Okay, so I was working with a friend's EK43 grinder and making pour over coffees, mainly on a Kalita wave.

First of all, I noticed that you are working with a much smaller average grind size to get comparable extraction yields. My normal grinder produces so many boulders that almost every grind setting feels a little chunky. I tend to favor longer brew times (4-5 minutes) with small pulse pours and a relatively quick draw down on the Baratza since it doesn't seem to grind evenly at fine settings.

Second, I have been getting very different draw down times with subtle variations in technique, even at the same grind setting, dose size, and coffee. I was playing a little fast and loose with the amount of water I poured and when I poured it, really just focusing on the time when all the water was added. I know the way to consistency will likely be to pay more attention to this, and develop a system like Matt Perger does for his v60s (http://www.youtube.com/watch?v=MPDfn--vxK8). The slurry was never allowed to completely drain after the bloom and the high and dry grounds were washed off the sides once or twice in the brew. Other than that, it was just pulse pouring every 20 or 30 seconds, using a spiral pour from the center, keeping the slurry filled but not too high. Each coffee was brewed with 207 F water, measured in the full 1 liter pouring kettle. 300 grams of water was poured over 18 grams of coffee.

Using a ECX Gedeo Yirgacheffe roasted by Mountain Air Roasting I got the following results:

• 1. With a grind setting of 7.5 all water was added by 2 minutes and the drawdown took 2 minutes. Extraction yield was 22%, just a touch of astringency but good flavor.
  2. With a grind setting of 7.5 all water was added by 2 minutes with a large final pour and the drawdown took 1 minute. Extraction yield was 19%.

I think this coffee wants to be somewhere between these two yields. It tasted best in a normal immersion cupping, at about 21% yield.

Using the same coffee (from shrub/marias), but roasted lighter, I knew I would need to grind finer. However, I got some strange results. Both cups were brewed a relatively fine grind setting (6 on the EK), or a fine drip/cupping grind. In the first cup, all water was added by 3 minutes and draw down took 2 minutes. This cup had a yield of 24% and tasted great. The next day I brewed the same coffee at the same grind setting. I added the water over 3 minutes, and the drawdown took 4 minutes. Interestingly, it only had a yield of 21% and was not as good. So despite brewing for longer, less coffee was extracted. I'm not sure how to explain this, except for maybe that the brew turned into a quasi-immersion brew, slowing down extraction significantly.

So any information you may have at speeding up draw down, pour agitation, fines migration, filter clogging, and achieving consistency would be helpful.

Yes, the grind size is smaller, since the fines contribute less to the overall extraction, so the modal sized particles need to extract more. I do two types of brews with my EK: a 20-21% cupping, that is usually polished through a V-60 instead of stirring / breaking the crust. These can be of variable draw-down time, depending on how well I hold back the fallen fines. Usually no more than 30sec variability, but I'm not timing. Extraction on this type of brew is pretty easy to keep consistent. I also do 3min V-60s, 18g/300g, with a few pulses during the first minute to run off the gas before laying clear column on top. These tend to be very consistent in time and extraction. At home, I have an Encore and a Chemex. I brew 30g/500g there, and drawdown times are all over the place, as is extraction. It is at home that I am having frustration. With the recent word that some newer made Chemex filters are clogging easier, it makes me wonder if our inconsistencies are of the same nature.

faustroll wrote: So any information you may have at speeding up draw down, pour agitation, fines migration, filter clogging, and achieving consistency would be helpful.

The most delicate dribble from as close to the surface as possible promotes the lowest fines migration. I think a transparent puddle is the hallmark of a good pour. The result is faster draw down and more even extraction (rounder, sweeter flavor).

I'll have to test the filters. Maybe just see how fast 200 grams of plain water flows through 5 random filters?

In the past I would get much faster draw downs with a v60, so I may try that too. I was getting used to the Kalita and feeling good about it because in theory it has better bed geometry.

Are you sure the extraction is more even? A plausible explanation for the opposite is if the fines move to the bottom of the brew bed or into the filter, they will only be touched by brewed coffee, not fresh solute.

I'll have to try your method of degassing the brew bed. I've only done a stir in the beginning with v60s, and I would sometimes stir after the last pour (like Scott Rao suggests in Everything But Espresso).