Compak K10 needs a little dosing love - Page 3

Grinders are one of the keys to exceptional espresso. Discuss them here.
User avatar
Peppersass
Supporter ❤

#21: Post by Peppersass »

I think this would be hard to test with a refractometer because it measures total-dissolved solids, which doesn't tell you anything about the specific mix of dissolved solids you have in the cup. It's conceivable that a combination of under-extracted and over-extracted flows could give you a % TDS reading that's in the sweet spot of 18%-20% extraction yield, though I'm only guessing about that.

I think the impact on taste is a matter of degree. If you have badly flawed extractions -- spritzes, gushers, donut extractions and so forth -- you'll likely have inferior taste in the cup. I'd speculate that for newbies the spritzes and gushers are the major contributors to bad coffee, whereas intermediate baristas probably have less impactful but still significant flaws like donut extractions. But if you have minor flaws they probably don't vary the taste as much as the normal and unavoidable shot-to-shot variation you get with even the best equipment and technique.

I don't think there's any such thing as a perfect pour, and I wouldn't know how to define that anyway. Some of the videos purporting to show perfect pours still have the occasional short-lived river of blond flowing through an otherwise perfect tiger-striped cone. Is that perfect or a flaw? If it's not a problem, then how long can that blond stream last before it's considered a flaw? I've had great shots from VST baskets, which tend to have multiple streams for a longer time before consolidation into a single stream than other baskets. Is that a flaw or a non-issue? Is an off-center cone bad? Not as far as taste is concerned, in my experience.

While I think it would be a good idea to test these things with a refractometer to see whether my assertion is correct, I have a feeling that it will only register a meaningful difference when the shot is obviously and badly flawed.

Aida Battle: Indigo Reserve from world renowned Finca Kilimanjaro in El Salvador
Sponsored by Aida Battle
User avatar
Viernes

#22: Post by Viernes »

Of course the refractometer doesn't tell you how the cup will taste and, as you said, it's possible to get a combination of over/under extracted flow which gives a good TDS reading; however to test this I use my refractonguemeter. :D

In my experience the refractometer indicates if something goes bad with the grinding. I'm not the only one.

http://shotzombies.com/2010/08/16/using ... equipment/

User avatar
Spitz.me (original poster)

#23: Post by Spitz.me (original poster) »

All "WE" ever want to do is make sure that we're getting the most out of our grinds. If the shot visually looks like we may not have gotten the best out of those grinds than we'll tweak the prep, even if it tasted good. I think this hobby always has us chasing after the perfect shot. There's always something off, IMHO. Blind taste tests and side-by-sides are interesting to hear about, but, can the shots actually be compared considering not all variables are constant always?

I didn't think I would need to redistribute the grinds in the basket, but really, I do with the k10. It doesn't extract evenly if you don't play with the grinds prior to tamping, locking and pulling. This is the same as the Vario. I've seen video of other grinders where the extraction LOOKS very even without any redistribution of the grinds. This is what I thought would occur and it hasn't.

Taste in the cup is paramount, if it tastes good than who cares, right? But, could it have been better? Probably, and that's the point.

Let's see if LOOKS actually tell us a story.
LMWDP #670

User avatar
LaDan

#24: Post by LaDan »

tekomino wrote:I think this would be really interesting to test and put to rest with the refractometer (paging Mr. Mitch :D ).
...
I do think that pour quality definitely affects taste.
I agree. Judging from my experience, my shots went from good to very good, when I figured out a way to get one stream.
I think getting refractometer readings on pulls that do not look pretty (and pretty to me is an even extraction) should show whether there is extraction difference and thus taste difference...
I am afraid there will still be a difference in the taste even if the refractometer will show that the extraction is OK in this case. If you have parts of the puck over extracted and parts that are under extracted then at least theoretically you'd get an average that is within the OK range, but the taste will be different between evenly extracted and averaged extracted for the same extraction value.

For me, when I got the nice even one stream center cone extraction within the first 12-15 seconds, the taste went up to the next level.

PS. This is with a VST basket.

