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

[Tex]

I was using my digital thermometer to log some roast data today and noticed the turbo oven has some serious dead band issues; maybe as much as 25°? I had one roaster that I PID'd to get around this, but it was clumsy to adjust the set point as the roast progressed.

I like the initial temp to be set @400°F for ~11 minutes, then I turn it up to 450°F until I hear the outliers of 1st crack, then it's back down to 400°F through 1st crack. When I hear 2nd crack's first outliers I crank it up to 500°F.

As you can see that'd be a lot of PID manipulation. The analog controls of the CO are simple to manipulate, but there's the problem of the dead band of the mechanical bi-metallic tstats.

I'm thinking that what is needed is a roaster with the elements always on @100%, with a variac to control the percentage of power being applied. I'm picking up a used CO top this week and I'll pull it apart and wire it directly to the variac.

My question is: Should I also set the fan up to be variable as well? A setup like this would pretty much mimic the Hottop, wouldn't it?

[J.Jirehs Roaster]

consider this.. (my current set up)
TO fan is on one plug... it stays on constant..
TO heat is on another plug, dial on high.. it is plugged into a relay controlled by the PID
the PID is a thermometer for the most part, I lift the handle on the TO to cut the heat (but not the fan) if I want to slow the progress of the roast and I lift the lid to see the beans and factor that into the roast curve (all mental/zen like roasting)

the PID is usually set around 430 F so it dose start to turn on and off as I get close but at its current settings I can easily find second crack if I want..

separate the heat and fan, keep the heat wired through the TO handle kill switch and try it.. add the variac with plug and and an outlet if it doesn't work.. I like modular then I can move things around between experiments

[Tex]

consider this.. (my current set up)
TO fan is on one plug... it stays on constant..
TO heat is on another plug, dial on high.. it is plugged into a relay controlled by the PID
the PID is a thermometer for the most part, I lift the handle on the TO to cut the heat (but not the fan) if I want to slow the progress of the roast and I lift the lid to see the beans and factor that into the roast curve (all mental/zen like roasting)

the PID is usually set around 430 F so it dose start to turn on and off as I get close but at its current settings I can easily find second crack if I want..

separate the heat and fan, keep the heat wired through the TO handle kill switch and try it.. add the variac with plug and and an outlet if it doesn't work.. I like modular then I can move things around between experiments

Here's the problem I saw as I was logging the data: The CO controls do not manage power, only the set point temp (ex: if you set the controls to 400°F, the power is on 100% until the tstat reads 400°F - then the element shuts off until the bottom range of the dead bands is reached).

If you're running the CO through the controls, then you're dealing with a thermostat if the tstats preset is reached. And a PID won't change the need to control the power - it's still running the element at 100% to a predetermined set point, then shutting the element off. Granted, it's turning the element on/off in a more controlled manner, but it's also only doing that for one temp. To change the PID set point requires changing the PID parameters, not that simple to do on the fly.

I'm going to wire the element directly to the variac (at house current) and manage the power and fan separately, just like the Hottop. This way I can run the CO at a fixed percentage until I reach a certain temp, then change the percentage until the next temp milestone is reached. It won't be programmable, but it will be very repeatable.

example: 50% power to 350°, then up the power to 75° until 400°, then drop power to 50%, and coast through 1st crack, and when 2nd crack begins run the power to 100%. The same pattern would probably be necessary for the fan.

It's a similar problem we have with PID's in espresso machines. A PID isn't very useful in maintaining intra-shot temps. The PID's strength is achieving predictable and stable starting temp in an espresso machine. Based on what I observed with the TurboCrazy I PID'd, the temp is better managed with power and fan control than by trying to manage temps using a PID's set point.

[J.Jirehs Roaster]

I went down that path you are on and found the Turbo Oven controls to be too inaccurate roast to roast.. bean to bean.. that is why I set the TO on full power and just turn on and of the heat with the kill switch in the handle in the lid.. I am my PID and the temp readings I am getting from the bean mass are my guide to slowing the roast.. the stir crazy heat is off the TO is on 100% (unless I raise that handle, then it is off till I lower the handle back down.. no variac, just separated heat and fan.. I may try and get the PID to work for me some day but for now this setup is working fine...  maybe the key for me is I am in the garage, it has its own electrical service so their is no great load on the system but me roasting

[blzrfn]

I split the wiring on my oven and run the fan directly to the outlet and run the element through a "router speed controller" that I picked up at Harbor Freight for about $20.  This allows me to control the amount of power going to the heating element without buying an expensive variac and also avoids the dead band on the standard thermostat.

[Tex]

I split the wiring on my oven and run the fan directly to the outlet and run the element through a "router speed controller" that I picked up at Harbor Freight for about $20.  This allows me to control the amount of power going to the heating element without buying an expensive variac and also avoids the dead band on the standard thermostat.

I've got one of those that I played with for awhile - varying the group pressure on Gaggia's and Silvia's, rather than adjusting the OPV. I hadn't thought of using it in place of a variac in this application. Maybe I'll use two: one for the element and one for the fan.

The only drawback is the lack of a way to measure the output accurately. I wonder if a Kill-A-Watt hooked up to the output, and before the element, work suffice?

Hmmm...

[blzrfn]

I'm sure it would I just adjust it higher if I want more heat and lower if I want less.  I am using two thermocouples in the roaster, one in the bean mass and one in the air space that I use to adjust my heat.  It isn't hard to reach over and bump it up or down a little if needed.

