SC/TO Profiles / Q&A / Tips & Tricks

[peter]

Any idea how much that batch weighed?  You should be able to do 400g pretty easily.  Maybe the beans were getting wedge between the arms and SC bed?

While the t-stat actually does function like a t-stat, the heating element is either on or off.  So you can forget about trying to set the dial by temps, and just turn it up far enough so that it'll stay either on or off.  You'll get the hang of how far the temps will coast upward when you turn the heat off, and how long it'll take for the temps to start rising when you turn it back on.  11min. to 1C isn't all that bad.

Please remind me how your temp probe is set up.

Did you open and close the chaff chute to either let heat out or hold it in?

[RamzyB]

Just weighed out equivalent of the amount I roasted, and green it was about 350 grams.  I didn't notice any beans getting wedged.  Should the stirring arms be very close to the pan?  As I made them, there is a gap.

I am using a CDN model DTP482 temperature probe through a hole I drilled in the chaff collector.  I set it up to lay the point down into the beans at as shallow an angle I could.  The probe goes bump bump bump as the stirring arms move under it.

I did open and close the chute on the chaff collector.  I generally kept it closed until I saw chaff zooming around, then I opened it and the chaff came out.  I kept that open/close thing going until I dropped the temperature to 350 and then left the chute open.

[peter]

My suggestion would be to have the arms as low as you can w/o actually rubbing.  You may just have a wimpy motor; I've noticed some like to reverse more than others.  But 350g should be no problem.  Or, there may be some friction along the shaft somewhere.

350g should be enough to keep the tip of the probe buried fairly well. 

I don't think you have to roast at a lower temp necessarily to get a lighter roast level - just end the roast sooner - unless you shoot for that amount of time after 1C for flavor development.  You'll have to factor in the amount of time it takes to vacuum out the beans too, because as you know, they're still climbing in internal temps until you can start cooling them.

[mrjay]

My latest fix to the SC shaft

Previously I had lopped off the plastic "bolt", drilled and tapped for a 6mm bolt.  That left a weak section in the center of the shaft that in hind sight, I probably could have prevented today's failure by judicious use of hi-temp JB Weld.

But alas, I didn't and (luckily) the shaft failed just as I was beginning a roast.  My latest fix involved cutting the entire center out of the plastic shaft drilling/tapping the shaft bottom and tying it all together with a 6mm all thread and a couple of nuts.  I now use a 6mm wing nut to hold the stir rod to the shaft and cover that with a few wraps of aluminum foil and a 1" copper cap.  The attached pic should explain it better than what I just typed.
This was a quick and simple fix and I hope it will last a while.

[stevea]

Hi, I made my first batch in the SC/TO.  I used 1 and 2/3 cups of Kenya Thunguri.  I tried to follow the roasting profile given early in the SC/TO thread, but the thermometer probe seemed to be measuring more of the air temperature and not so much the coffee temperature.  Total roast time was 17 minutes and it went into first crack at around 11 minutes and it was truly slow.

Ramzy - like ~7-8- years ago, when I first started playing w/ an SC/TO (invented by an SF guy named Suzuki IIRC, look around coffee geek ~2007/8) ) I did a long series of experiments.   My *OPINION* is that any hardware than can't get you to 2nd crack in ~7 minutes has inadequate power and will likely leave you with "baked beans" flavors.  Sivetz (the late author of "Coffee Technology") recorded that commercial large scale roasters operate in ~6 minutes.   By "Baked bean" flavors I mean that you end up with a chocolate-y simplistic "McDonalds coffee" flavor regardless of the varietal character.   There is absolutely nothing wrong (and a lot right) with creating an 8  or 9 minute profile schedule that allows for more delays for browning reactions and maillard reactions on the +- side of 1st crack, *BUT* you need to be able to power your way up hard to 1st crack and then out of the rests and into the edge of 2nd crack fast.

Long slow profiles (like your 17min) kill off the caffeine (and what is the point of this coffee obsession w/o caffeine ?) and ruin the natural acidity of the bean.  So you get a good deal of roasty-chocolate flavors and no "bite".  IMNSHO there is absolutely no justification for any profile over 10-11 min.

