After several extensive searches regarding ET probe placement on a gas roaster, I decided on putting the TC as shown above. The brown wire on the upper left of the chamber is the one I added. Consideration was given to placement in the drum, in the exhaust air chamber, or outside the drum as shown.

I'm logging with an Omega HH806AU. I have an analog gauge in the exhaust chamber (on the right side of the bean chute) which seldom reads above 350' F, so I avoided that choice.

The temps I get with the current location run in the high 500 to low 600 F. readings until toward end of roast when I reduce heat dramatically and spool the airflow up. Additionally, Artisan flat lines the graph reading of the ET at @ 570 degrees (even when the Omega shows higher), so I am not able to see any correlation to burner adjustment on the chart. I am satisfied with the BT readings and graph, but would like to be able to see the relationship between ET and BT more clearly.

I can relocate the probe, but wanted suggestions here before I start tapping another hole in the machine.

Can you feed one thru one of the loading chute's 4 screws? It would take a bit of modification, but should work well. I'd opt for the right side, facing the roaster, if your drum turns clockwise, the left wouldn't work well. I'd imagine your upper right would be the best option, as long as you can route it out of the path of tumbling beans. You might be able to, quite easily, by dog-legging the probe a bit after insertion.

There's a delicate balance when it comes to ET measuring. Too close to the drum and you get artificially higher than actual temps from the radiant heat from the drum itself. Too close to the middle and you risk the probe getting hit by tossed beans, which is no good either.

I currently roast on a 3k USRC machine at work with a thermocouple measuring the temp (of preheated air, of course) going into the drum and one measuring the temperature as it exits the drum. The average of these two readings proves to be useful in lieu of a true ET, and the set-up avoids the pitfalls that you and TomC are describing. Readings rarely exceed 450F. I have never seen the USRC sample roaster close-up, but perhaps a similar arrangement would suit your needs. Cheers,

Ol'Timer wrote:<image>Additionally, Artisan flat lines the graph reading of the ET at @ 570 degrees (even when the Omega shows higher), so I am not able to see any correlation to burner adjustment on the chart. I am satisfied with the BT readings and graph, but would like to be able to see the relationship between ET and BT more clearly.

Check your Artisan settings, Config>Axes...>Temperature Axes Min and Max. Change max to the highest you want on your graph.

Thanks, guys, for the responses. Tom, I considered in the drum placement , but wanted to avoid possible problems with bean collisions. The drum is only 6" in diameter and I prefer to run 454g batches, so I did't want to run the risk of getting too close to the drum.

Being a visual guy, I would like to be able to look at the graph of the roasts real-time, so averaging multiple temps doesn't give me the chance to see the ET, BT relationship I'd like.

tamarian, I have adjusted the axes, but the problem is that the recorded temp maxes out before it gets to the top of the axis.

I'm considering removing the analog dial in the exhaust channel and positioning a TC in it's place to be able to record to Artisan. I'm just not convinced that position represents a reliable environmental temp.

The object of this for me is to see the relationship of the ET and BT so I can get a predictable behavior to gas flow/air flow adjustments in the roast.

I really believe I can do this where the probe is currently if I can figure out how to get the software to show the actual ET.

You're really not that interested in the actual ET (whatever that is), so much as you are in a measurement which varies in the same way a fair average would, to provide a sufficiently accurate picture to use for whatever purposes you consider important -- comparing respective RoR, or specifying a "drop dead," MET for instance.

As a stopgap, the first thing you might want to try is programming an "offset" into the datalogger.

But as you already knew when you started this thread, sooner rather than later, you'll want to relocate the probe.

At a guess, the hopper (aka "bean chute") location where the current analog probe is located is a good site. At least it's out of the way of the convection coming from the live flame; the convection, by the way, which along with radiation from the drum is probably giving you such high readings now.

Finding the best site for a probe can be tricky. Before doing anything permanent, ask Dan. I know for a fact that USRC sells Sample Roasters with pre-mounted ET thermocouples (mine, for instance). You might as well use the official location.

Twenty days after ordering, and with a functioning 1kg roaster in place, I'm cracking. How did you maintain your sanity? Or, did you?

Six weeks to go,
BDL

Ol'Timer wrote:tamarian, I have adjusted the axes, but the problem is that the recorded temp maxes out before it gets to the top of the axis.

Check Tools>Extras>Input filters. Something in your Artisan config is set at around 570F or 300C, which causes the flat line. If all else fails, do a Help>factory reset for Artisan. Note that changing the axes limit will only work for new profiles, so you can test changes on the next roast, or test with sacrificial beans, green or roasted, and play with the config while test roasting to see if it wil budge.

I really believe I can do this where the probe is currently if I can figure out how to get the software to show the actual ET.

It should work, my ET (fluid bed inlet) goes way above 570f/300c.

I had thought I read once upon a time that ET's should not go higher than 500°

Am I wrong about the number or about the whole fact there is a limit?

boar_d_laze wrote:You're really not that interested in the actual ET (whatever that is), so much as you are in a measurement which varies in the same way a fair average would, to provide a sufficiently accurate picture to use for whatever purposes you consider important -- comparing respective RoR, or specifying a "drop dead," MET for instance.

As a stopgap, the first thing you might want to try is programming an "offset" into the datalogger.

But as you already knew when you started this thread, sooner rather than later, you'll want to relocate the probe.

At a guess, the hopper (aka "bean chute") location where the current analog probe is located is a good site. At least it's out of the way of the convection coming from the live flame; the convection, by the way, which along with radiation from the drum is probably giving you such high readings now.

Finding the best site for a probe can be tricky. Before doing anything permanent, ask Dan. I know for a fact that USRC sells Sample Roasters with pre-mounted ET thermocouples (mine, for instance). You might as well use the official location.

Twenty days after ordering, and with a functioning 1kg roaster in place, I'm cracking. How did you maintain your sanity? Or, did you?

Six weeks to go,
BDL

You are right on target, Rich. I did talk to Dan about the probe placement and decided to avoid inside the drum based on his recommendation. I have ordered a new TC from Omega with a compression fitting that will fit in the location where the gauge is, so it won't be a big deal to remove the analog gauge and insert a TC that will log to the computer.

I strongly suspect this is the location where Dan will install your ET thermocouple. Willem Boot recommends the exhaust air path as the location of choice for the ET device.

As for my sanity, it was gone long before I ordered the roaster! But, we are having fun!

tamarian wrote:It should work, my ET (fluid bed inlet) goes way above 570f/300c.

I thought I looked everywhere for the culprit, but lo and behold, the location you pointed out had the max ET temp. set at 572. Thanks for your help.

The ET thermocouple should be outside the drum. The location just beneath the hopper will be a good start:



This is my thermocouple: