View Thread

GreenBean wrote:

I see that you have started a thread on another coffee forum discussing this same subject i.e. advice on installation of thermocouples in your Hottop. It is difficult to follow your current thoughts on what you propose to do across the two threads. There are many different approaches possible and you have received good advice from a number of contributors. The solution proposed by okmed looks good to me provided nylon ferrules are only used on the cold side of the fitting. My views on what I perceive to be your current proposal are:

1. There is no substitute for dismantling and measuring your Hottop with your proposed solution in mind. I believe Hottop have made changes to the design of key elements such as the drum. Measuring your own Hottop is the only way to be certain that your solution will fit.

2. Choose the thermocouple length to allow maximum safe protrusion into the roast chamber. This will provide the most accurate readings. The drum drive motor can be disconnected to allow the insertion or removal of the thermocouple.

3. My view is that 3 mm diameter probes offer the best compromise of smooth output, speed of response and robustness (of the sheath and thermocouple wires). 1/16? thermocouples will work but you should expect less smooth temperature and RoR curves and may have issues with breaking thermocouple wires.

4. Fibreglass insulation provides good thermal insulation to the thermocouple wires but is not ideal to insulate them electrically from any mains connections within the roaster. To ensure a safe installation you should route fibreglass insulated thermocouple leads through suitable silicone sleeving/tubing.

5. The rear wall of the Hottop is too thin to provide a robust threaded connection to the fitting although okmed reports good results so far. The connection may well fail after a time. In this case you would still have the option to fix it with J B Weld.

You have asked if RoastLogger will graph the fan and heater settings. I have deliberately made the decision not to chart this information as I believe it encourages users to think, and attempt to control their roaster, based on elapsed roast time. Elapsed roast time is a poor indicator of roast stage and attempting to control a roast based on this produces poor results. I have covered my views on this in several posts on other coffee forums. In summary, RoastLogger allows you to control the roast by making changes to the heater and fan based on bean temperature. This produces much better control of a roast and more consistent results in the cup. RoastLogger provides a simple set of tables that allow you to script the changes in settings. These changes are applied exactly as scripted in each roast and RoastLogger indicates when each change is applied. Charting this would not provide any additional useful information.

BarryR wrote:
4) Isn't fiberglass an electrical insulator?

BarryR wrote:
...The XCIB is 6.4mm diameter with an exposed tip, so I imagine this would combine the benefits of ungrounded (not needing grounding) with the rapid response time that the exposed tip provides.

BarryR wrote:...Has anyone mounted the XCIB type without the epoxy? If so, which thermocouple or other method was used....

BarryR wrote:...I recall on poster epoxying a sleeve but then holding the probe itself in place with basically friction fitting in a silicone (or such) sleeve. Wouldn't this tend to shift?...

BarryR wrote:...I believe the XCIB is stated to be very durable but for the BT probe, wouldn't the beans constantly banging into the exposed tip put it at risk for breaking?

BarryR wrote:
...I'd like to use a back (or front) nut to hold the fitting in place rather than relying on the thickness of the backplate.
My remaining questions are:

Do I need to tap the nut (as other's have described for the 1/16th" solution) or can I purchase it (if the latter, what nut do I purchase)? Are there compression fittings that would have a standard (non-compression) thread on the part that I need to secure through the backplate so I could use standard hardware at that end? If not and I need to tap the nut, what nut do I start with and what tap would I need (I'm guessing 1/8" since that's what OKMed used in the backplate)?

I hope I haven't exhausted your patience.

smico wrote:

Have you guys considered RAG's solution:
http://homeroaste...post_45922
I am champion of improvisations, but if I would need to do it again, I would do it exactly like that.

smico wrote:

I used to use 1/8 inch probes, fixed on the back wall, but they were quite slow to respond. Not critical for BT, but kind of annoying for the ET measurement.
Truth to be told, I simultaneously switched to insulation from the back-wall, and to 1/16 inch probes, so I don't know how to attribute which change had more effect.
Your solution seems simple and effective.

smico wrote:

Would it be possible to tap the hole in the back-wall, and thread just a few threads on the outside of coupling so you can screw it flush in? I am not sure if the back-wall is too thin... Your uncle might know.