So I've always used/use a common SC unit (with a rake), and a Galloping Gourmet 1400W TO unit (the two will blow a 15A breaker !) and for years I've used a 2.25 inch springform pan as spacer .  I pre-heat the unit (no beans) for 2.5-3 minutes, add room temp beans and am seeing 1st crack in 4-5  minutes, 2nd crack in under 8 minutes UNLESS I add delays.  .  I have measured results and I believe that this combo will roast 250-600grams of green beans adequately, but the sweet-spot is 350-375g of beans.  350g of green beans (my standard load) is ~2.4 cups of greenss.  My experience is that tiny loads like your 2/3rds cup (perhaps 110g) do not roast evenly and *should* incinerate in ~7 minutes!

My suggestion for your next roast.
0/ Pre-heat the unit to operating temps. *ALWAYS*; this is a basic pre-req.
1/ bypass the controls and try using full power/no-power and your nose as instrumentation.  (Wax -on Wax off, sensei; once you understand - then use automatic controls to do your bidding
2/  Run the beans to the early edge of 1st crack <5min, then stop the power for ~60 seconds.
3/ Then re-apply full-power till you decide they are finished (maybe the early-edge of 2nd crack for city+)
99/ Cool beans rapidly!

You can't get a "bright" flavor at 17 minutes or even 12 minutes.  10-11min is the upper bound IMO.  After that you are "diddling your own beans" and killing the acidity.
((NB: a LOT of owners of very expensive roasters will be unhappy w/ my comments))

My *SUSPICION* is that you have an underpowered TO.  I've asked repeatedly on this forum and elsewhere and no one wants to answer, but my expectation is that some of the 1200W "Suppernhaus" units lack the power to roast 350g of  beans in a rapid time.  Decrease the load, but increase the heat containment.

  I initially set the TO at 300 to preheat, then 400 to roast.

Big mistake =- set it to "Heart of the Sun" until you approach 1st crack.

  After first crack began, I reduced the oven to 350.

That's fine, but you'd better be at 1st crack in <5min.  Between 1st & 2nd you are burning off "brightness" at an alarming rate.  Plan accordingly.

The stir arms kept reversing direction as someone had warned they might.  That doesn't seem to hurt anything as long as they keep moving, but it seems like I am at the upper limit of how many beans I can put in, and it would be nice to put more in so the temperature probe is buried in the beans.  I'd like to find out more about a more powerful motor that someone had mentioned.

When you have fewer beans the stir-arm is FAR more-erratic.  IMO DO NOT try anything less than 200g of beans in an SC.  \*WEIGHT* the input beans.

There was hardly any smoke compared to the hot ai huge  poppers.

Yep - and the roast level is slightly more uneven (which is a huge ADVANTAGE!)
We really *WANT* some beans on the more acidic unroast edge and some farther into the *$ bold well-roasted edge to make a good mix.  You can't achieve that on some instrument  of perfection and science.

Welcome to the Art of Roasting !

[RamzyB]

Thanks very much for this thorough reply.  I have in fact, lost some of the brightness, and it's good to get pointers on why.  I have been roasting 350g of green per batch, but apparently I need to increase the heat at the outset.  I will try the TO at full heat for the beginning.  Do you also recommend using the SC heater element?  I have it wired separately so that I can turn it on/off.

With regard to the metal shaft I made, I was unhappy with my wire arrangement as it was not stirring very well, so I put the original shaft back in and have been using the original stirring wire.  That, however, leaves beans unroasted out at the perimeter.  I saw a picture of Peter's stirring wires, and that design looks like it would keep the beans moving in under the TO element better.

Temperature data point:  I have an infrared temperature sensor and it measures bean surface temperature within one or two degrees of the temperature probe, so I am satisfied it is buried deep enough into the beans to trust.

Also, when the SC heating element is turned on, I noticed some beans being dragged around on the flat side and burning while the top side was green.  Not good.  I either need to not use the SC heating element or figure out a way to agitate and turn the beans.

This concept of striving for an uneven roast:  If I allow the beans at the perimeter to remain pale brown and roast until the beans in the center are smoking, is that desirable?  I can try a batch that way and see if I get the flavors back.

