Having experienced a good shot with my Macap M4 one day, then seeing the same coffee, grind and dose choke the machine the next day, I can appreciate what Jim is saying. Varying the grind on my M4 to fix the flow rate often produces a bad tasting extraction. Then I'm pulling shot after shot, varying the grind and trying to find another sweet spot -- sometimes with little success. I'm going to have to try varying the dose instead.

My assumption up to now has been that the M4 isn't stable and consistent -- i.e., with the same coffee and grind setting, the particle size and distribution may not be the same from one shot to the next (I grind by the shot.) But of course they're reasonably stable during the same session. The variation is always from one day to the next. So it would seem that humidity and perhaps aging of the coffee are causing the variation in flow rate, and the real problem is that tweaking the grind setting to correct it throws off the extraction. It's an exercise in frustration to correct the grind because there are too many moving parts.

Jim, have you figured out what it is about the burr size and/or geometry that accounts for the more constant average large coarse particle size on the high-end grinders?

Ken Fox wrote:A disproportionate number of the lower end grinders are going to be found in the homes of the least experienced readers, whose taste may be less experienced.

I could be wrong, but I don't think we're necessarily discussing subtle taste differences here. Jim has been talking about correcting obvious flaws in the flow rate by varying dose instead of grind. Even raw newbies can tell the difference between between 1.75 oz in 25 seconds versus 2 oz in 15 seconds versus choking the machine. While it may be true that most of us who are new to the game (and perhaps many experts as well) can't determine whether it's strawberries or raspberries in that fruity SO, we can certainly tell the difference between a nicely balanced shot and one you spit into the sink.

I haven't, but John has gotten close with his distribution and up close look at the particles (referenced TGP pages). The surprise was that the conicals seem to have wider distributions of coarse particles, rather than the tight ones you expect form a high quality grinder.

The engineering department at John's university was unkind enough to need the equipment for silly things like teaching, so his access was limited. In particular, he didn't get the chance to test different grinder settings. That prevented us from nailing the issue down completely

But suppose, when adjusting the big conicals, the quantity of fines changes, but that wide coarse distribution doesn't move a lot, then the grinder becomes bulletproof. You adjust the grind, and flow rate of the shot changes while the extraction remains the same. On the Jolly, you have that very narrow coarse distribution, and if it moves, all the extraction characteristics change. Some of the testers thought it was the best grinder when it hit its sweet spot, but they all thought it a lot more finicky than the big boys.

As I said in my response to Ken, the biggest surprise of the TGP is that grinder quality does not equal grind quality, it equals grind consistency.

So my slightly subversive question in this thread is "do these quality issues indicate that we are using our grinders incorrectly? It would be very funny if it turned out that something basic like a Rocky could produce a killer shot time after time if you didn't mess wit the grind setting once you hit a good shot.

another_jim wrote:As I said in my response to Ken, the biggest surprise of the TGP is that grinder quality does not equal grind quality, it equals grind consistency.

From a practical standpoint, the above is irrelevant, because few people here are going to repeatedly toss shots until they get a perfect one; they will toss the grossly awful ones but will suffer along with the mediocre to decent ones, at least as regards flow rate and flow characteristics.

Over my 10 years or so experience with Cimbali Juniors and Cadets (among the better, small planar grinders), I needed to change the grinder settings multiple times each week in order to get good shots. Being a home user making 3-5 shots per day (on average), I could have easily tossed 25 to 35% of my shots on the basis of parameter shift on a day to day basis. Back in that era, I was updosing and dosing by sight, making the potential loss of coffee even worse. If you are lucky, the "bad" shot goes in a milk drink, but if you are not and if you are either rushed for time or you just don't want to throw away so much of your precious coffee, you drink most of what you get, regardless, and try to remember to change the grinder the next time around.

If you do decide to act on Jim's suggestion, you still have to make the bad shot in order to know that the "dose" needs changing, rather than the grinder, so that still equals a tossed shot that many will just drink due to the above reasons regarding coffee waste and time. If you have a "better" grinder then you aren't going to notice these differences very often in the first place, so you won't change your dose anyway, if the observation you are using to change your dosing is the flow rate and shot appearance.

Now, with a "stable" of Cimbali Maxs and a Compak K10 WBC, I toss about 1 shot every two weeks, using a gram scale and dosing at 14g. That isn't to say that I drink every shot, but those that I don't drink are either ones where I'm testing out a coffee, or those I don't like the taste of, that on "eye cupping" grounds were good shots.

Taking Jim's argument to its logical conclusion, with a better grinder, does this mean that if we got a good shot with our high end grinder yesterday, but the shot tastes inferior today, that we should change the grind to accommodate a different dose, since the flow characteristics are apt to be unchanged day to day given the better grinder?

I think you have an interesting idea, Jim, which might bear some sort of standardized testing, but in the absence of blinded testing, I regard it as a hypothesis without any real proof. Since most reasonably good tasters could detect the difference between a shot made with 15g and one made with 16.5 (updosed to account for grind changes), I don't even know how you could possibly test this, except with uneducated palates.

Count me as a skeptic.

ken

Kudos
Great thread. Not a new methodology - folks have suggested altering dose rather than grind, especially for stepped grinders with electronic dosing - for some time now. But this is an excellent approach for helping to relate changes in dose and grind to taste in the cup.

another_jim wrote:But suppose, when adjusting the big conicals, the quantity of fines changes, but that wide coarse distribution doesn't move a lot, then the grinder becomes bulletproof. You adjust the grind, and flow rate of the shot changes while the extraction remains the same.

