Behmor 1600 Coffee Roaster - 1st Look - Page 5

[howard seth]

$100 a week, for three weeks, in roasting Hell for not following the CHARTS? I suppose that might get me to dust off my graph paper.

Howard

[stevendouglas]

Howard,

You raise good points, but since this thread already had several posts (by Mike) with time versus temperature, I thought it was appropriate to use this as a base for trying to figure out the best place to place a temperature device to get a better understanding of what's happening in the roaster termperature wise.

As for a review of the machine. Tom (who sells the machine) at SweetMarias has a pretty good write up on the Behmor http://www.sweetmarias.com/prod.behmor.shtml.

My experience and answering your questions (keep in mind I've only had it for a couple of weeks and roasted about 10 pounds of coffee in it):

1. How much does it cost
2. Does it look like it is robust enough to last for a while - (my several Hearthware models (Gourmet and Precision) conked out pretty quickly.)
3. How much can you roast at one time.
4. How noisy and smoky is it. and
5. Perhaps, most importantly - Does the coffee it roasts taste good!

1. $300
2. Yes, it looks fairly robust. I would say it's about on par with the Hottop (which is 2.5 times the cost of the Behmor). But I have no idea how long it will last...nor will anyone else for another 5 to 10 years.
3. Up to 1 pound
4. It's very quiet (about the same as my Hottop) and produces very little smoke (far less than my Hottop)

5. I've been very happy with the coffee. It's bettter than the coffee from the Hottop - IMHO. According to Tom at Sweetmarias, "We are impressed by the consistency of the Behmor; in our tests the cup quality compares favorably to our commercial Probat drum roaster.". Of course, Tom is selling the Behmor, but to my knowledge, he's never said that about any other roaster that he sells.

Again, I've only had it a couple of weeks, but I have absolutely no regret with this purchase - zero, nada, not! Will I in three weeks, three months, or three years? I'm betting I won't....but even if I did....it's only $300. I've certainly eaten more for less in the past!

Good luck!

[Ken Fox]

$100 a week, for three weeks, in roasting Hell for not following the CHARTS? I suppose that might get me to dust off my graph paper.

Howard

This part of the "home roasting appliance spectrum" is a really dodgy place.

When one feels he has outgrown the cheap little air roasters (which generally cost from maybe $60 to $200), but doesn't want to lay out the serious cash to buy a real, small commercial or sample roaster (price of admission, at least $3000+) one finds oneself in never never land. I have heard, several times, the plans of various people to sell something really substantial in this price range, but it never materializes.

So, one has the Alpenrost (or whatever they call the current iteration, if still on the market), the Hottop, BBQ roasters, various home made thingies, and this new product, the Behmor. Of this group, the only product having any real sort of track record is the Hottop, which is neither cheap nor perfect but seems to have found its niche.

The homemade thingies can't be evaluated since they are all different. The same can be said of the BBQ roasters. I'm absolutely certain that some BBQ roasters are really good, but by the time you look at all the different versions people have cobbled together, it is impossible to generalize.

And then you have the Behmor. What exactly do we make of this product. I don't know, and until I did know more than I know now, I'd certainly not be a buyer.

ken

[mike]

It's been interesting to read the posts here, and those elsewhere.

I remember when I first met Dan, we used to have these long conversations about espresso process back in the Silvia days when a lot of machines were just ridiculously off on parameters. It sounds oversimplified, but if you apply decent water at roughly the correct temperature and pressure to decent coffee, you will get decent espresso.

I think with roasting it is similarly simple. Can you apply heat at roughly the correct temperature at roughly the right times to decent greens, and monitor/control the process to a point that you can get the results you want?

Based on my first 5 loads, I think the answer on the Behmor is "maybe". The roast consistency is good, the agtron spread was good, and I've not yet seen a roast bake or stall. The roast times are about where I want to see them be for body development.

It does concern me that the heaters cycle of during critical portions of the roasting cycle, and for the length of time that they do. There may be enough thermal mass inside the roaster, the drum and the greens themselves that the whole thing variation gets damped. What I'd really like to be able to see is the actual temperature profile of the bean mass as it goes through the roast cycle. As I said above, I've not seen a load bake, stall or burn, so I suspect that the applied heat is roughly correct.

I'm also missing the ability to more closely monitor and control my roasts (including 'dumping' the beans quickly), but as Ken points out, this comes with the terrain of moving off the tiny poppers.

And, while I'm appreciative of all the data that has been collected and posted on this thread, based on the roasting times and results that I've seen (all of which have been relatively normal), the temps don't really correspond to bean temps either in actual temperature and likely also variation of temperature.