User avatar
LaDan

#25: Post by LaDan »

Peppersass wrote:I think this would be hard to test with a refractometer because it measures total-dissolved solids, which doesn't tell you anything about the specific mix of dissolved solids you have in the cup. It's conceivable that a combination of under-extracted and over-extracted flows could give you a % TDS reading that's in the sweet spot of 18%-20% extraction yield, though I'm only guessing about that.
Oops, didn't see your post. You've said the same thing already. Great minds and all that jazz... :D

User avatar
haunce

#26: Post by haunce »

I WDT and its consistent every single time.

User avatar
LaDan

#27: Post by LaDan »

Spitz.me wrote: Taste in the cup is paramount, if it tastes good than who cares, right? But, could it have been better? Probably, and that's the point.

Let's see if LOOKS actually tell us a story.
For me, the K10 still needs good distribution and tamping in order to get a nice looking pull. Here is what I do, step by step:

In the order of importance, loading the basket is the most important part. If I don't load it perfectly, no WDT or Stockfleths will help save that loading.

For the sake of explaining, imagine for a moment that the basket is a clock with the markings of 6, 9, 12, and 3 o'clock.

Doesn't matter if you grind the whole dose into the dosser or load while grinding. Let's say your are done grinding and the dosser has the grounds.

Start by tilting the pf 30-45 degrees by pushing the pf handle down on the K10 fork. You are starting to load 6 o'clock. 6 o'clock is down and 12 is up 30 degrees. Slowly pull the dosser lever and gently fill the 6. While still pulling the lever slowly, start turning and tilting the pf toward 9. Try to create an even level rather than hips. That's why you are pulling while tilting and turning. Continue to 12 and then to 3 and back to 6. Now you have one level which should be around half the height of the basket. Continue in the same manner until the basket is full.

You are filling from the center of the basket to the tip. You are filling the corners so that the grounds fall partially on the walls and partially on the bottom. That's why you are tilting 30 degrees. Continue to dose slowly and even up whatever deeps you have in the basket until you are done. Weigh the loaded pf to see if you have the desired dose weight.

What you should NOT do, is load one side substantially more than the other and then fill in the low part. You should try to keep everything even while you dose and turn/tilt the pf.

OK, done with the loading. Now you have a pf with a pretty good distribution right from the grinder. Go the next step to make it even flatter by doing "half" WDT. Take your pin or whatever it is you are using for WDT and work THE TOP ONLY with the tip of that pin. What I mean is that you only stick the pin about 3mm into the grounds. You do not want to touch the bottom of the grounds, they are fine. You just want to smooth the top with the WDT. Do not go deeper than about 1/4 deep into the basket. It takes all of a 3 seconds. ;)

Now the top is very fluffy and even. You can take your tamper and tamp if you want. I on the other hand go and do a quick Stockfleths to 'polish' the top and/or to remove a 0.2-0.4g that I might have over dosed to bring it to exactly what I wanted. (Today it is 18.2g :roll: Call me OCD).

Now tap the pf strait down lightly once of twice to settle down the grounds. Put your tamper in and tamp strait down. No nutations, nothing else. You want to twist to polish, go ahead, but don't press again after polishing.

That's it. Pull the shot. I bet you $20 you'll get one stream one cone in 15 seconds or less. If you are not, you are not following this correctly.

Call this the LaDan Distribution Technique - LDDT. LOL. It sounds better in Spanish, "El DDT", the famous wrestler move. Haha.

Like this. This is right after loading:

La Marzocco · Home: customized for espresso aficionados
Sponsored by La Marzocco · Home
User avatar
Peppersass
Supporter ❤

#28: Post by Peppersass »

LaDan wrote:In the order of importance, loading the basket is the most important part. If I don't load it perfectly, no WDT or Stockfleths will help save that loading.
I agree completely. It took me a while to realize this. Now I take great care with this step.

My technique is different, but the end result is pretty much the same. Rather than dose directly into the pf from the doser, I dose into an OrphanEspresso plastic dosing tray that sits on a small scale that sits on the fork. Obviously, the main reason I do this is to weigh the dose, which I feel is critical for consistency.

Of course, I could remove the basket from the pf and use that instead of the dosing tray, but I've found that no matter how careful I am, the distribution is not always even when I dose from the doser directly into the basket. It can be messy, too. So what I do is dose into the dosing tray and then transfer the grounds to the basket, which is mounted in the pf.

