This is a multi-part message in MIME format. Yesterday I made the chaff collector modification described at the end of SM's tip sheet to my I-Roast. I wasn't having a problem with my roasts being too dark. (I'm not quite sure what "too dark" means. You can set the temperature as low as 325 degrees. Are some people getting too dark of a roast even at 325 degrees?) However, in the hopes that the modification would result in more consistent results, I gave it a try. I observe the following changes: The good: a.. As promised, the chaff collector is now clamped in and doesn't rattle. b.. No chaff that sneaks out. (But then there wasn't much escaping before the modification either.) The bad: a.. The lid doesn't clamp tightly onto the roasting chamber anymore. Not sure if this is significant or not. This might be a plus for those that were having problems with the lid being too tight. I suspect this will become more of a problem with use. b.. The gasket falls out of the lid now. The gasket nipples don't hold in the recessed holes. Not a big deal. c.. The modification makes a huge difference in the display and bean temperature (as measured by a TC). Before the modification, the display temp would read about 40 degrees than a set temperature of 350 F and about 80 degrees lower than a set temperature of 450 F. After the modification, I'm getting about 70 degrees lower than a set temperature of 350 and about 130 degrees lower than a set temperature of 450F. Using Tom's slow roast profile (2:00 at 340 F, 3:00 at 400F , 6:00 at 450 F) I reached City at a total of about 8:30 (Kenya Kora, line voltage 117.5 V). After the modification I barely reached 1st crack at the end of 15:00. I would never have guessed that what seems like a minor modification would make such a major difference. Is this difference because the chaff collector is tight or because the lid is now loose? Fortunately, the modification is easily reversible. -David |
I also made the modification, but now am now having a serious problem that
may or may not be related. I actually think my heating coil is giving out.
Background: I also had a drop in temperatures after the modification, but it
was good in that it put my I-Roast on parity with Tom's roasting profiles,
with the only difference being that my machine was now generally 50-60F less
than the set temp (vs. Tom's 40-50F). Close enough. (So David, no, until
recently I didn't see the large temp declines...until...)
However, lately I had my first sour roast (With Kora Peaberry), so I started
profiling the temperature each minute, and here's where it gets goofy. Over
the past three roasts (all post modification), I went from a first crack at
5:46 @ 360 degrees (internal I-Roast temp) and a peak temp of 410 at 9:05
(stopped here). This seems normal, but was still a little sour, so I wanted
to bump up the second stage to 425 from 400 (my new profile was 2 min @ 350,
3 min @ 425, 6 min @ 460). Unfortunately, in the two most recent roasts, I
now cannot ever get over 347 degrees even with the temp set to 460. I even
thought I had mis-programmed it the first time, so I re-programmed the
settings again but the temp curve was exactly the same, looking as if it the
coil wasn't heating up over 350.
Here's my latest run:
Set3 Onboard Vent (exhaust air temp)
1 350 300 172
2 350 307 234
3 425 320 268
4 425 325 288
5 425 330 302
6 460 335 311
7 460 335 316
8 460 336 329
9 460 342 329 (Up to here, the onboard temp looked the
same as the previous batch, thought the Vent temps averaged 17.7 degrees
cooler...very strange)
10 460 345 351 (* See note below -- I plugged up 1/2 the
exhaust vent)
11 460 358 367
12 460 360 375
*Note that when I saw I was going to have the same temp profile (and not
wanting to waste another batch of Kora Peaberry!), I covered half of the
vent with a towel. Note that while the exhaust temp increased 46 degrees
from 329 to 375, the onboard temp only went 18 degrees from 342 to 360. I
had been getting onboard temps of 360 (and first cracks) at 5:46 before, so
I know my unit was previously capable of much higher temps.
Is anyone else noticing temperature declines like this? This doesn't seem
to be modification-related since I had good batches after the change. I have
only done 16 batches since July 2nd, so I'm curious if others are finding
that this temp drop is just the "breaking-in" process, or if it's just
broke. I'm going to switch to less-prized beans for future testing (testing
pre-sets, undoing modification,?).