I'd like to be able to see that the roaster is staying within some relatively reasonable temperature ranges despite the cycling - that would give me some comfort. If it isn't, it might be necessary to adopt an approach where the heat levels can be more carefully controlled - perhaps a PID-like approach.

Related to this is the relative pointlessness of the pre-programmed roast profiles. I'll not try to comment at an expert level on this (as I'm no expert), but the profiles don't reflect what I've seen professional roasters do, and Tom's comments at SM seem to confirm that the profiles are little more than curiosities. Even if I had the simple adjustment capability of reducing inflow heat after the completion of first crack, that would be nice.

I'm actually shocked that no one (to my knowledge anyway), hasn't yet completely bypassed the control board and directly controlled at least the heating coils.... The other thing to look at would be the airflow, but I think that is constant?

Anyway, I'll probably run the roaster as is for a while longer. Dan is close by, and maybe we can rig up something with the Fluke. It would seem to me that putting a probe near the front bottom of the drum (just barely shielded from the heater, as close as you could get to the bean mass) might be the best place to get some even halfway reasonable approximation, but I'd have to go have another look.

[prof_stack]

Well, happy New Year everyone! May it bring great roasts and wonderful coffee.

The Behmor smaller-mesh drums are coming on-line now and will be in our hands shortly. It looks like it will cost $20. Nice.

Ken Fox, you don't have to buy a Behmor to use it. Come to Seattle this week and I'll give you full hands-on experience so you can find out for yourself what the roaster does and doesn't. Or perhaps someone in your city will extend a similar offer.

Just so you know I'm a fan and not a fanatic about the 1600, two days ago I brought out the TO/SC to roast 1/2# of Yirg peaberries outside. The temperature was 35F. Too cold to trust the Behmor.

The TO (1470W) did a wonderful job getting to FC in 10 minutes and SC in 13:30. Very Earthy aroma and a wonderful capp yesterday and today. I won't be selling this roaster any time soon.

[Ken Fox]

Ken Fox, you don't have to buy a Behmor to use it. Come to Seattle this week and I'll give you full hands-on experience so you can find out for yourself what the roaster does and doesn't. Or perhaps someone in your city will extend a similar offer.

I'm not looking to buy another roaster as I am very satisfied with what I have, a modified commercial 1lb sample roaster that is a copy of an old Probat-Jebez-Burns model.

I am also not seeking to become "Mr. Anti-Behmor." I do however personally know three well respected individuals whose taste I know, and who have personal experience with this roaster or its roast product. I am not going to name them because if they wanted to express their opinions online, they would have done so already.

One has received samples from an enthusiastic owner; he deemed them baked and undrinkable. The other two own the roaster themselves. One had his first ever roaster fire with a Behmor, and considers the product dangerous. Both he and the other owner told me that none of their roast results were better than "baked."

Roast temperature graphs that have been posted for this machine do not inspire confidence and in fact support the impressions of these three individuals that I reference above.

This gets us back to the whole issue of the validity of home roasters' evaluations of their own roast products. I have already expressed my opinion on this topic a number of times and specifically want to make the point that I am NOT judging the roast results of anyone posting on this thread. I have not tasted your roast results and therefore have no opinion on them.

My general opinion of home roasters' evaluations of their own roast products, however, stands. This extends to my own opinion of my own roast results. I regularly send out samples to people whose taste I respect, in order to seek feedback and be sure that I am being objective on this subject regarding my own results.

ken

[prof_stack]

All right, Ken. I think I now know where you're coming from. Thanks for spelling it out.

Happy new year to you! May all your roasts be excellent.

[mike]

Ken, trust me, all of us would prefer to be on a high quality sample roaster.

To me, this goes back to one of my original posts. The idea that it takes more than a few hundred dollars to deliver water at approximately the right temperature and pressure to a cake of coffee to create decent espresso to me is absurd. The fact is, a PID'ed Silvia will produce better espresso than you will find in 99 out of 100 shops. That to me is what this is all about - even if it is short of the ultimate.

If a $2 popper can be manipulated to produce approximately appropriate levels of heat and air flow, I feel fairly certain that the Behmor can produce decent roasts either in close to stock mode or with modification .

