I-Roast experiment

[keno]

While I'm relatively new and inexperienced when it comes to homeroasting, I've read about a lot of people who complain that coffee roasted with the i-Roast lacks the depth and complexity of drum roasted coffee (e.g., Hottop). People usually say the i-Roast results in coffee with a sharp (i.e., acidic) bite.

I noticed that the Hottop and i-Roast profiles are quite different. The Hottop starts at a low temperature and ramps up very slowly and continuously over a relatively long time and takes 15 plus minutes to complete a roast. The i-Roast heats very quickly to a higher temp and then tends to level off, with a shorter overall roast time of around 9 minutes. Here is a typical comparison of roast profiles for a few machines (including the i-Roast and Hottop):



From previous experimenting with a defective i-Roast that ran too hot I've discovered that the i-Roast doesn't really do what it purports to do. The temperatures you program aren't the temperatures that are reached in the roasting chamber. In fact, it often gets hotter slowly over time. Also, the onboard temperatures that the machine reports are typically a lot lower than the temperatures measured using a thermocouple inserted into the center of the deflector.

So, I thought I'd try programming the machine to the minimum temperature for the maximum time (that's 325 degrees F for 15 minutes). I measured the temperature with the onboard thermometer and a thermocouple. I used 2/3 cup of Colombian beans (wet processed to avoid chaff buildup that would tend to increase the temp faster). The results are pictured below:



As you can see the result is a lot closer to the lower longer ramp up of the Hottop. First crack occurred at around 9:00 remaining and second crack at around 5:00 remaining. I stopped the roast at 4:00 remaining (for a total roast time of 11:00).

I tried the same thing the other day with some Brazilian beans (but didn't bother to measure the temp since I wasn't even sure the experiment would work). I didn't tell my wife that I did anything different and she commented on how much better the coffee tasted! So, that's what led me to measure the temps and try it again. I'll try to post some comments on how this batch tastes in a few days.

Has anyone else discovered any preferred profiles for the i-Roast?

Cheers,
Ken

[popeye]

I've been using an iroast 2 for about a half-year now. One thing i figured out after a few weeks is that the programmable temperature isn't actually in 5 degree increments - the roaster only has 3? temperature ranges, and will adjust your inputs accordingly. For example (and these numbers are not correct, just example numbers) if you select anything from 360-400, the roaster will put out 380F (or what it thinks is 380). I was dismayed to notice this - it's pretty deceptive - and i posted and confirmed it over at coffeegeek (although i'm too lazy to link that post). I get around it by using a sliding cover over the vents on the back and manually adjusting my temps (although if you do this, you have to seal most of the bottom vent and also some of the crack between the side panel and bottom - air enters through it). This gives me a ~20 degree "boost" that i usually engage sometime around the first crack (by closing more vent). I'm thinking about PIDing the entire thing, because after a half year I can do two roasts of the same bean at as close to the same profile as possible, and still get different results.

[Rainman]

ya know, I'm thinking of doing the exact same thing. All the jabber on this forum talking about controlling temp for 1 min, then at a different temp for 2, etc.. has me feeling a bit guilty, that #1. I have no idea what my iRoast is really doing (I'm just trusting the technology), and #2. I've been too lazy to look into it any more and do something about it. Nobody has really criticized the iRoasts, other than to say that they probably rush into 1st crack and produce too bright a cup as a result... sure seems that fan goes on and off very frequently, and Jim configured the PID on his P1 roaster http://www.homeroasters.org/php/forum/viewthread.php?forum_id=32&thread_id=326 to control in-flow air temp fluctuations to < 5 deg F (he thinks he can taste a difference, but admits others might not..). I would think that controlling the speed going into 1st crack, and maybe improving the cooling cycle (or just dumping the beans into a different roaster to cool - I have an old HW Precision that might work for that) might open up the body in some of my coffees.

Ken- where did you place your TC?

Ray

[keno]

One thing i figured out after a few weeks is that the programmable temperature isn't actually in 5 degree increments - the roaster only has 3? temperature ranges, and will adjust your inputs accordingly.

Yes, I think I've noticed the same thing. Basically there are three fan speeds which control the temperature. The temperature control settings are like the walk buttons on NY city streets - there only to placate people who like to think they have some degree of control!

Ken- where did you place your TC?