Craig ("glad he ordered a second 5lbs of Kora Peaberry") Wichner
From: homeroast-admin
[mailto:homeroast-admin] On Behalf Of David Altekruse
Sent: Sunday, August 01, 2004 12:09 PM
To: homeroast
Subject: +I-Roast Modification
Yesterday I made the chaff collector modification described at the end of
SM's tip sheet to my I-Roast. I wasn't having a problem with my roasts being
too dark. (I'm not quite sure what "too dark" means. You can set the
temperature as low as 325 degrees. Are some people getting too dark of a
roast even at 325 degrees?) However, in the hopes that the modification
would result in more consistent results, I gave it a try.
I observe the following changes:
The good:
* As promised, the chaff collector is now clamped in and doesn't
rattle.
* No chaff that sneaks out. (But then there wasn't much escaping
before the modification either.)
The bad:
* The lid doesn't clamp tightly onto the roasting chamber anymore.
Not sure if this is significant or not. This might be a plus for those that
were having problems with the lid being too tight. I suspect this will
become more of a problem with use.
* The gasket falls out of the lid now. The gasket nipples don't hold
in the recessed holes. Not a big deal.
* The modification makes a huge difference in the display and bean
temperature (as measured by a TC). Before the modification, the display
temp would read about 40 degrees than a set temperature of 350 F and about
80 degrees lower than a set temperature of 450 F. After the modification,
I'm getting about 70 degrees lower than a set temperature of 350 and about
130 degrees lower than a set temperature of 450F. Using Tom's slow roast
profile (2:00 at 340 F, 3:00 at 400F , 6:00 at 450 F) I reached City at a
total of about 8:30 (Kenya Kora, line voltage 117.5 V). After the
modification I barely reached 1st crack at the end of 15:00. I would never
have guessed that what seems like a minor modification would make such a
major difference. Is this difference because the chaff collector is tight
or because the lid is now loose?
Fortunately, the modification is easily reversible.
-David |
Craig, The temperatures you posted are similar to what I was getting after the SM tip-sheet modification. But I have no theory as to why you are only now having the same huge drop in temperatures I had. If the air leakage theory I posted in my updated post is true, perhaps your seal was barely OK after the modification but just now settled in enough that it is no longer making a good seal. Seems like a stretch, though, doesn't it? None the less, based on my experience once I undid the modification, I'd still be tempted to undo the mod and see what happens. (Note, though, that it isn't all that easy to get the gasket nipples to fully seat in their original holes.) If you don't have any other ideas, you don't have much to loose to try it. -David <Snip>http://sweetmarias.com/maillistinfo.html#personalsettings |
Well shiver-me-timbers...Preliminary test: I un-did the mod, and using new beans (a confounding variable), had the same too-hot roasts that I did before, meaning that it looks like my heating coil isn't failing and that there was likely to have been a leaky seal, as you suggested. I'll have to try again with the Kenya Kora Peaberry to be sure, since the beans I used had lots of chaff which definitely heated up the chamber toward the end of the roast (ever heard of Hanghibetta Estate??? Got it free at the SCAA meeting so it seemed like a good test pilot...ie Free). As a possible warning to those who did the chaff modification, I was rather careful with the seal, so this might serve as a good warning for extra vigilance. The numbers for my last four roasts are below. The narrative is 1, 2, and 3 all used the modified chaff collector and the same bean. Roast 1 worked, though was a tiny bit sour. Roast 2, 3 didn't work, and the temp readings cluster closely together (though the longer roast time of #3 seemed to make a surprisingly passable, if uninspiring, caramelly and chocolaty cup). Roast 4 undid the Chaff modification, and also used a different bean. This 4th roast gave my standard "too hot at the end" roast curve (almost 30 degree increase per minute). This is the reason I did the mod in the first place. Set 1 2 3 4 1 350 306 300 309 2 350 309 306 307 307 3 425 330 322 320 330 4 425 338 