readeral's KKTO

[Postmaster]

Hi Al,
Glad to see that your still having a crack (pardon the pun.)
I have built a different KKTO. (A bit different to the normal KKTO) More like a stir crazy.
In regard to my original thread re the Kmart turbo oven, I agree that it is under powered.
I put another plate made from a piece of metal from an electric stove grill plate (non reflective)
I also ground back the contact stop so that more pressure is placed on the points to keep them closed (not sure if this has made any difference)
I can now get it 220 deg C. So I hope this is enough.
I need to learn how to roast before I can see if its a success or not. Did try a roast, but I learned that it was a mixture of beans. Need to try one type first up.
I have successfully roasted popcorn. Heated to 220 deg dropped in corn. Temp dropped to 160 deg. Keep temperature between 180 deg and 200 deg and got great result. Corn popping everywhere.
Next step is Coffee beans
Regards
John

[readeral]

Oh, also, every single time I heard the cracks. This was a revelation. I don't think I'd really communicated how underpowered these TOs are, but honestly, now knowing I can actually hear the cracks when approaching with enough energy is very exciting.

[readeral]

So, the readeral TO has been modified - basically along the lines of http://coffeesnobs.com.au/roasters/3...urbo-kkto.html with an SSR in the place of the thermostat. It was a VERY simple process. I ducked out to buy a switch from Jaycar and he was done before I got back. Thus no photos.. :P

We went ahead and connected up his TC4 to test it, and the result was for the most part stunning. We did 3 roasts, the first was a Guatemalan, the second a Vanuatu (I think) and the last a Nicaraguan. Sorry I don't have any profiles - they're all saved on Adi's computer. Hopefully he emails them to me.

- The first roast (mine) had a preheat of the chamber to 210 degrees, turning point of 60 and was done in 15 minutes. Dropped at a light roast soon after first crack. Mostly because we weren't sure how fast we were heading into 1C (more on that soon)
- The second roast (Adi's) similar time and similar drop, again, we weren't quite sure where we were.
- The third (Aaron's) we pushed hard. Preheat of 270 degrees (I went inside while this shenanigans played out - otherwise I wouldn't have permitted it!) and it hit first crack in just under 9 minutes, rolling into 10 minutes, and dropped at the first snaps of second crack at a little over 12 minutes.

One thing that was apparent was that because of the (much) shorter roast times (10+ minutes shaved off) the evenness of the roast was diminished. This was partly because we didn't know how the roaster would behave. We had the power setting on 100% from the get go, but I think a slower drying/ramp would probably help even out the roast.

We had an issue where the thermocouple I normally use is a grounded thermocouple (one of Andy's for the DMM) and so we were getting erratic readings from the Arduino. We tried to shield it with some heat shrink, but I'm still not 100% sure it was operating quite right, as the Arduino was suggesting that the thermocouple was picking up temperature drops far faster than I recall of it's actual ability, given it's sheath diameter.

I also think maybe the temp readings through the Arduino were about 10 degrees off, as we were dropping the light roasted beans at 212, which would to me have been more like a medium roast. I also tried a probe down below the roasting chamber, which interestingly after the drop from preheat, didn't get much above 110 degrees. I find that quite interesting, but I haven't worked out quite what it means yet.

So the next steps is to re-assess the accuracy of the thermocouple and make adjustments as necessary, and then go ahead and re-learn this far more powerful system!
I recommend doing an SSR and switch (electronic or manual) mod to anyone trying to use a Kmart/Aldi turbo oven. I don't know how to do it - I won't tell you how to do it, find an electrician mate to do it for you, but I promise it's worth it. Frustrations with the KKTO essentially gone.

Another thing I discovered when we had the turbo oven apart the other day was that there isn't any exhaust from the system at all. Where previously I thought that the air was able to escape through the TO unit (hence the vents in the side of the unit) - but actually that's just to circulate air around the electronics to stop them overheating. The KKTO is still essentially a closed system, which means all smoke and moisture remains in the system. I'm even more convinced that I need to generate some air transfer, either by widening the gap in the silicone tubing, or some other means.