[Tex]

I'm sure it would I just adjust it higher if I want more heat and lower if I want less.  I am using two thermocouples in the roaster, one in the bean mass and one in the air space that I use to adjust my heat.  It isn't hard to reach over and bump it up or down a little if needed.

Two t/c's...? Why didn't I think of that? That's why I bought the Tenma 72-7712. Duh!

T1 to measure the bean mass temp & log data
T2 to measure air temp and make temp adjustments using variac

What air temp do you shoot for as you're ramping up to 1st crack? I currently preheat to 200°F before I dump the beans. The temp immediately drops to ~145°F before it begins climbing again. This is with the TO's power at 100%.

[blzrfn]

I'm sure it would I just adjust it higher if I want more heat and lower if I want less.  I am using two thermocouples in the roaster, one in the bean mass and one in the air space that I use to adjust my heat.  It isn't hard to reach over and bump it up or down a little if needed.

Two t/c's...? Why didn't I think of that? That's why I bought the Tenma 72-7712. Duh!

T1 to measure the bean mass temp & log data
T2 to measure air temp and make temp adjustments using variac

What air temp do you shoot for as you're ramping up to 1st crack? I currently preheat to 200°F before I dump the beans. The temp immediately drops to ~145°F before it begins climbing again. This is with the TO's power at 100%.

Well I am still doing lots of experimenting.  For instance, last night I removed the "heat shield" to see if I could get faster ramps using some of the radiant heat from the element.  Still not convinced it made a difference but fun trying.  Before I removed the shield I would ramp up to 475-490 and hold depending on the roast desired.  My "favorite" profile so far preheat to 200, dump, beans drop to the 150 area but begin rising quickly.  Hold air temp around 350 until 5 minutes, bean temp usually right around 295-305, then full blast until I reach my final temperature mentioned
above.  Total roast times with a three minute finish from FC to end is around 12-14 minutes roasting a 300-350g batch.

My next modification is to create a way to close the chaff vent so I don't lose as much heat when I don't need to.  Excuse my rambling.

[anastasio]

My next modification is to create a way to close the chaff vent so I don't lose as much heat when I don't need to.

Check out my Chaff Ejector mod on page 21 of this thread, it might be what you are looking for.
Basically, I have a hole in my aluminum spacer covered with a flap of spring steel that I can push open when I want to let some chaff out.
Works great.

Good luck!

[sea330]

Hi, how did you remove the teflon.Rich

[anastasio]

It wasn't easy.
I used a power drill with a sanding wheel, something like this:

80 grit Flap Sanding Wheel - 20,000 RPM - 1" x 3"
http://www.amazon.com/80-grit-Flap-Sanding-Wheel/dp/B002UFVF3E

Get a good mask so you don't breathe it. Be patient.

[Tex]

Check out my Chaff Ejector mod on page 21 of this thread, it might be what you are looking for.
Basically, I have a hole in my aluminum spacer covered with a flap of spring steel that I can push open when I want to let some chaff out.
Works great.

Good luck!

A link would be useful when you're trying to direct someone to a thread or post within a thread - click on the subject of a particular post and copy that URL into your reply.

ex: This will give you richdel's brief tutorial on making pizza dough: http://www.greencoffeebuyingclub.com/index.php?topic=10640.msg163195#msg163195

[Tex]

Here's a clip showing how I measure bean temp.

UFO-CO thermocouple placement

[blzrfn]

I thought I would share a couple of general profiles that I have been using lately which net different results in the cup.  This is just the beginning of my experimentation, but these results are after over 100 successful(ish) roasts using my GG/UFO roaster.  The UFO has the heater disabled and the GG wiring is split so that the fan is on full all the time and the heater element is controlled using a router speed controller.  My batches are all 285g to keep the results more consistent.  I don't have the records in front of me, but the exact numbers are not the important part to me.

First profile - this has generated a slightly sweeter cup with a good overall balance.
Preheat to 350F as measure by the BT probe, let temperature stabilize for at least 5 minutes.
Drop beans in and switch heater to full.
When ET Probe returns to 350F adjust heat imput to coast until 4 minute mark.
Switch back to full until 400F ET, coast until 8 minute mark (BT is usually around 360F at this point)
Switch to full again until 450F ET and allow temp to raise slowly but not exceeding 475F. 
1C is usually around 12-14 minutes and roast is concluded between 3-4 minutes thereafter once I hit my final target BT (usually 425-430) pre -2C

Second profile - this profile tends to accentuate the brightness in the cup and is what I prefer for DP Ethiopians especially
Preheat to 400F as measured by the BT probe, let temperature stabilize for at least 5 minutes.
Drop beans in and switch heater to full.
When ET Probe hits 400F reduce heat imput to try and stabilize until BT Probe reaches 360F (this averages around 8 minutes on my setup)
Return to full heat until ET reaches 450F and reduce heat to a slow rise, not exceeding 475F.
1C is usually around 10-12 minutes and roast is concluded between 3-4 minutes thereafter once I hit my final target BT (usually 420-430F)

These aren't scientific by any means, but the results have been repeatable and enjoyable.  I have tried other suggestions and with modifications have settle on these two profiles as my starting points depending on what I want from a bean.  The second profile may be too much for a softer, lower grown bean, but I don't know for sure since all of my current inventory is high-grown, hard bean coffee.  Feel free to try this out and comment or make any suggestions or questions.