325 325 335 5 425 336 325 330 336 6 460 378 331 335 340 7 460 392 333 335 347 8 460 395 342 336 380 9 460 408 347 342 403 10 460 410 347 345 410 (#4 stopped at 9:22) 11 460 358 12 460 360 If some of the roasting pros are still reading, take a look at #4. It seems too flat for too long, then rockets up to second crack when the lid fills with chaff. Any thoughts on optimization? The consensus target profile seems to be fast to 300, 20d per minute to first crack, then 10 degrees per minute to second crack targeting 3-4 minutes. Yes? The catch-22 now seems to be either overheating due to inconsistent chaff, or underheating due to inconsistent sealing. Hmmm, I think I'm going to look at other ways to lock down that chaff collector without messing with that seal. Any ideas? I'll also have to try a decaf as a test, as I hear they have much less chaff, and should give a clearer picture of if the low temp was due to a leaky seal or just the overly effective functioning of the chaff collector (likely the leaky seal, but still need to test). Craig (wishing it was just a broken I-Roast) Wichner |
Craig, The plot thickens. As I mentioned in my update (which I posted yesterday afternoon and reposted last night but neither got through until a few minutes ago), I crafted an aluminum foil gasket that I put under the chaff collector ring. This raised the collector up enough that it was a tight fit and didn't rattle around any. I doubt it is air-tight, though. And I'm sure it won't last long as the ring will eventually cut through it. The foil gasket did significantly reduce chaff from reaching the top screen. There is still some tiny chaff that gets through the ring's holes, though, and that eventually forms a thin layer that covers most of the screen. I am hoping you will go ahead and come up with a good way to keep the chaff collector tight and "sealed" and let us know the results. Based on my one roast after my foil gasket, I don't think tightening the gasket ring has much if anything to do with changing the roaster temperature. And why should it? What difference should it make if the chaff blocks the holes below the ring or the screen above it? Either way it is blocking some of the air coming out. Not a lot either way, in my opinion. If you put your hand over the screen there's still a lot of air coming out, even near the end of the roast with lots of chaff. My suspicion is that the most significant variable in all the variations we are seeing in I-Roasts, especially with regard to the modification, is the seal between the chaff collector lid assembly (is there an official name for it?) and the roasting chamber. If this is true (and it is not much more than a theory at this point), then the reason you are getting higher temperatures than Tom's might be explained by his I-Roast having more a little more leakage than your machine does now. -David --------- |
This is a multi-part message in MIME format. Hi David, I do think that a tight chaff collector will at least slow the rate of clogging of the top screen, thus resulting in a longer, slower, and most importantly, a more controllable temperature ramp toward the end. Just a few minutes change would have a significant effect on the roast temp curve. At least that is the hypothesis I'm working on (my testing only slowed down by the large excess of roasted test coffee I'm working through as I type). And for a little while with the modification I was hitting Tom's roast times (4min for city, 5 for full city, 6 for Vienna), so the optimist in my is hoping the solution is something reproducible like securing the chaff collector rather than creating varying gaps in a rubber seal. :-) The "Key" to the problem: I did find that a spare housekey, placed at the top of the chaff collector, seems to hold the collector on perfectly when the lid is tightened. This isn't an ideal solution as the key could slip, and it won't hold the chaff collector down if I remove the lid (for example if I wanted to quickly clean the top screen if the temp was getting too hot). I'll try a roast in the next couple of days, one "normal" (too hot), and one with the key, and see if there's a marked temp and taste difference. Another potentially confounding variable. Yesterday I removed the four screws at the bottom of the glass roasting chamber and pulled out the metal