The thermocouple was threaded in through the holes in the vent cover and chaff collector on the top of the machine. You have to ever so slightly drill out a hole in each. I then ran it down into the middle of the bean deflector, figuring that this central location would provide a pretty reliable indication of the temperature.

By using a low setting and extending the roast time (to about 12:00 to 13:00 total), I've found the results to be better than the shorter built in programmed roast profiles. A little less acidic, with more complexity. In trying to avoid too much acidity with the programmed roasts I often found that by increasing the roast time that it quickly went too dark and then would taste ashy. It's much easier now to target the right roast level without overshooting because of the lower temp (which still gets well over 400 by the end of the roast).

Motto: Do not rely on the onboard temperature indicator on the i-Roast, it is worthless and wrong. I think this may be the source of a lot of frustration that people have had with the i-Roast. Use a thermocouple to plot your temps and use that info to develop your own roast profile that suits your tastes.

Let me know if you try this and what kind of results you get.

Cheers,
Ken

[ppopp]

I've wondered about this - why the fan sometimes changes speeds when jumping from one temp to another, and other times why it doesn't. So what are the three actual roast temps for the iRoast and iRoast2? Are they different for the iR1 and iR2, since the consensus is that the iR2 roasts cooler for a given programmed temperature.

As my expertise has increased with my iR2, I've found I've progressed to lower and lower temps. Now I'm thinking I should just roast on the lowest temp for 15 minutes and see what happens.

I feel a little duped.

Peter

[keno]

Peter,

The people at Hearthware tell me that the i-Roast 1 and 2 are identical in terms of heating. I think any reported differences you see (e.g., in the figure above) are due to measurement error. The difference is that the 2 allows you to program 5 stages instead of 3 and it will store in memory (when unplugged) up to 10 roast profiles. It also has a few minor improvements to the roasting chamber. As one person said Hearthware should have called it the i-Roast 1.01.

As to the three fan speeds, that is my experience. I'm guessing that the heating element is always on and the fan speed serves to "control" (but I use that term very loosely) the temperature (i.e., lower speed = higher temperature). But, in any case, actual temperatures seem to have very little to do with what you program. For example, if you program 400 for the entire cycle, you would expect a quick rise to 400 and then for the machine to cycle right around 400. Moreover, since the machine is programmable in 5 degree increments you'd expect it to have +5 degree temperature accuracy. But it doesn't do anything like that. My thermocouple measurements show it shoots well beyond the programmed temperature. Even the onboard thermometer shows that it keeps rising throughout the roast and surpasses the programmed temp.

So, by all means try roasting a batch on a lower temp setting for a longer time and see if you think there's any improvement. It seems better to me, but I'd like others experience and input too. Please post a response after you give it a try.

Cheers,
Ken

[wmfamily]

I tried the keno challenge tonight and set my IRoast 2 to 325 for 15 minutes roasting a Sweet Maria's Espresso Monkey blend and was surprised that the roast barely reached a city roast. I didn't have any way to check the bean temperature so it hard to say what's going on based only on visual cues. Of coarse the ambient temperature may have had something to do with it (below 70 degrees) but I ended up shutting it down with 10 seconds left and pushing preprogram number 2 to finish it off. One thing that I did notice was that once it hit the second crack point I did not get the little divots that I usually get. Hmmm.... Once finished though evenything looked good now I only have to wait a bit to find out how it tastes.

To me this is very interesting and will have to try it with some other beans. Maybe something that I don't have to roast so dark.

One question/concern that I have would be extending the roast too much and how that could negatively affect the flavor.

[keno]

Hi wmfamily,

Glad to hear you tried the experiment. Sounds like your machine may not be roasting as hot as mine. Even at 325 (and in a room with an ambient air temperature of about 60 degrees or less - my guesthouse) I could not go more than 12 or 13 minutes with the roast getting too dark (e.g., I start to notice small spots of oil rising to the surface).

A few questions for you: (1) Do you happen to have a voltage tester and know what voltage your electrical supply is? Mine is 120, a lower voltage would account for the difference. (2) What was the highest temperature you got on the onboard thermometer when you press the temp button while the machine is in operation? Toward the end of my roast I get readings slightly above 350 degrees, but the actual temp as measured by my thermocouple is quite a bit hotter (upwards of 450 degrees).

Maybe I've managed to get a second defective i-Roast (this machine is a replacement for another defective unit). I may try running it with an extension cord to reduce the voltage.

