cycling on and off of the SSR which generates heat. Hopefully someone who has more knowledge about this will chime in.

The docs with the SSR say that any load under 10 amps doesn't require a heatsink, and I'm taking them at their word.

GVDub wrote:I've just wondering why some enterprising company has made a PID with an SSR built in. After all, they make ones with mechanical relays built in, and the mechanicals have a much shorter MTBF.

They are available. In the Watlow line, a SSR output is usually identified by "K" in the model number. But the capacity is usually fairly limited, around 0.5A. That's not enough oomph to switch the heater in a boiler.

Jim

ddr wrote:I am not sure about the relationship between the wattage through the SSR and heat generated. From what I read it is the cycling on and off of the SSR which generates heat. Hopefully someone who has more knowledge about this will chime in.

Heat generated inside the SSR is directly proportional to the load current (amps) while the SSR is conducting. Because it is a semiconductor device, there will be a small voltage drop across the load terminals. Most SSR spec's I've seen say the max voltage drop will be 1.6V while it's conducting. The one's I've actually checked have all been below 1V.

In GVDub's application, his SSR will probably only generate around 2W; certainly no more than 5W (3A x 1.6V = 4.8W).

When considering the maximum temperature an SSR might reach in service, the worst case condition is when the load current is on for a long period of time. For espresso machine applications, the max SSR temperature generally occurs during initial warmup of the boiler. Once the boiler is up to temperature, and the heater is being switched on/off only to maintain the boiler temp, the SSR temperature drops significantly.

Just like the heating element in the boiler, I believe that frequent switching is also better for the SSR because it minimizes the temperature differences between the on and off parts of the cycle. So the mechanical stress reversals inside the SSR are kept to a minimum.

Jim