base & gasket. The gasket had a _significant_ amount of chaff in the seal around the metal base, which it seems could lead to an increasingly leaky and therefore steadily cooling roaster. I don't know the expansion effects of glass vs metal, so I don't know if at roasting temps the metal expands more than the glass, therefore compressing the gasket, or vise versa, which would increase the leakiness, particularly if chaff is present. Any engineers out there? Anyway, so now my gasket is clean. Time for another cup of coffee... Happy hunting, Craig From: homeroast-admin [mailto:homeroast-admin] On Behalf Of dna Sent: Monday, August 02, 2004 11:40 AM To: homeroast Subject: RE: +I-Roast Modification Craig, The plot thickens. As I mentioned in my update (which I posted yesterday afternoon and reposted last night but neither got through until a few minutes ago), I crafted an aluminum foil gasket that I put under the chaff collector ring. This raised the collector up enough that it was a tight fit and didn't rattle around any. I doubt it is air-tight, though. And I'm sure it won't last long as the ring will eventually cut through it. The foil gasket did significantly reduce chaff from reaching the top screen. There is still some tiny chaff that gets through the ring's holes, though, and that eventually forms a thin layer that covers most of the screen. I am hoping you will go ahead and come up with a good way to keep the chaff collector tight and "sealed" and let us know the results. Based on my one roast after my foil gasket, I don't think tightening the gasket ring has much if anything to do with changing the roaster temperature. And why should it? What difference should it make if the chaff blocks the holes below the ring or the screen above it? Either way it is blocking some of the air coming out. Not a lot either way, in my opinion. If you put your hand over the screen there's still a lot of air coming out, even near the end of the roast with lots of chaff. My suspicion is that the most significant variable in all the variations we are seeing in I-Roasts, especially with regard to the modification, is the seal between the chaff collector lid assembly (is there an official name for it?) and the roasting chamber. If this is true (and it is not much more than a theory at this point), then the reason you are getting higher temperatures than Tom's might be explained by his I-Roast having more a little more leakage than your machine does now. -David --------- |
I think you make a good point about the location of the blockage not making a difference, but I think that the chaff is a very significant factor in the temperature of the I-Roast, at least in my case. I have a first-run (preordered) I-Roast which will not roast decaf at all. I haven't done any modifications, and the temperature with decaf beans won't get above about 360 degrees. It appears that to get to higher temperatures, my machine requires the restriction in airflow that the chaff provides. Your point that the fit of the chaff collector assembly to the roasting chamber is also important is well taken, though. I guess the bottom line is that airflow may be the most important variable in I-Roast performance, no matter how it is affected. -Christian On Mon, 2 Aug 2004 dna wrote: <Snip> |
<Snip>
I roast decaf now with the standard chaff collector and I block 1/2
of the exit air flow with tin foil -works great.
I did a bunch of crazy testing with homemade chaff collectors. It so
happens that if you remove the top screen, a 4" piece of ducting fits
in perfectly. So I bout a couple chimney caps at home depot, lined
them with aluminum screen on the inside, to create basically a giant
chaff collector. I was excited: it was cheap ($8), fits perfectly in
the top, looks nice ... but I was having trouble getting the heat
adjustments in the roast settings right. I think in HVAC terms that
the chaff/stock chaff collector is creating a degree of "backflow
pressure" and my oversized homemade chaff collector relieved the
backflow too well. I need to spend more time with this, but it has
great potential. If anyone else wants to tinker with it, it's 4" OD
ducting that fits snugly in the top once you remove the fine screen
of the chaff collector. I left the interior ring in place in my tests
to help keep the heat in the roast chamber.