Let me know how it tastes.

Cheers,
Ken

[wmfamily]

Ken,

To answer your questions...

1) I've never checked the voltage. Never really had a need to before. I'll pull out the volt meter this weekend and check.

2) The highest temperature registered fluctuated between 350-352.

I checked the beans today and noticed they are not showing any oil to speak of. Normally when I roast this dark the oil shows up a bit but not what I would call "greasy" (that doesn't even sound good but describes the beans I see in the stores).

If I have time tomorrow I'll do a test pull and see what they taste like. We are putting our pinewood derby together tomorrow night and I won't get back until later. Of course right now I'm buzzing on some very good Indian Monsooned Malabar "Elephant" coffee from Sweet Maria's (the last of it for now) so I may risk a quick cup.

John

[wmfamily]

I was finally able to find/make time to test the beans from my little test and have to say I think that we are onto something here. This was some of the best coffee I've roasted for espresso and I have to say it was noticably smoother from my previous roasts from the same batch of green beans. Of course a side effect of this has been that I've had several espressos today and wanted another this evening but my wife was out and I hate to drink alone.

I just finished roasting a SO Kenyan coffee and will test it out tomorrow in a presspot. I have high hopes for it too.

Now the big question is whether the poor IRoast's life will be shortened by this full bore roasting. This evening's roast went 14 minutes before I hit the cool button. I don't know that these things are built well enough to run the fan at full speed for that amount of time.

John

[topcat5]

I had concluded the same thing about my I-roast and have since abandoned it for roasting and now use a SC/TO combo. Sometimes I will use the I-roast if I need to roast some coffee in a hurry or don't feel like setting up the turbocrazy. Since the I-roast's performance is very dependent upon the bean, the ambient temperature, and even the voltage I found that profiles on this machine were pretty useless. However I have gotten my best results by setting it to the lowest temperate for about 5 minutes then bumping it to the max temperature for the remainder of the roast. I just manually go into the cooling cycle when the beans appear to have gotten where I want them.

[keno]

Hi John,

Thanks for posting your results. That's interesting because that's what I found too: tastes smoother and less acidic or bright (which is a common complaint about the i-Roast). It would be interesting to try to experiment a little with this and see if any modifications to the roast profile would improve it. I might try bumping the temp up a little after 1st crack is reached.

Not sure about the durability issue. I think the most important thing there is just to follow Hearthware's advice not to do back-to-back roasts and to keep it under 7 per week (not a problem for me).

Cheers,
Ken

[wmfamily]

I'm going to try a little differenct profile from the 15 minutes @325. I set it for 10 minutes at 325 and the remaining 5 minutes at 400. The 10 minutes brought the beans to the verge of 1st crack and once the temp increased over 400 (on the silly display) it reached 1st crack and then started into second just before starting the cool down which is what I was shooting for. I'll have to let it rest for a day or so before testing but am courious how it will turn out.

BTW has anyone determined about where the fan speed shifts down? I know there are three speeds and it drops to the second speed at ~400 degrees but when does it drop to the third speed?

John

[keno]

Hey John,

Please post about how that roast profile works. I've been thinking about trying something similar, but haven't had a chance yet.

I think the fan speed change is somewhere around 440 to 460. I might experiment with this to try to figure it out. If I do, I'll post the results.

Cheers,
Ken

[wmfamily]

I've let my previous test espresso roast to rest a couple days and I have to say that this method/profile looks promising. The one thing that I don't like is there is not much room for error when it comes to timing. I will continue to mess with the timing and post my results.

As for the espresso my wife is very impressed. She liked the last batch and likes this one even more. BTW this one was Sweet Maria's Liquid Amber blend. Even though we like the other SM blends this is the one we keep coming back to as our favorite.

I also roasted up a batch of my coveted Ethiopian Harar Horse Lot 30. I haven't had a change to taste it yet but will in the morning since I have a little time before heading off to work to share a cup with my wife. I messed up and roasted this batch on the 325 for 15 minutes profile and was worried that I would run out of time before it finished but had plenty of time left (~10 seconds) before hitting the cool button.

John

[steamer]

The 3 speeds will be begining of roast, mid roast change and cooling speed. I have had mine for over year and a half. All coffee and no espresso. Once you like your setting for each origin then it's a breeze to work with. I do use 7 of the 9 for different beans, I leave one for testing.