I do have an initial impression that the P1 profile can produce decent coffee. I have a lot less confidence in the other profiles, especially P3 through P5, which to my limited knowledge could very well produce coffee that is 'baked' due to the long time to bring the coffee to 1st crack. Also, again, I'm concerned about the cycling behavior of the heaters, and wonder if a tighter band thermostat might be needed, but don't have anything to base that on until we get some idea of how much the temperature varies - the system might have enough damping to smooth that out. And finally, I should would like to have the ability to manually adjust the temperature, especially after first crack.

I feel fairly confident that if we could get a reasonably accurate measurement point in the unit that has some relatively close approximation of the temps that the beans see, we would be farther along. It would also be interesting to figure out the air flow characteristics. From the folks that I know in roasting, taking a look at accurate graphs of those parameters would get us farther, and at least address to some degree whether appropriate levels of heat and air flow are being applied. I've not seen anywhere that Joe of Behmor has posted his roasting profile characteristics (not just the heat, but air flow, etc) or what he used to arrive at those decisions, but it would be interesting to delve in deeper. I've got to admit, I know just a bit about profiles and for me it'll be a learning experience to get deeper into what is covered in roasting theory, and how well the 1600 implements the theory - most of what I can go on is Tom's comments at SM and a bit of personal knowledge.

Until I have the time to put in a thermoprobe and mess around, I can't quite answer any of those questions. Maybe later in the month. What I am ultimately looking for is whether this roaster is going to work well in stock mode, or can be modified relatively cost-effectively to do what I want it to do. If the solution becomes as simple as a cheap Auberins PID and SSR on the heater, that's not a real issue.

[Ken Fox]

If a $2 popper can be manipulated to produce approximately appropriate levels of heat and air flow, I feel fairly certain that the Behmor can produce decent roasts either in close to stock mode or with modification .

I do have an initial impression that the P1 profile can produce decent coffee.

Until I have the time to put in a thermoprobe and mess around, I can't quite answer any of those questions. Maybe later in the month. What I am ultimately looking for is whether this roaster is going to work well in stock mode, or can be modified relatively cost-effectively to do what I want it to do. If the solution becomes as simple as a cheap Auberins PID and SSR on the heater, that's not a real issue.

There is not a piece of coffee related equipment that I use that I have not modified, and it may well be that whatever flaws are present in this roaster as delivered can be overcome with some tinkering.

Keep us informed of what you do, and especially your results.

ken

[howard seth]

.... Thanks Stevendouglas and Ken Fox for your responses to my questions. Perhaps my little FreshRoast 8 will keep chugging along for a real long time - like the Little Engine that could - but I'm only hoping it lasts at least till you guys hash this Behmor stuff out, and I can reap the benefits from all your First Look labors.

(By the way - and off topic - I noticed last night on an episode from The Soprano's Marathon Special (HBO) - Mrs. Soprano was unpacking 'My' Elektra Semi-Automatica espresso machine - supposedly bought from Williams Sonoma.

To bad The Sopranos is no longer being produced - perhaps we would have gotten to see some goon doing some home roasting eventually.

Howard

[Marshall]

To bad The Sopranos is no longer being produced - perhaps we would have gotten to see some goon doing some home roasting eventually.

Howard

I remember them doing some home roasting, but I'm pretty sure there wasn't any coffee involved.

Of course, one of their classic scenes was Paulie Walnuts in the espresso shop grousing about them "stealing our culture" and getting rich off it, while he shoplifted (a moka pot, I think).

[stevendouglas]

I did a P1 roast last night and charted two temperatures this time.

Obviously, neither T1 nor T2 represent the bean temperature. Since T1 would never have reached 1C and T2 would have smoked the beans.

1st crack started with slightly more than 5 minutes remaining (i.e., 15 minutes into the roast) and then I manually hit the COOL with 1:10 remaining (for a total roast time of 18:50).

T1 - The thermocouple was located below the drum and slightly behind it so that the drum and beans were sheltering the thermocouple.

T2 - The thermocouple was located in the back of the machine centerline, about 1" above and in front of the top heating element.

A couple of points of note:

1. T2 @ t = 14:30 - Large temp increase. I'm not sure what causes this or if it's a problem with the data. However, I seem to remember the voltage dropping. Another run will confirm one or the other.

2. T1/T2 @ t = 12:30 - The fan comes on and you see T1 rapidly increase and T2 drop.

3. T1/T2 @ t = 8+ These fluctuations are caused by the heaters cycling on and off. Obviously, this is more noticeable on T2 because it's located directly in front of the heaters.

4. T1/T2 @ t = 10:30 (during cooldown) - I opened the door with 10:30 remaining in the cooldown. You can see the drop in temperature, and then around 9:20 I closed the door and you see the temp increase.