Tom
--
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
"Great coffee comes from tiny roasters"
Sweet Maria's Home Coffee Roasting - Tom & Maria
http://www.sweetmarias.com Thompson Owen george |
I actually put together a page with some of the photos I have taken of alternative i-Roast chaff collectors. Maybe someone has some better ideas. My next attempt will have a primitive exit air flow adjustment which (I think) work really well - I made a drawing of the basic idea.http://www.sweetmarias.com/hearthI-Roast-chaffcollect.htmlTom -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - "Great coffee comes from tiny roasters" Sweet Maria's Home Coffee Roasting - Tom & Maria http://www.sweetmarias.com Thompson Owen george |
I noticed that my iRoast chaff collector was loose, and that sometimes I got excessive chaff on the screen in the lid. My solution was very simple: make a wad of aluminum foil about 3/8 inch in diameter and an inch long. Place it on top of the inner chaff collector in the indentation, at right angles to the handle of the lid. When the lid is put on, it holds the chaff collector down tight - no rattle. My roasts are now much more consistent, with less chaff buildup on the top screen. Pflash |
Craig, I switched to hotmail so I can sent "plain text" messages which I can't do from my ISP's web email application. Thanks for the update on your I-Roast experiments. I'm an engineer but not a mechanical engineer. Most metals expand much faster (the termal conductivity of steel is 60 times than glass) and about 40% greater than glass. Here are some thermal expansion coefficients (fractional expansion per degree X 10 ^ -6) at 20 degrees F: Glass = 5 Pyrex = 2.2 Quartz (fused) = 0.33 Aluminum = 13 Brass = 11 Copper = 9.4 Iron = 6.7 Steel = 7.2 The above information has much less to do with the Purdue engineering courses I took 40 years ago than my skills at using Google. ;) You make a good point about the seal below the roasting chamber which I've been ignoring. Keeping these seals clean and in good shape is no doubt very important to get consistent results. Since my last post on this topic several days ago, I've roasted 3 more batches with the chaff colector ring clamped down (and the gasket back in its original position). (I also got SM's Variac in an attempt to reduce line voltage as a factor in roast variations.) I don't think my times and results are much different from what I was getting without the chaff collector clamped down, however. What do you think of Tom's no-chaff-collector experiments? One of the things that was better when I was roasting beans in a popcorn popper was that chaff was not a factor in roasting time since all chaff was blown out into a bowl next to the popper (but unfornately, everwhere else as well). I'm not thrilled that I have to factor in how much chaff beans have in my I-Roast profiles. -David MSN Toolbar provides one-click access to Hotmail from any Web page – FREE download!http://toolbar.msn.click-url.com/go/onm00200413ave/direct/01/ |
Hi David, I have spent the last few days exploring re-roasting. Seemed too much of a shame to loose that much Kenyan. I got it passable, but I'm curious if others re-roast to the full profile, or do as I did, which was to just fill in the gap they missed the first time, in my case ~5 minutes at 365-395 (internal I-Roast temp) until I got the color I wanted. It helped, but it's not anywhere near what it could be. It's still better than most restaurant coffees... David wrote >>Most metals expand much faster (the thermal conductivity of steel is 60 times than glass) and about 40% greater than glass. << Perfect, so the gasket should seal even if it is relatively full of chaff. I'll probably still clean it every 3-6 months. David wrote >> I don't think my times and results are much different from what I was getting without the chaff collector clamped down, however. << I haven't tested yet. Looks like it will have to wait until Monday for me, unfortunately, as I'll be at a 3-day conference starting tomorrow. The good news about this problem is that I have researched/learned much more about roasting profiles and roasting chemistry...and what underroasted and re-roasted coffee tastes like. More good news from my education (including reading the 187 page PhD thesis on coffee roasting athttp://e-collection.ethbib.ethz.ch/cgi-bin/show.pl?type=diss&nr620reference by Jim Schulman) is that I know in the future I'll be able to roast the same bean and get a variety of great tastes/flavors from it. The bad news is that the I-Roast is built to roast well _its_ way (which is pretty darn good on average), which will make it more difficult for us, vs. the Popper-People, to do dynamic roasts (changing airflow/temp as we roast). However, this same problem should result in repeatable roasts once we have the parameters dialed in (as long as we don't want more than 3 temperature settings, fiddle forever with airflow exhaust rates, or change beans/ambient temp/humidity/etc). I think Tom's drawing of the ideal chaff collector is great and on target. I think the ultimate solution will be something like that, with as little air volume as possible, with an airflow adjustor at the base or top, and a chaff collector (like the lip at the 4" to 6" transition that he has) that is easy to clean. Ideally it will also allow me easily secure a temp probe in there. I'll run by the local hardware store with my lid in hand and see if inspiration strikes. I'm not thrilled about the chaff variable either, but hopefully we'll find an easy solution. Craig |