When this issue first came up, I compared shots side by side single dosed and from a full hopper on my Compak, and on some of the Titans, and on the Vario, when they were here. In these comparisons, I did make the required adjustments. I could taste no consistent short term difference between the shots.

This is the kind of experience that helps me guide my procedures.

cafeIKE argued that there is a systematic taste difference, but I guess no one has comprehensively tested it.

Comprehensive enough for me.

If you don't want to single dose grind, you certainly don't have to. Depending on usage rate, maintaining a certain degree of pressure with a "bean column" can certainly be managed for optimum results. That seems clear enough.

If Jim has tried side by side tastes of "bean column" shots and single dose shots, that carries a lot of weight on that particular issue. If a bunch of guys wearing lab coats conduct a similar experiment to a statistically significant level, those results might carry more weight, but I won't wait around for that to happen.

RapidCoffee wrote:I get a more consistent result when there is an adequate bean mass in the hopper.

Consistency versus proneness to channeling is a different issue again; one that I didn't test. It's simply not an issue on my machine at reasonable doses. I've stopped looking at how pretty the pours are, since it does not correlate at all with taste; but I doubt I'll be nominated for an espresso porn Shotty (tm) award any time soon.

I agree it would be nice to look at the grind distributions, but then again it would be even nicer to have a discriminant for the distributions of good versus bad grinders first.

another_jim wrote:Of course, the grind and flow changes as the hopper empties. Of course, the taste will be worse if you don't make the required adjustments.
...
In fact, your observations actually are an endorsement of single dosing. When I single dose, the grinder is in an identical state every single shot. If you don't refill the hopper to the same weight after each shot, your grinders are not the same state, and the grind adjustment/dose drifts further and further from optimal with each shot.

Taking the above at face value, one should adjust the burrs continuously while grinding a single dose.
With an initial 250g load on the burrs, and a 15g shot, the load varies from 250 to 235 or 6%.
With a single 15g load, the variation is 100%.

The fact that one has to adjust the burrs as the hopper empties is the single most compelling argument AGAINST single dosing A single dosed shot will have one grind at the start and an entirely different grind at the end. Such variation is unlikely to promote the most even distribution and could contribute to a plethora of defects.

With a mini hopper, frequent replenishment and a suitable weight, one can keep the load, and grind, variation within a few percent for the whole shot for all shots. The entire shot has nearly the same distribution from top to bottom. That is not possible when single dosing.

This is a great discussion. My brief personal experience with two different grinders has been, like John's, that I get more consistent results day to day if I keep some beans in the hopper.

I started out with a Macap M4 stepless doser. I was simultaneously trying to learn how to make espresso and trying to evaluate lots of different coffees to see what I liked. I quickly dispensed with the hopper and moved to grinding by the shot. I used a Bumper rubber tamper stand with the tamper in it as a lid for the grinder throat. Without the rubber holder, the tamper would have contacted the nut on top of the motor shaft, scratching the tamper's stainless surface. Basically, with the M4, there's no way to weight a single shot of beans all the way through the grind without some sort of cylindrical device.

I had consistent problems with the grind setting needing constant adjustment with the M4, always from one day to the next and frequently during the same day or even the same session. This was true even after grinding about 10 lbs of coffee over a period of about 7 weeks. It seemed to take an inordinate number of attempts to dial in the grind, and it would change significantly from day to day, and not in a predictable manner. I had some issues with channeling, too, and had to resort to WDT (even with the GS/3, which is pretty forgiving.)

All this led to my trading the M4 for a Baratza Vario (on Jim's recommendation, I might add.) This is probably a temporary move while I wait for the next generation of more user-friendly conical, doserless grinders. The Vario is much more stable than the M4 and I feel the grind quality is significantly better -- fluffy, no WDT needed, etc. I started out using the Vario in single-shot mode, but found that the grind had to be adjusted every day, and sometimes between shots, even after a few weeks of break-in. Much more consistent than the M4, but still annoying.

So, I decided to abandon single-shot grinding and try putting 6-8 oz of beans in the hopper. The results have been dramatic: the grind setting is much more stable from day to day, and almost never needs to be changed during the same day. Sure, the grind has to be adjusted as the beans age, and sometimes as the hopper gets close to being empty, but it's much less frequent than before. Some days I have to adjust after the first shot, other days I don't. I feel like this is slow-motion compared to what was happening before.