I used K type thermocouples. And I believe the readings are reasonably accurate. I removed some bad data points (about a dozen, the temp was logging ever 2 seconds so there were a lot of points).

To Mike's point earlier, manual control of the heater and the fan would allow much more control of the temperatures in the roaster. However, I don't know how the heater elements would fair under the increased temperatures (i.e., what is the purpose of cycling the heaters in the later part of the roast?).

[prof_stack]

Wild looking data! I applaud your efforts.

The bean temperature is obviously somewhere in-between the graphs. The cycling of the element should not have any significant effect on the final roast product, if any at all.

How did the beans smell after cooling? More importantly, how did they taste after proper degassing?

But please keep us apprised of your findings.

----------------------------------------------------------

But I wonder if modding the Behmor really defeats the purpose of the Behmor: Making great 1#, 1/2#, or 1/4# roasts with somewhat automatic controls for a really reasonable price.

Oh, I understand the motivation to have more control and tweak it to our own roasting standards. When I use the TO/SC I track the temperature every minute (not seconds) and adjust as needed. I bought the Behmor for the ease of roasting with little fuss, although I discovered that there is a learning curve for it as there is for any piece of coffee-related equipment.

There is no "set it and forget it" or "close the door and ignore".

[popeye]

Interesting results. They are a little scary looking, as i don't want my beans going from 300 to 500 and back repeatedly. But i have to say the cup indicates the roast - however it occurs - is very good.

One question, though. I don't know much how the heater in the behmor works, but it seems some/most of the heat is infrared. Can a thermocouple adequately measure infrared heat? Wouldn't a white and a black thermocouple register different temps because of different absorption of the infrared radiation? I should have a better understanding of this, but i don't. Off to wikipedia!

[stevendouglas]

i don't want my beans going from 300 to 500 and back repeatedly.

I don't think the temp of the bean is changing so rapidly. Note that T2 (in front of the heater element) varies between 550 and 475. Prof is right the bean temp is somewhere between T1 & T2.

I think you are on to something with the infrared/radiant heating. I do not know much about that either.

Like both you and the Prof, I like the result and one of the advantages of the Behmor is the simple operation.

My conclusion in this is:

1) The result IMHO is very good regardless and as I've said, I like the coffee roasted in the Behmor better than the coffee roasted in the Hottop at this point.

2) Neither of these two temperatures can be used as a surrogate for actual bean temp. I'll continue to look for a surrogate location, but I am not confident that there is one for bean temp. The alternative is to find a way to measure bean temp in the Behmor. I don't think that is impossible but it isn't easy either.

[prof_stack]

The Behmor did a wonderful job with some pre-blended green beans from a top-rated roaster in Seattle. Best roast yet from the 1600. I let it go into second crack for a while and found that a lot of smoke doesn't mean that the beans are overly roasted. In fact, they taste awesome! Nothing half-baked here!

Or maybe its because my living in Seattle next to Charbucks has spoiled my taste buds!

[Ken Fox]

<image>

I did a P1 roast last night and charted two temperatures this time.

Obviously, neither T1 nor T2 represent the bean temperature. Since T1 would never have reached 1C and T2 would have smoked the beans.

1st crack started with slightly more than 5 minutes remaining (i.e., 15 minutes into the roast) and then I manually hit the COOL with 1:10 remaining (for a total roast time of 18:50).

Thanks for going to the effort to produce this graph and to post it.

As you state, neither of these roast temperatures are useful, and I hope that neither of them are truly representative. To state that the actual roast temperature is somewhere in between also gives no information you can use or that could help anyone to evaluate or criticize the roast profiles.

From your verbal description, I would like to point out that there is no roaster I have ever heard of that produces a good roast product, with the onset of first crack FIFTEEN MINUTES into a roast. In fact, it is my opinion that if the heater element is actually on for this length of time, that your final roast product would (probably without question) be improved by reducing your batch size, which should reduce the time to the onset of first crack to something halfway reasonable, say, under 10 minutes. It would also shorten/improve your cool down time, as below. One other thing it might do, given a faster roast and cool down cycle time, is allow you to run more batches in a given period of time, somewhat compensating for the reduced output that lower batch sizes would give you. IF the roaster will not allow you to do this in the stock configuration, I'd seek out a way that I could accomplish this by modification.