After that, I need to determine the best place to have an ET probe - I'm thinking probably above the bean line poking out of the perforated chamber - however I do know that's the direct path of the heat from the TO (down around the inner edge of the pots and back up through the middle). I'd prefer have one hovering in the middle of the air somewhere, as that would be more accurate, but alas I am not a magician.
Does anyone have any recommendations on where to put the ET?

[readeral]

So back in Sydney - did a roast on Tuesday and although I didn't hit my target temperatures easily, I managed to juice it out of this Turbo Oven.

Managed to hit first crack with both my types of beans - but my mileage varied beyond that point. The best way for me to manipulate the temperatures was to lift the turbo oven off the pots, thus cooling the little sensor down so it returned to it's 'On' state quicker. Obviously though, that lost me a lot of energy in hot air and moisture escaping. Not a long term solution, but did ok in my desperate state of no roasted beans.

One thing I have been observing with this Turbo Oven is that once I hit 130 degrees C in my bean temp, the RoR starts to dive from ~10 deg/min, and by the time I hit 150 deg C it's usually at or below 5 deg/min. This is probably a function of the deadband - but such a dive over 2-3 minutes means that the RoR stalls - it basically doesn't get back above 6 deg/min after that point. Useless. Ideally it would drop at a more constant rate, and hit 6 degrees more around 180/185.

So today my buddy and I pulled the Turbo Oven apart. Gave it an opportunistic clean, and had a look inside at how it's built.

No electronics in this one - mechanical timer, mechanical bimetal adjustable thermostat, and the handle cutout switch. That and the element and motor and that's the whole device. Was interesting to see the fan that pushes hot air out before it reaches the electrics above it. I always thought that was a sort of air outflow from below - but really it's just pushing out any air that might have consequently got hot above the element.

Additionally, we turned it on with the element guard off just so I could feel the airflow - it moves right across the glass, and basically none of the heat goes down. Makes sense given the angle of the fan wings. Although this is great for a gentler heat, it means that KKTO users lose heat across the exposed glass, and also lose heat in the unwrapped edges of the roasting chamber (well I do, I have only wrapped the outer pot).

So... what are we to do:

The first thing I'll be trying is to roast with the element guard off. Risky, yes, but I'm curious to know how terrible that thermostat really is, and if radiating heat from the guard does make a difference. If it does make a difference... Next my mate who is trained as an electrician is going to pull out the thermostat, and fit an SSR and an on off switch so I can manually control the element instead. After that, we might consider hooking up a TC-4 controller after I make a trip to the US later in the year.

[Shano592]

Hi Al, I think I have the same kmart TO as you do. I am finding almost the exact issues that you are.

I reckon I will probably end up chasing a 1400W turbo oven as well.

Still, I'm liking that this KKTO is capable of roasting so much in one go.

[readeral]

So did you get your hands on the new turbo oven? Looking back through your thread, our roasters look very similar. I'll be curious to read of your experiences when you roast again. Only 'issue' with 700-800g batches is managing to drink it all so you can roast again

Haven't yet, no. I'm in Hobart for summer, go back to Sydney in a week so will get a hold of one then. I honestly wish I had access to a 1400W Rank Arena, as I like the light. Also I've been staying with the in laws who have been enjoying daily coffees, so moving through the beans has been no problem Ultimately I want to be roasting for friends, so TBH if I have to consider 500g charge each roast I would get quite frustrated.

I looked at the rating on my TO and it says "1200-1400W". How there can be a 15% variance I don't know, but it certainly seems to me that it operates more like a 1200W than anyone else's TO seems to. It's been run in 3 different houses, and seems ill equipped to push past 1st crack at 600g on all circuits. It could still be the deadband issue. I'll do an experiment to determine how broad the deadband is on it, but have other priorities right now.

I'll also try the changes to the heat guard as that could be a design flaw causing more trouble for me.

[Dimal]

Only 'issue' with 700-800g batches is managing to drink it all so you can roast again

Never found that to be a problem really...

Mal.

[Pretzal]

So did you get your hands on the new turbo oven? Looking back through your thread, our roasters look very similar. I'll be curious to read of your experiences when you roast again. Only 'issue' with 700-800g batches is managing to drink it all so you can roast again

[Dimal]

Yes mate, it was a solid element type.

