Low cost thermocouples for the IKAWA

Discuss roast levels and profiles for espresso, equipment for roasting coffee.
GDM528
Posts: 847
Joined: 2 years ago

#1: Post by GDM528 »

Below is my cheap hack to install a thermocouple into the roasting chamber for my IKAWA Home. These are my starting tools and materials:



There are many suppliers for high-temperature Kapton tape, for example: https://a.co/d/4FqVFI3

And these are the thermocouples I start with: https://a.co/d/aSU60Zr
Are they traceable to the National Bureau of Standards? Heck no..,. whatever. They're only $10 for a pack of five, so you can pick through and find one that looks like it will survive the beat-down from the spinning beans, and be reasonably accurate. A lot of latitude around the word 'reasonably' IMHO. Probe placement will likely have a bigger influence on the readings than the variations between probe samples, but at $2 each I'm not gonna sweat it.

These thermocouples are unique in that they have a polymer jacket, probably Teflon. Many of the other thermocouples options are wrapped in fiberglass, but I worry about getting glass fibers in my coffee - no thanks. I've also hand-built the bare wire onto Kapton, but that process is very labor intensive. These cheap Teflon jacketed probes produce less emotional trauma when they break down over time (maybe 12 months or so).

I start by straightening the wire as flat as I can, pin it down and lay a 6-7cm (2.5") strip of Kapton on the wire:



Then I bend the tip so it will extend into the spinning bean mass:



Then I lower into the chamber and tack down the tape. This is where that long wooden swap comes in handy.



Next I cut short strips of Kapton around 4cm (1.5") and trim off the corners. Then I use tweezers to carefully position the strips perpendicular to the wire, on each end. I've found this 'H' pattern to hold up well over many roasting sessions.



After I've bent and shaped the wire across the foam strip, I use flat pliers to squish-squish-squish the wire to flatten the insulation that crosses the foam strip:



Then I tape down the wire on the outside of the shell, and done!


You'll end up with a lot of excess thermocouple wire after you've trimmed the installed portion. There are instructions online for how to use that leftover wire to make your own thermocouples. I use a portable car battery charger and some #2 pencil leads to weld the wires together and form the bead - wear safety glasses!

User avatar
drgary
Team HB
Posts: 14362
Joined: 14 years ago

#2: Post by drgary »

Gary,

These are great instructions. Thank you!

A caution light went on when I saw the thermocouple sheath material described as PFTE, which has a maximum tolerance of 500°F / 260°C. One of the people rating the thermocouples noted that the packaging says they'll withstand 400°F, not C. Am I wrong with this? If not, what other thermocouples would you recommend?

Does something like this work? More expensive, I know.

https://www.amazon.com/dp/B076DG626V/?c ... dp_it&th=1
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!

GDM528 (original poster)
Posts: 847
Joined: 2 years ago

#3: Post by GDM528 (original poster) »

Good catch Gary!

I know this may be splitting hairs, but their documentations states the 'test' range is 400C without explicitly stating if they pass or fail at that temperature ;) Elsewhere, one of their marketing graphics says "Can withstand 200 C" So, yeah, kinda marginal.

FWIW, the Inlet temperature setpoints are quite a bit higher than the actual temperature in the chamber. I generally use 40degC as a starting point, but the drop will increase as the temperature rises. I generally don't roast any higher than 225 degC in the chamber (Full City roast level), which equates to around a 270 degC Inlet setpoint. I taste a significant uptick in 'burnt' notes if I go any hotter.

I checked the insulation with a precision PID-controlled soldering tip, and observed the insulation starts to soften above 230 degC and gets gooey by 250 degC. So, mechanically speaking, the insulation is just barely hanging in there at the highest temperatures I hit during the roast. Fortunately, the highest practical temperatures in the chamber can't get high enough to decompose the insulation.

The highest inlet setpoint for the IKAWA Home is 290 degC (which I tried) but that only reached about 240 degC inside an empty chamber and 235 degC with beans loaded in the chamber. I don't think any decent greens deserve this sort of treatment, but if one is interested in French roasts, they should consider more robust thermocouple insulation.

The Kapton-wrapped thermocouple you linked is an excellent alternative and essentially immune to the highest heat levels anywhere in the IKAWA. It's a heavier gauge wire, so it could be more disruptive if placed into the spinning bean mass. That might actually be a good thing if one puts some thought into where to place the thermocouple.

User avatar
drgary
Team HB
Posts: 14362
Joined: 14 years ago

#4: Post by drgary »

The crucial question about the PTFE insulation is at what point it starts to off-gas. Does that happen once it starts to soften? Or does it completely have to melt? I've got some leftover thermocouples with fiberglass sheath and have ordered kapton tape. I'm thinking of wrapping the insulation that goes into the chamber in tape to see if that eliminates any glass intrusion and will post here if I have that working.
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!

GDM528 (original poster)
Posts: 847
Joined: 2 years ago

#5: Post by GDM528 (original poster) »

Per Wikipedia, outgassing lethal to birds kicks in above 250 degC. So, I can't program the inlet setpoint high enough to reach that temperature in the chamber.

User avatar
drgary
Team HB
Posts: 14362
Joined: 14 years ago

#6: Post by drgary »

Sorry to belabor this. I just used an IKAWA recipe that heats the chamber to 500°F / 260°C and holds it there for awhile. Here's the graph. But I think you're saying that the inlet temperature doesn't carry into the chamber temperature. Have you measured the temperature drop just above the inlet port?

Gary
LMWDP#308

What I WOULD do for a good cup of coffee!

GDM528 (original poster)
Posts: 847
Joined: 2 years ago

#7: Post by GDM528 (original poster) »

The inlet setpoints are referenced to a sensor tucked under roasting chamber, adjacent to the heating coils - that's nearly equivalent to putting a thermocouple in the flame burner of a drum roaster. The inlet setpoint is the hottest point in the system, and everywhere else will be cooler - a lot cooler. For example, this post shows the drop between setpoint and in-chamber: Ikawa Home thermal performance

Although technically accurate, it seems a bit unfortunate that IKAWA chose to display the actual setpoint temperature for their control loop. Those setpoint temperatures (e.g. 260 C) would incinerate the roast! Fortunately, the actual in-chamber temperature is considerably lower: 30-50 C lower, depending on the setpoint (lower setpoint = lower difference). And it's also fortunate the drop is stable enough to make the relationship between inlet setpoint and BT predictable, hence the spreadsheets and web pages that obviate the need for an in-chamber temperature sensor - yes, I know that's heresy ;)

I've never attempted to measure the temperature at the bottom of the chamber, given that there's already a more accurate and stable version of that reading via the inlet setpoint. Rather, I've been targeting the top of the spinning bean mass, that reflects the cooling effect of the greens.
★ Helpful

GDM528 (original poster)
Posts: 847
Joined: 2 years ago

#8: Post by GDM528 (original poster) »

Update:

After a dozen or so roasts, many of which with inlet setpoints as high as 270 degrees C, here is what the teflon insulated thermocouple probe looks like:



The insulation doesn't show any signs of discoloration/softening/flowing. Looks like the cheapo Teflon-insulated probe is hanging in there.

Disclaimer: putting a fluoropolymer in the roasting chamber could theoretically lead to dead parakeets, flu-like symptoms, etc. I don't have any pet birds and I aggressively vent my roasting exhaust outside so my neighbors can deal with it ;) Melting and/or discoloration of the Teflon can be a signature of outgassing.

User avatar
drgary
Team HB
Posts: 14362
Joined: 14 years ago

#9: Post by drgary »

Sold! I'll install mine soon.
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!