You will probably also find that an interval of 3.5 or 4 minutes (possibly slightly longer) between the onset of first crack and the end your roast (assuming it is somewhere near the onset of 2nd) is better than a shorter interval. In my own experience, with my own roaster, an interval of 2.5 minutes produces very flat tasting coffee compared to the longer 3.5 or 4 minute interval. In order to have an interval of ~4 minutes, there needs to be a reduction in the heat input (or an increase in the venting/heat loss) after the onset of 1st crack, as a consistent heat input through the whole roast will result in an excessively fast temperature ramp up during this critical phase beginning with the onset of 1st. This is just a general observation on how roasting beans behave, that I believe is probably valid for most any kind of roaster.

Fairly rapid cooling is important. The cool down period programmed into this roaster (which may be as quickly as its fan can accomplish) seems to me to be very excessively long, but should be improved by reducing batch size, as above.

ken

[mike]

Two things I take away from this:

First, it looks like the heater cycling is somewhat damped in terms of temperature. I really don't have any idea what the temperature is that the beans are seeing, but it seems reasonable to think that they are seeing a temperature between the two measurement values and that they are probably seeing further damping by the mass and screening of the drum. So, I am somewhat pleased to see that there are not wild temperature swings as the heaters turn on and off. Aside from that, as Ken says, there isn't much to say.

Second, I wholeheartedly agree with Ken that if you are seeing 15 minutes to first crack, you most assuredly need to reduce batch size. As I said before, I don't have a fundamental concern with the roasts that I have been doing so far because I am hitting first crack in the 9-ish minute range.

Not related to the data, but to points that Ken and Tom at SM have raised, the coffee is definitely entering into first crack with too much energy. I've done only one roast to date where first and second crack have not run together, and it is a major shortcoming that we cannot stretch the time between first and second crack. As Tom mentions on his site, P2 is perhaps the best profile in the machine, however the lack of control (locked in on time) to move the temperature adjustment points during roasting is a huge drawback. However, P2 may very well be where I start investing more of my effort from here.

I think ultimately the change that is going to have to happen is to open up the unit and bypass the control system for the heater, and install something like a variac, or possibly a PID with a suitably placed TC. I'm afraid I know very little right now about the behavior of the fan, afterburner, etc, and I'm unsure how many test roasts I have time to do to reverse engineer the whole thing.

Again, I think Joe has done a great job here. I haven't heard anybody yell about the myriad of things he could have gotten wrong like rotation speed, design of the drum, design of the unit. He did a whole lot of things right. I will continue to admit that I am short on detailed, scientific knowledge on roasting theory, and do not know that Joe has the time or the inclination to reveal all of the work that he put into that aspect of the unit for a number of reasons.

However, I think if you simplify this down, all of the complaints that I have seen have been the amount and timing of heat application. Boy, that sure sounds like a simple thing to fix to me, albeit at violating the warranty and incurring some additional investment. Right now I have another project on my hands, and no easy access to a variac, so I don't know if I will be able to try this in the near future, but I suspect somebody will soon. I'd be interested to hear whether Joe has interest in extending control of the heat to the user - it would seem a relatively simple modification to the machine, albeit again with some incremental expense to change the state machine.

[cannonfodder]

Those heater on/off cycles look like very large swings. I have never profiled my roaster so it my have the same swings, but I don't think it does. Lets PID that sucker and see what happens.

Those roasts times look long, but I am just an amateur. My roaster ran 12-15 min first cracks and borderline baked the beans. I reduced my batch size to half the rated capacity of the roaster which reduced the roast time. To further accelerate the roast I closed off some of the air intakes for the roast chamber. I regulate the roast profile by cracking open the chaff tray and charge chute to allow more air to flow through the roast chamber during the drying stage. I get first crack in 8-10 min depending on the bean and draw out the first to second transition to 2 min, or at least try to. My roasts got sweeter, deeper and more flavorful, and it is a drum roaster.

My next step will be to hack the machine to allow complete control of the heater and fan, or build a two pound sample roaster. The more I learn, the more I learn how short most home roasters fall. Don't shoot me, that is just a broadly generalized observation and I have no first hand experience with the Behmor.

[DavidMLewis]

One question, though. I don't know much how the heater in the behmor works, but it seems some/most of the heat is infrared. Can a thermocouple adequately measure infrared heat?

I have come to this conclusion with the Hottop, and it's even more the case with the Behmor because of its quartz-tube elements. A probe measuring environment air temperature in this sort of machine doesn't tell you nearly as much as it does in a roaster whose primary heat transfer path into the beans is convective. A probe inserted into the bean mass will tell you the result of the heat transfer, but we don't have that here. My temptation would be to take the profiles as given, and adjust the bean mass so that the cracks match what the roaster profile expects to be occurring.

Best,
David