Exactly the same as the one pictured as it turns out - 1.4KW. I thought it was only 1.2KW (been a few years) so went out back and dragged the box out to have a look. Never had any issues getting up to temperature with this unit. My only problems were not being able to see what was happening in the bowl or hear first and second cracks. Didn't have a CS Roast Monitor back then though, otherwise I might have persisted using it and just rely on the profile...

Mal.

[readeral]

Thanks Mal, Bosco_Lever has got me on to a Tiffany at JB Hifi. Last I looked it was solid element, which is backed by the picture they have on the web, but as Bosco says, they just reuse images all the time.

With your brief foray into the KKTO, did you have a solid element one? Any thoughts?

[Dimal]

G'day Al...

If you can locate a Tiffany branded one, they are pretty decently made and not expensive...
Tiffany OVT01 Reviews - ProductReview.com.au



Mal.

[readeral]

Ok - so I think after Pretzal's thread, and discussions with others as well as a lot of reading of this and other forums, I've decided that my Kmart Turbo Oven must be significantly underpowered compared to others.

I had the impression that although Turbo Ovens were varying in power, that they'd be mostly the same. So the strange things I've been posting in this thread about control and etc. are really symptomatic of my desire to get more juice out of the system in the first place.

With a 600g batch, I barely push past first crack - certainly couldn't get to second crack without nastily baking beans, or on the flip side, charring from cranking it too early (which is what I have been experiencing with the softer beans). It means I don't really hear a first crack as there is no energy there to draw it out.

So.. I'm on the hunt for a new unit.

The Kmart one might just be a dud unit - I don't know, but I'm not inclined to spend another $40 to find out. Unfortunately I just moved house, so all the extra bowl etc. got discarded before the move (so I can't just use it for it's original purpose anymore.)

The Aldi ones seem to be the same unit rebadged IMO, so don't really want to go there either.

Does anyone have recommendation for a 1400W TO? I've tried to find a Rank Arena one, but they don't seem to be available anymore. Rank Arena now have a new style of turbo oven. Other recommendations would be welcome!

My current one will be subject to experimentation, but I don't want my future roasting to be halted because experimentation didn't work out.. so the goal is to get a new TO first.

Al

[readeral]

Also - for the record - I'm not sure my TO will ever push to something like 220 degrees (not that I'm personally interested in getting there) with more than 600g of beans, just don't think it has enough chuff - so control at the end of the roast is more about trying to milk more out of it than tone it back (a la Pretzals difficulties).

[readeral]

Gosh it's hard to find information on packed bed roasters! My comparison was only in context to the book I was reading, and the comments Rao makes on roaster types. He omitted packed bed due to, I expect, (according to the googles) it not being a type used in the specialty scene.

Really what I'm doing is trying to work out how I can reduce roast times, while not charring my beans, which is the idea behind the thought experiment. High temperature and optimum batch size will reduce those times, but it doesn't solve the defect problem. Basically I'm keen to increase my Δt, as I really struggle to have good roast development without pushing it darker than I'd like.

I should experiment with cold drop - I'm not sure I've actually done one yet, but that would probably be close to what I'm going for.. I'll do that next time and see how it goes.

Warning - ahead is more wild conjecture - confessing this up front.

However, yesterday I went ahead and tried out my idea with a 700g roast of a columbian, large beans but evenly graded. Figured it would be easy to pick roasting defects with them being so even and easy to roast. The approach of preheating without the roasting chamber wasn't nearly as difficult as I had anticipated - the only technical drama was that I'd tested the rotation of the motor prior to kicking everything off, and didn't think about setting back up the alignment for inserting of the inner chamber once all was preheated. That's ok, a bit of a wiggle and all was fine.

So I sat the TO directly onto the lip of the outer pot (after carefully ensuring all prior chaff was cleaned out) and preheated it to 200 degrees or so - wasn't exactly sure of the temp, and the drive shaft hole didn't help any... I took off the TO, threw in the roasting chamber and started it off at 225 on the TO dial.

The roast seemed to start quite smoothly - I didn't have a turn around metric as my thermocouple was attached to the inner chamber rather than the outer, but I chickened out and didn't set the roaster on maximum 250 degrees initially. In hindsight, it's something I definitely should have done because my RoR dropped below 10 at 6:20 when the beans had only hit 120 degrees. This was my first alarm bell.