Another advantage to filling the hopper is that I can use the timer. Turns out that once the coffee is dialed in, the timer is quite accurate and repeatable from day to day, well within 0.5g (closer to 0-0.2g.) I'm still weighing the grounds because I'm OCD , but may dispense with that eventually.

The bottom line here is that I had to change my preferred preparation method to get the results I wanted. It's not the end of the world: being able to change coffee every 6-8 oz is fine -- it's at most 5-7 days worth of beans. If and when I upgrade to a top-of-the-line grinder, I'll keep the Vario for an alternate coffee or always-available decaf. My sense is that for non-cuppers, multiple grinders may be a better solution than single-shot grinding when you want to vary the coffee. I know I'm not the first to come to this conclusion.

Like John's observations, mine are anecdotal and unscientific, and unlike John's they're based on a lot less experience. The caveat is that the observations have been made while learning from scratch how to make espresso, and also while switching grinders and machines (and having some flow rate problems with the latter.) That's a lot of variables and moving parts.

cafeIKE wrote:The fact that one has to adjust the burrs as the hopper empties is the single most compelling argument AGAINST single dosing

Part 1:

There's one and only one compelling argument for doing anything in food -- taste differences. For instance, I adjust the grind and dose when the shot is over or under extracted, if the flow is too slow or fast. But I've never tasted a shot that said to me that the hopper was too full or too empty. If it really affects the taste, then how? What is the taste defect caused by a too empty hopper?


Part 2:

Taste apart, I can appreciate that if grind adjustment becomes a lot more fiddly with single dosing, one may want to go to a full hopper.

However, I have never seen this. I have a schedule of dose weights and grind settings for my daily use of three to four different coffees. In general, the numbers will stay the same over the three to six days after the coffees degas and before they stale. So either I'm not picky enough; or John, Ian, and Dick are too picky.

I think the difference is here: I mostly don't care if a shot flows 25 or 35 seconds, since these variations, provided the puck is well packed, almost always are a result of changes in the dwell time and early flow rate. In my experience, these variations have no effect on taste. Shot time variations due to channeling, too much or too little flow during the middle of the shot, or gushing at the end, do have a taste effect. But for these variables, the hopper fill is not a factor.

So I'm wondering if people who think there's something wrong with single dosing are responding to changes in the dwell time or early flow. If this is the case, I'd urge them to do some taste testing, since I don't think it makes a difference.

I know this initially sounds implausible. But think of the endless preinfusion experiments, all leading nowhere except as channeling prevention. A rich (rather than skimpy or gushy) flow in the middle of the shot, and the shot staying steady in flow towards the end, rather than gushing and blonding, are much better diagnostics of a good extraction. The pressure profilers seem to moving to this conclusion as well. So it may be better for your peace of mind to simply ignore the first 15 seconds of the shot, and check in for the second half.

Please note, I'm not saying what happens inside the puck in the first fifteen seconds of the shot is unimportant -- on the contrary, I think it's all-important. Instead, I'm saying that what one sees in the first 15 seconds of the shot is not a good indicator of its quality.

another_jim wrote:What is the taste defect caused by a too empty hopper?

If one has to grind finer as the hopper empties, the taste defect of a hopper too empty is identical to a grind too coarse.

another_jim wrote:I have a schedule of dose weights and grind settings for my daily use of three to four different coffees.

I'm very curious as to what the heck you drink. Today I switched from [ 5 pounds of ] Supreme Bean Ring of Fire to Caffe Fresco Black Hand. At EOD yesterday I pulsed the dregs out of the MKX, dusted and swept it clean. As is my habit, before charging with Black Hand today, about 18 hours later, I pulsed the MKX again. A full 5.7g of detritus my reward.

Perhaps the difference we are tasting is due to my preference to drink the same coffee for several pounds and yours for several coffees per shot.

I don't give a rat's patoot about 'standard' numbers.
I adjust grind, dose and temperature by taste.
I end on colour, not time or volume.

With the 'hopper' at 'mid-level' shots :

• are less finicky, as in ±0.5g vs ±0.2g when single dosed or the level is too 'low' or too 'high'
• distribution is a non issue, a gentle shake, rap and light tamp of any size and shape
• the taste more balanced to my liking
• about the only time I see a spritz is if I'm too lazy to grind out stale coffee or when grinding the heel of the batch out of the grinder.