Speculation
I'd be surprised if the coarse peak did not move with grind adjustments. After all, you are physically changing the separation between the grinder burrs. The most direct impact would be on the coarse peak, with only an indirect effect on the fines.

another_jim wrote:Once the extraction rate is right, the grind setting should be sacrosanct, and you should control the flow rate by changing the dose.

Just can't resist
Good grief! This sounds suspiciously like the moron mantra. Shouldn't we be changing dose, grind, brew temperature, and at least half a dozen other parameters simultaneously?

RapidCoffee wrote:Just can't resist
Good grief! This sounds suspiciously like the moron mantra. Shouldn't we be changing dose, grind, brew temperature, and at least half a dozen other parameters simultaneously?

Consistency is the hobgoblin of small minds

RapidCoffee wrote:Shouldn't we be changing dose, grind, brew temperature, and at least half a dozen other parameters simultaneously?

Bad John... let's not do that one again for a while.

Ken Fox wrote:If you do decide to act on Jim's suggestion, you still have to make the bad shot in order to know that the "dose" needs changing, rather than the grinder,

Most of the critique here is good, but this particular bit is nuts. Once the shot has been dialed in, the flow rate and shot timing will change over time. The usual response is to change the grind setting and keep the dose constant. I'm suggesting changing dose and keeping the grind setting constant. It's not about "having to make a bad shot," but about the best way of keeping ones day in day out shot making on track.

Again, we do this all the time. I don't even think about it anymore. I look at the flow of the shot, and reach over and adjust the grind setting for the next shot if necessary. Instead, now I'm testing making a mental note of changing the grind quantity a half gram one way or the other for the next shot. Anybody who weighs their dose can do this without adjusting their SOP in the least. People who time their dose can adjust grind time by a few tenths to do the same if it can be done simply. People who dose by basket fill may not want to try this until the efficacy is confirmed, since they will need to change their SOP.

Once I have the operational details down, I'll start logging the shot ratings and see if they drift more or less this way. I'll think about a way of doing blind tests, but the only thing I've come up with so far is pretty Rube Goldbergish.

another_jim wrote:Most of the critique here is good, but this particular bit is nuts. Once the shot has been dialed in, the flow rate and shot timing will change over time. The usual response is to change the grind setting and keep the dose constant. I'm suggesting changing dose and keeping the grind setting constant. It's not about "having to make a bad shot," but about the best way of keeping ones day in day out shot making on track.

Again, we do this all the time. I don't even think about it anymore. I look at the flow of the shot, and reach over and adjust the grind setting for the next shot if necessary. Instead, now I'm testing making a mental note of changing the grind quantity a half gram one way or the other for the next shot. Anybody who weighs their dose can do this without adjusting their SOP in the least. People who time their dose can adjust grind time by a few tenths to do the same if it can be done simply. People who dose by basket fill may not want to try this until the efficacy is confirmed, since they will need to change their SOP.

Once I have the operational details down, I'll start logging the shot ratings and see if they drift more or less this way. I'll think about a way of doing blind tests, but the only thing I've come up with so far is pretty Rube Goldbergish.

Jim,

Of course we all (or most of us) adapt our shot making parameters to what we see coming out of the PF on the preceding shot; that goes without saying. And of course, if we have any experience with this process at all, we sometimes modify which parameter we change to fit circumstances. For example, occasionally I forget to adjust the grinder after a substandard shot and I am left with a pile of grinds that I know won't work well with my usual approach, so I updose or downdose a little to try to make it all work. OR, as would be more common for me, I'm coming to the end of a grinder hopper's load of coffee and I don't want to toss the grinds that would have made the last (substandard) shot due to popcorning and resultant grind variation. So I updose that shot in order to use it in a cappa.

I find a lot of this discussion to be akin to the stereotypical "how many angels can dance on the head of a pin" discussion, or hearkening back to when I was young, a conversation I used to have with a friend about whether we were lazy which explained our unhappiness with our "jobs," or if we just hadn't found the right job, yet. Of course we never reached any sort of conclusion on that one, either

For example, one can discuss Dr. Illy's findings on fines and whatever, but what was his methodology? Did he "dose per shot," as is advocated by quite a few people here, in spite of the fact that no commercial grinder in creation is designed to be used in that way, with beans popcorning up towards the ceiling? What is the effect of popcorning on the distribution of grind particles? I haven't a clue and I can promise you that there isn't a soul posting on this thread who has any idea on this, either.

I have a pretty sensitive BS meter, and it is redlining right now

ken

This is testable and refutable, so by definition, not BS. The logic of your objections, on the other hand, eludes me. It sounds mostly like sour grapes, since the full hopper requirement on the Max disqualifies you from doing most routine coffee and methods testing on your own. You should trade them in for something that works for more than one coffee at a time.

Anybody who wants can try this for a week and see if they are getting more, less, or about the same shot to shot consistency.

The Rube Goldberg method (if anyone has a better trick, please speak up quickly) for blind testing is to prepare two baskets at a time of the same coffee over the course of several days as the settings drift. One basket would hold the original grind setting fixed and vary the dose, the other would hold the dose fixed and vary the grind setting. The shots do not have to flow the same; this is, after all a consistency test, but they should be both be acceptable shots, just as they would normally be. Cover the baskets with lids, so the fill level isn't a give away, shuffle, and make shots. Then say which shot you like more. If the scores consistently favor the fixed grind shots, the method is a keeper. If it's only a small win, a tie, or worse, then it's probably not worth while switching.