The roast progressed toward 200, but the RoR dropped below 2 at about 195 degrees and approached zero, which was not ideal at all, and I eventually hit 200 around 27 minutes and entirely ran out of steam 2 minutes later, dropping the roast at 203 degrees. Can't say I was all that pleased.

The result though was a very nicely even (if light) roast, the inside of the bean a nice even colour to the outside so had developed nicely, and the aroma of the beans was much fruitier and actually perceivable than I have so far encountered from a roast in my KKTO.

But - what I learned was that indeed my tipping/scorching was being caused by the too hot surface of the roasting inner chamber (This wasn't a purely scientific exercise, I must confess I did change beans to the Columbian as I had run out of the Brazil I'd been roasting) and that I could successfully obtain a similar ramp profile to my prior roasts without the downside of the roast defects. Alas my TO was not powerful enough to carry 700g of beans into a reasonable drop temperature, and I had to pull them out before I started baking it, so I think I have learned that my roaster is happiest in the 600-650g range (which is where I've been sitting most of the time for all my previous roasts) - this could be different for a cold start, and certainly due to the large size beans could be different with a smaller denser bean (in either direction).

So either my mixing action has a problem, but KK you've seen my videos and it seemed fine. Still seems fine to me when I look into the roaster. OR I'm pushing the boundary of what is a suitable preheat without damaging beans. I feel that preheating the outside chamber is a good compromise - but it's definitely a compromise... I think next time I muck around with this idea (probably a month or so) I'll preheat the outer pot, then will throw in the inner chamber for a minute so I can get a turnaround reading, and then throw in the beans, and definitely stick to maxing out the TO temperature dial.

One other thing I was wondering is how airflow affects the performance of the Turbo Oven, and whether bleeding a little bit of air out of the system toward the middle of the roast would encourage the TO to cycle on more regularly, inputting more energy into the system (rather than just trying to maintain the set level). I don't know. I know it seems counter-intuitive, but maybe it'd work?
I also wonder whether (as Mal pointed out in a thread today, and also one from 2009...) the higher relative humidity inside the chamber might impact the ability of the system to sufficiently remove moisture from the beans - that is, is all the moisture retained within a nicely sealed chamber actually stopping the bean temperature rising beyond a particular point?

So my next experimental step will possibly be to play around with the silicone hose that I have sitting between the two pots. I might carefully (avoiding burns!) roll the hose up at 6 minutes to encourage a bit more air flow (won't be much) out of the system and to hopefully draw a bit more output from the TO.

All these are just thoughts - I am highly likely to end up right back where I started, in the footsteps of those who came before, and roasting much like Bosco and KK suggests. However - I'm never one for following something that works and not asking questions - I always want to know why it works and how I make make it work better. *shrug* I can't help myself. Always a student, never the masterer.


But all of this experimentation aside - I really do need to find a bog standard approach to using my KKTO.
I tried charging at 750g, but I think that's too much - I'll stick to 650g from now on, and I think I'll try roasting from cold, just sitting it on 250d the whole way and see where that gets me. If it's satisfactory without roast defects, I'll be a very happy man.

[Koffee_Kosmo]

Technically speaking the most descriptive design features of the KKTO is
Heat = Forced Fan " Convection"
Roast Chamber = Would be closest to a "Packed Bed" design ( without the turning bowl & agitator used instead )

I have conducted many many trial experimental roasts and most turn out good To be on topic with drop in heat temps I have experimented with Cold drop - works equally well with small and medium batches producing a slightly longer roast but with full body flavour
So the method is Drop in cold Turn on the TO to 250 C and only turn down only " if required" after first crack

Some KKTO builds may be more thermaly stable than others so the owners need to make small adjustments to suit ones own roast style and eventual taste they are trying to create
There is no right or wrong Keep in mind that the key things that make this roaster work so well is
1) Roaster Chamber Volume
2) Heat - Convection type - and with preferrence to the 1400 Watts models
3) Agitation - good constant mixing "Like a Rolling Stone" agitation - not a large mass of beans pushed around the roast chamber ( It took me a few months to get this right so agitation is very important ) Refresh and adjust agitation as needed by visiting the YouTube videos I have posted - there is one for a double solid blade with springs, and also one with the early model wire style

Hope that helps
KK