Adding heat storage to a KKTO

[readeral]

Hi all,

I've been thinking about this for some time, and hoping I've come up with a beginning to the solution.

Before y'all tell me that you love your stock KKTO setup - that's totally fine, but this thread is specifically about trying to increase the heat stability of the system.

My setup: I've modified my Turbo Oven to be under the control of a TC4 for it's heater pulse cycles (e.g. at 50% it's on for 3 seconds, off for 3 seconds), and I use the PID of my software to set my temperatures. This has decreased my roast lengths from 25 minutes (as my stock TO was crap) to 15 minutes. This was a great start. The KKTO is also triple wrapped in a fire blanket, and has insulation under it's false floor.

What I'm finding is that the whole system is highly susceptible to changes in the TO heat level, as the fan stays at a constant temperature, and the majority of the heat that the beans receive is fan-forced convective heat. A drop in element temp is a very quick drop due to the high airflow.

I've placed a probe below my 'roasting chamber' and observed that it doesn't get above 100 degrees. The system just doesn't hold much, nor does it circulate much outside the roasting chamber. (Turn it on empty and you'll see chaff swirling around in the chamber perfectly happily without falling through.)

SO... I wanted to solve this problem without doing a complete rebuild and was doing my head in, so I tried to come up with some way of 'filling' the space below the roasting chamber with something to store heat and then support the heat stability of the system. Stored heat coming from below, not just via the fan-forced TO.

I've found a bunch of ceramic pie weights on clearance, so bought 900g of them. They're 7 or 8mm in diameter, and I'll try to come up with a way to restrain them from the driveshaft (and the driveshaft hole) and put them into the roaster below the roasting chamber. It'll be a much longer heat up and temperature stabilisation, but hopefully it'll mean that I can start the roast with a more stable environment.

I roasted today, so it'll be a week or two before I experiment with the pie weights, but I'm hoping it works! It might not make a big difference, especially given it's only 900g of ceramic, but I'm hoping it'll be enough.

Eventually the goal is to gain control of the fan in the TO too, and start with a lower heat from above and lower airflow, so I don't strip the moisture so quickly (and hopefully retain some more of the aroma of the beans). I'm also going to see if I can work out using PWM to regulate the power output of the element (sooner rather than later... but it's the same solution for controlling the fan, so they come together) so I don't need to pulse it on and off, but rather up and down.

Al

[Dimal]

G'day Al...

Quite a long time ago, the prospect of utilising 'ceramic heat beads' (or similar) in this manner was discussed but I can't remember the details. I can't even remember what type of roaster was being considered. For what it's worth, it should work as you hope but to what degree, it is difficult to say of course but you will no doubt keep us informed of how it all progresses.

All the best with your experiments...

Mal.

[Koffee_Kosmo]

I was thinking about this some years ago and I will share where my investigations led me
My research led me to Zeolite balls / marbles, they hold 4 times the heat of boiling water but that was dashed as I found out that Zeolite stores heat and only releases it when water is added

Then I researched natural rocks still not enough heat

The only static thing that I think might work but haven't tried yet is an aluminium heat sink

On another note you can use a second heat source, like a cut to size electric frypan as the base under the perforated pot

KK

[readeral]

Thanks for the comments guys.

Yeah, so I'm wanting to avoid a rebuild, so the second heat source isn't an option. I know it's a solid suggestion, but I had to call in so many favours to get this KKTO built, I don't think I could do it again! The best I could do (and the results have been excellent) was electronic control of the TO.

I'm not expecting to get much from it, 900g of ceramic is a pretty small amount, and I like the quick control of a KKTO roaster. But I figure that if it's able to just smooth out some of those temperature changes, esp. toward the end of the roast, and start with a gentler fan, then I'll be happy enough.

I've decided I'll get an aluminium 'egg ring' and fix it to the false floor with permatex, that should do the job of keeping the weights back.

I'll be putting a thermocouple probe in with the weights to know when they've got to preheat temperature, and also to measure the effect of removing the turbo oven off the pot set, and then to measure the heat difference between ET and the weight temp once the lid is back on. One risk is that these might quickly drop in temperature below the ET I want to maintain, and as a result actually slow down my roast because they then need to take on more energy, while the coffee beans are needing to do the same.

Standby for feedback!

[ArnhemR]

Interesting thread, i will be watching to see what you find out!

[MrJack]

Don't worry about the weights dropping below the ET - without a chemical reaction going on in side (or if they get red hot), that's pretty much thermodynamically impossible. Heat only moves from a region of high temperature to one of low temperature.

They will definitely cool if the ET falls (that's the point!). It is possible to model this mathematically - it's relatively straightforward if your only looking for indicative numbers.

Also, I could be missing something here w.r.t the KKTO design, but If you're main concern is compensating for the drop in air temp when the element power drops, perhaps the best place for the beads is in the air stream between the element and the beans (rather than underneath them?)

[readeral]

Thanks for the comment MrJack. That's very true, but practically very difficult to achieve.

When I say dropping in temperature it's "below the ET I want to maintain". That is, I want to maintain 265c (which is what I'm normally going for at the moment). If I remove the turbo oven lid (to put in my beans) and the ET drops to 200 degrees and the beads drop to say 220 degrees, then after the ET has returned above 220 degrees and as I push toward my steady ET of 265, I'm pumping energy into the beads as well as into the coffee, which might slow down my roast in the opening stages.