One reason I went to the SC/TO was that I had been using hot air popcorn poppers and the flavor was not what it should be.  I was advised to try the SC/TO as it would be a longer roast and retain flavor.  I would love to hear opinions on different roast strategies.

[RobertL]

My *OPINION* is that any hardware than can't get you to 2nd crack in ~7 minutes has inadequate power and will likely leave you with "baked beans" flavors. 

Wow! 

[grinderz]

Be advised that stevea's preferred roast profile deviates a bit from accepted "norms."  But that's cool. Not everybody likes the same thing.

[peter]

Be advised that stevea's preferred roast profile deviates a bit from accepted "norms."  But that's cool. Not everybody likes the same thing.

Maybe he'll play along in the next swap.

[sea330]

Coffee swap using only SC/TO roasting method  would be fun, maybe Stevea  would  jump in,  I have so much roasted up now I need to wait a week before roasting up some,  I will roast up 350g just for SG using his method,  most of my roast run 12 to 14 min. With my 60 rpm drive motor mod, I’m getting very even roasts,  some of the naturals will look a little uneven but that is expected. 

[RamzyB]

I would like to hear more about the 60 rpm motor.  Type, source, any issues with installation?  Does it have more power than the SC motor, or just go faster?

As I have been getting more experience roasting with the SC/TO, I have been taking the beans into first crack (7 minutes or so time), then removing the TO while the SC heater is fully hot.  I can then see the beans as they darken, and nudge beans from the perimeter into the center, and all the beans begin cooling in the absence of the TO on top.  Maximum temperature hits around 426F but drops to around 406F (11 minutes or so) as the first crack popping tapers off.  I have roasted both the Chelfit and oaxaca coffees this way and got excellent finish tastes as the cup cools.  Beans go into the cooling strainers (fan blowing up from below) looking somewhat mottled, but by the time they are cool, they look a lot more even.  I don't think I am hallucinating that...but who knows?

thanks,
Ramzy

[stevea]

My *OPINION* is that any hardware than can't get you to 2nd crack in ~7 minutes has inadequate power and will likely leave you with "baked beans" flavors. 

Wow! 

Read that carefully - I am not suggesting a 7min roast is best, but you need that much power to ramp at a proficient rate.  OTOH commercial air roasts are faster !

Be advised that stevea's preferred roast profile deviates a bit from accepted "norms."

FTFY 

  But that's cool. Not everybody likes the same thing.

Of course.   Keep in mind that I mostly prefer ethiopians and s.american beans so a sidamo is about as acidic as it gets - I'm trying to preserve acidity.

FWIW SM's defines "baked beans" a lot differently than I do.   As in under-roast or "case hardened" (toasted on the outside, raw on the inside), therefore "sour" (acidic) and poor flavor.   My issue is that long roasting time remove the caffeine bite and the acidity and the fruity/winey varietal flavors - leaving you with something a lot closer to cocoa than coffee (I haven't had it in years, but that's my recollection of MacDonalds coffee).

I may be extreme - but I've never tasted a 13 minute roast to FC+ that I thought was anything but inferior to a <10min roast.  Time is a double edged sword - you gain caramel and maillard products esp in the lower temp range, and some desirable carbonization at the upper - but lose acidity and caffeine and varietal flavors.

[sea330]

I permanently mounted my TC no problem, the only thing is it needs to be long enough so you don’t need to disconnect. I made a 4’ 20 gauge K type TC just for this application. Works perfect.

[antoine_t]

any TO/SC users still?

I picked up a combo a few weeks ago and just go into roasting


i want to buy a variac, either staco or powerstat (superior electronics). 

anyone d/c their heating element and connect it directly to the variac?

if so, were you happy with the results?

[RamzyB]

I still use my SC/TO.  It plugs into a standard (US) 110/120 VAC outlet.  I have the heating element on the SC wired to a separate switch so that I can turn it on and off without affecting the stirring motor.  I mostly leave this bottom heating element off, as that gives me a better roasting profile targeted at first crack around 9:30 into the roast.