The dosing tray can be slightly squeezed, and using my finger I can gently tap the grounds into the basket. This method gives me precision control over where the grounds land. In particular, I try to coat the perimeter of the bottom of the basket as evenly as possible. From there, I try to get an even amount of grounds into all sectors of the basket. Rather than rotate the pf and/or basket, I just move the lip of the tray around the basket.

When the coffee reaches the upper edge of the basket, there's usually a little extra in the dosing tray, which I pile into the middle. Then I Sockfleths to even out the dose and fill in the edges near the top. I take a lot of care with this step, too, making sure to fill in any visible gaps.

Like you, I tap the pf two or three times lightly on the counter to settle grounds. Then I do a light N-S-E-W tamp and one quick nutating tamp to make sure the coffee at the edges is well compressed. I've found this step to be necessary for avoiding donut extractions. Then I tamp lightly straight down. The critical part of this step is to make absolutely sure the tamper is level in the basket so that the puck ends up level, too. Then I do a partial twist to polish the top, and I'm done. The polish isn't necessary, I just like to do it.

I was inspired to go this route by Ken Fox, who extolled the virtues of dosing into a ramekin and scooping the coffee evenly into the basket with a small spoon.

User avatar
Spitz.me (original poster)

#29: Post by Spitz.me (original poster) »

Just wanted to follow-up.

As an FYI - my machine's pump was acting up badly and was 90% of my problem. I still need to prep my grinds with the k10, but not as much as I thought I needed to when my pump was acting up. Now that my machine pulls at 9bar consistently I'm getting full bang for my buck with the k10. I'm really tasting it now.

I think I'm still fishing for grinds for too long, but I get all the grinds I need.

I have found that I'm wasting MUCH more coffee than before. AT the end of a session I'm throwing out close to 6g or so of grinds after cleaning.

Whatever... it's awesome.
LMWDP #670

User avatar
Peppersass
Supporter ❤

#30: Post by Peppersass »

That's a lot more than I throw out. I'd say my waste factor averages about 1g-2g.

What I've found is that the K10 retains about 1g of grounds on the first shot, but disgorges them on subsequent shots.

- First shot, I weigh out 1g more beans than the target dose. This is probably overkill and .5g-.7g would probably do it. But I want to make absolutely sure I get enough grounds to make up the target dose.

- After every shot, I pulse and sweep at least three times. If I have the time, I pulse and sweep until the acid brush pulls out virtually nothing. Note that if one of the chamber sweep arms lands in the chute opening, you have to pulse again to get it out of the way. You have to be able to reach all the way in to the vicinity of the motor shaft.

- Second shot, I weigh only .5g more beans than the target dose. I could probably go even lower than that because there's almost always excess grounds in the doser after I weigh out the dose. But again, I like to be absolutely sure I don't come up short on grounds.

- By the third shot, I can weigh out the exact target dose and will always have that weight or a little more in the doser.

- When cleaning up, I use a 1" paint brush to sweep the chaff off the top of the ledge in the burr chamber. I also use a thin angled grinder brush, which I stick down into the grooves of the rotating burr (with the grinder off, of course) to get what grinds I can off (i.e., pushed down.) Then I pulse and sweep until absolutely nothing more can be swept out of the chute. The I sweep the doser, using the 1" paint brush to sweep grounds off the upper parts of the vanes, etc., and to sweep the bottom of the chamber. I've removed the plate that sits over the doser exit hole, so I'm able to push the paint brush into the hole to clean it, too.

- When I'm all done with the cleaning, I have 1g-2g, sometimes less, in the dosing tray I use to catch the waste grounds.

- I use a high-powered hand-held shop vac to suck out stale grounds about once a week, when I do the detergent backflush of my espresso machine, or when I change coffees. I've built a little adapter hose that I can stick into the chute and seal it while I run the grinder. That gets anything loose out of the grinder.

My theory is that grind retention is mostly a matter of ambient humidity. People who live in dry, western states (like jammin') have reported different amounts of retention than I see. I tend to see somewhat different retention in the winter and summer, when the heat or AC is running all the time, than I see in the fall when we don't heat or cool.

I also think that some grounds are retained when grinding the first shot, but virtually all of them come out later in the session. What I see for waste is usually very close to the excess weight of the beans over the ground doses.

I've read a lot of posts from Versalab M3 owners about the "egg" of grounds that almost always drops out shortly after grinding. I've wondered if this is the same phenomenon, and perhaps a trait of conical grinders.