So, I haven't been very happy with my coffee for the last few weeks. IMO one of the greater limitations with my roaster is airflow, at the best of times I think its sub-par. Nonetheless I usually manage to get pretty good results, or at least I think I do!

About 6 months ago I cleaned out the main airflow exhaust tube, up to the roasters side of the impeller. I was having airflow issues and it was pretty gunked up. That fixed things substantially, but issues arised again recently so I took the motor and impeller out. This is actually not a difficult task, but was made so by the lack of room and maneuverability around the roaster, and the close proximity of my espresso machine if I were to drop the heavy motor while removing.

Anyhow, I got it off and I think I've found the the source of the problems...


Impeller, sucks air/chaff from roast chamber and cooling tray


Impeller housing. Has exhaust hole (not visible) to chaff collector cyclone.


Scraped off impeller.


Scraped off impeller housing. What you can't see from this photo is the extra depth, the build-up was maybe an inch thick around the outside of the impeller and the same depth in front of it. You can see through it now, down the tubing that sucks the air from the roast chamber. In this photo, that tube hasn't been cleaned yet, so the small obstructions there were removed. You can see the airflow damper in there, its in the full open position (ie minimum obstruction).

My last roast from it, before I decided to pull it apart was fairly smokey, hence the final cue to rip it open. It was a different blend to my normal, but its actually come out alright this morning anyway.



So, all of this was downstream of the roast chamber, which is very clean. The net result is a smokey roast, which I believe is something Probat offers as an 'aroma roast'. For me it was a shock, and of course a potential fire hazard, as well as just being gross. The normal blend I roast certainly suffered because of it.

Hopefully this will help encourage others to check their exhaust systems, particularly if they are difficult to access and you've been putting it off.

Cheers, Chris

That first picture is a little disturbing.

The first and the second...

So I am curious, have you noticed a significant increase in airflow in the roaster now? As I understand it, this does seem disgusting, a probable fire hazard, and potentially air-flow-restricting, but the air that passes through this is all exhaust, correct? It's not recirculated back to the roasting chamber? I'm just wondering how this hurt your roast. Lower airflow means higher environmental temps in the roast chamber, but it seems like to some extent you ought to be able to adjust for that if you have a probe monitoring the temperature.

Sorry if I've missed something basic there. I hope your coffee gets better again in any case!

Presumably, the soot buildup lowered the airflow.

When roasting, the rate at which the beans heat up depends both on the temperature inside the roaster and the airflow. Like in a convection oven, faster airflow at the same temperature means faster roasting. Therefore, lower airflow means either a slower or a hotter roast.

Probat roasters have variable dampers to modulate the airflow, other brands also have dampers or variable speed fans. By throttling the airflow and raising the temperature, you generate more smoke and exhaust it more slowly. In reference to this, Probat calls the hot, smoky roast you get with the airflow shut down "an aroma roast."

As far as most modern roaster masters are concerned, the term "aroma" is a euphemism for "stink."

another_jim wrote:the term "aroma" is a euphemism for "stink."

That seems pretty typical of 'good to the last drop' type coffees as well. Thanks for the enlightening explanation.

shadowfax wrote:the air that passes through this is all exhaust, correct? It's not recirculated back to the roasting chamber?

Yeah, gross indeed, but totally exhaust air.

shadowfax wrote:I'm just wondering how this hurt your roast. Lower airflow means higher environmental temps in the roast chamber, but it seems like to some extent you ought to be able to adjust for that if you have a probe monitoring the temperature.

Correct, but since I normally roast for a few friends I do a kilogram at a time. You get a fair bit of thermal inertia and it makes the roast difficult to control - especially once it goes exothermic around first crack. It changes the behaviour from its norm, and from what I've become accustomed to.

another_jim wrote:When roasting, the rate at which the beans heat up depends both on the temperature inside the roaster and the airflow. Like in a convection oven, faster airflow at the same temperature means faster roasting. Therefore, lower airflow means either a slower or a hotter roast.

Normally true, but in my case the heating element is inside the roast chamber, not below it. Less airflow means hotter, and smokier.

dialydose wrote:That first picture is a little disturbing.

In reality, the second is worse!

another_jim wrote:As far as most modern roaster masters are concerned, the term "aroma" is a euphemism for "stink."

Since my roasts have been going downhill recently, then I would think thats fairly accurate!

shadowfax wrote:So I am curious, have you noticed a significant increase in airflow in the roaster now?

I haven't actually done my first roast since scraping it out. I imagine I will notice a fair difference. I think its one of those things where the performance has gradually been changing (deteriorating) over months so you don't really notice it on a roast by roast basis. I did check the airflow by turning the roaster on while it was partially dismantled, and it was still alot less than I would like. But as I mentioned earlier, I think thats one of the limitations of the machine.

Unfortunately, that also means the chaff cyclone doesn't work as well as it should, since there is not really enough airflow speed in there. That in itself probably helped to cause the build up in the first place, which I assume is compacted chaff.

I started noticing excessive environmental temps (up by 10-20C over normal) and unusual bean temp behaviour (faster runaway after first crack). That and more smoke than usual coming out from the sampler hole while having a look finally gave me the nudge to pull it apart. Should have done it long ago.

Cheers, Chris