However, if using the beads I can maintain them at say 240 degrees (that is, if they don't drop very far at all) then I'd be able to get much more benefit from them stabilising the ET very early in the roast.

At the moment, I preheat to 265c, my ET probe drops to 195c after the turning point, and reaches my target ET of 265 at around 6:30 and stays there pretty much right up until first crack when I start playing with my ET temps a bit more.

The ideal would have some sort of hopper to deliver the beans into the environment without the massive air temp losses, but then that just ads more metal to try to insulate against heat loss (plus a lot of cutting into stainless steel..) If there was a safe way for me to get a tube through the TO glass, that would be the best option, as lots of heat gets lost through the glass anyhow.

[Dimal]

Probably the one thing against using beads of any kind, will be the significant interstitial spaces between the beads which will not help either heat retention or conduction. If you know of a pottery makers supply warehouse somewhere near where you are, you could see if they stock different grades of "Ceramic Grog" (not booze ). This is basically screened and graded ceramic material from a very fine almost dust-like size up to quite coarse sizes in the multiple millimetre range.

I'd reckon that this would have to be more efficient than something like the beads.

Also, with regard to the heating element control you have, something that incorporates a relatively high frequency Pulse Width Modulated (PWM) output control (in the high hundreds to low thousands Hz), would provide a much more granular control of the heat output from the element. There may be some mod's around for the system you already have that might be worth researching...

Mal.

[readeral]

Yeah so the beads are a first step I reckon, I'd thought about getting something made up (there's a ceramic/clay shop about 40 metres from my apartment!) but wanted a non-destructive (or at least, easily reversed) approach as a first step. Then I'd think about lowering my false floor for something more permanent.

But thanks for the suggestion! I'll look into the grog...

Yep, we're working through the PWM options at the moment - both for element and fan. The bits and pieces around the web about TC4 and aArtisanQ are quite difficult to piece together, so can be hard to know what we need! But we're putting together a ZCD at the moment, and will pick up some new SSRs soon. Hopefully will be up and running soon!

[jbrewster]

Yeah so the beads are a first step I reckon, I'd thought about getting something made up (there's a ceramic/clay shop about 40 metres from my apartment!) but wanted a non-destructive (or at least, easily reversed) approach as a first step. Then I'd think about lowering my false floor for something more permanent.

If you decide to go ahead, blending a large proportion of grog (as much as you can get in there and have it still hold together) into kaolin clay should give you good results, ideally you'd also get somebody to fire it in a kiln for you.

Alternatively a commercial 'dense castable' refractory commonly used in pizza ovens and such (also, for furnaces that melt metal), could be a good option, in your application it wouldn't be necessary to actually fire it (there are binders in commercial refractory which behave like cement) ideally you'd still have some insulation on the back side though, but as you're not melting metal that's probably less of a concern

[MrJack]

Probably the one thing against using beads of any kind, will be the significant interstitial spaces between the beads which will not help either heat retention or conduction. If you know of a pottery makers supply warehouse somewhere near where you are, you could see if they stock different grades of "Ceramic Grog" (not booze ). This is basically screened and graded ceramic material from a very fine almost dust-like size up to quite coarse sizes in the multiple millimetre range.

I'd reckon that this would have to be more efficient than something like the beads.

Mal.

I would have thought that the dominant mode of heat transfer with whatever thermal mass is under the beans would be convection, not conduction. In which case, surface area is king, and beads would be ideal.

In addition, ceramics generally have low thermal conductivity and often also low heat capacity (relative to metals) - which is why they are used as refractory materials (insulation).

So perhaps some cast iron or ball bearings would be a better choice.

Though if you are putting them into an area which doesn't reach temperatures greater than 100°C normally, it won't help you anyway.

[readeral]

Though if you are putting them into an area which doesn't reach temperatures greater than 100°C normally, it won't help you anyway.

I'd be preheating it without the roasting chamber in.

[MrJack]

I'd be preheating it without the roasting chamber in.

In which case, they will likely cool down during the entire roast (which might be ok).

[Dimal]

I would have thought that the dominant mode of heat transfer with whatever thermal mass is under the beans would be convection, not conduction. In which case, surface area is king, and beads would be ideal.

My understanding was that the "beads" or whatever ends up in there, were going to be below the floor plate, therefore mitigating the effectiveness of any convection...

Have a good understanding of thermodynamics as such, used to be heavily involved in coal fired thermo-generation plant design, many years ago...

Mal.

[readeral]

"beads" or whatever ends up in there

It'll definitely be beads to start :P

Oh. And to clarify - not below the floor plate to begin with. That'd be a part of the 'second step' if it ever comes to be. I can't be bothered pulling the floor out right now.

In a sense, the way I intend to work to begin with, there will be two 'convection zones'. One within the roasting chamber, assisted by the fan force, and one below the roasting chamber, with largely natural convection (if what I say about that space below is correct - I dunno, maybe I'm wrong there!)
However - given the space between the false floor and the bottom of the roasting chamber is so close, I might be getting a tiny bit of radiant heating (although the beads won't be red hot, so actually, probably won't...) and as the beads heat the base of the chamber, further conductive transfer (from roasting chamber) to the beans. In which case, maybe having some stainless steel balls in there would make the most sense... who knows.

What I know is: I've bought pie weights. I'm gonna give them a crack. I'll let y'all know how it goes :P

I keep bugging my wife about this. She keeps reminding me that she's an Electrical engineer, not a Civil engineer. I keep reminding her that I'm a musician and at least she did 1st year general engineering subjects