Perhaps more exalted equipment would relieve me of these delusions.

Actually, you drink 5.7 grams of stale coffee per shot. I drink fresh coffee on each shot. With no hopper, I can pulse the grinder empty each time. (it takes 4 pulses on the Compak, and ends up with about 5 grams, used to take two on the Mini for about 3 grams, not an issue on the Vario)

You just keep making more and more arguments for single shot dosing

Actually, not. I grind into the bin before every shot. The 'stale' coffee is the same. No so on single dose, change-a-blend.

Grind a very light coffee.
Pulse 'til hell freezes over.
Grind a very dark coffee.
See any light spots?
Pulse 'til hell freezes over.
Grind a very light coffee.
See any light spots?

cafeIKE wrote:Taking the above at face value, one should adjust the burrs continuously while grinding a single dose.

Maybe so, if the goal is to produce a grind of specific consistency, and if that grind has a noticeably different taste. In your case, it does. In my case, it doesn't.

My completely unscientific counterpoint is that, single-dosing with my un-hoppered and de-dosered SJ, I'm able to leave my grind setting at a single point for about 3-4 days and produce shots that are quite consistent to my palate. Then again, my bean selection as of late has been limited to some home-roasted Sidamo and Costa Rica, and some Metropolis Redline. All of my homeroast is usually gone within 5 days of roast, and the Redline doesn't last more than a week past roast. I weigh both before and after grinding, and generally lose 0.0 to 0.1g in the process. According to the user manual for my scale, this appears to be within limits for error (±0.1g at <50g), and is small enough that it may not be significant. For all intents and purposes, I'm not losing anything during grind.

I've gone through combinations of the above-mentioned methods, sometimes out of necessity, and still don't notice the differences in flavors to which Ian (and a few others) are referring.

With my single-dosing method, shots :

• aren't finicky at all, because my OCD has me measuring pre- and post-grind. With doubles, I've tried and like the Sidamo at 14.5-15.0g, whereas RedLine tastes better to me in the 14.0-14.5 range. Updosing does me no favors. I had very little good luck finding a great sweet spot with the Costa Rica, but then again it didn't knock me out as a great cupping coffee either.
• get distributed into my basket with no need for WDT or other measures. Maybe I'm lucky, or maybe it's the 2L bottle hack...
• the taste is balanced and consistent through the life of the batch
• spritz only when I've not been paying attention to dose/tamp

I don't think that Ian or anyone else is delusional, and re-state my position that this may be much more equipment specific than anyone has yet considered. Either that, or my taste buds are shot from drinking too much espresso. This may ultimately be more damning of my palate, but as Popeye says, I yam what I yam.

-s.

OT: Ian, could you clarify something?

cafeIKE wrote:Actually, not. I grind into the bin before every shot. The 'stale' coffee is the same.

cafeIKE wrote:At EOD yesterday I pulsed the dregs out of the MKX, dusted and swept it clean. As is my habit, before charging with Black Hand today, about 18 hours later, I pulsed the MKX again. A full 5.7g of detritus my reward.

So your MKX retains ~5g of grinds? I'm laboring under the impression that espresso roasts are roasted to within a 15°F range, and the color difference would be slight at best. I'd personally have a hell of a time trying to see the difference between old/new. How do you tell? When you switch coffees, are you pulsing until you don't see anything come out from the old, then adding new? If so, how does this address the grounds that may be present within the chamber?

Come to think of it, this makes me wonder about another issue. With single-dosing, I don't worry about half-crushed beans that are hanging out in the throat or burrs. I know that the Mazzer hoppers have a gate that allows you to cut off the supply, allowing the beans between the gate & burrs to grind, then clearing. Given that these were designed for commercial environments, I'd imagine that this was designed so that, at the end of a day, the PBTC could close the gate and save a majority of whatever was left in the hopper for the next day, thereby minimizing wasted beans. For the SJ, this would be around 40-50g of waste. A pittance for a shop, but for those of us playing along at home, that's a couple of triples, or 3 doubles! I suppose that you could chop the hopper for home use and just close the gate per shot, but that would make it act like a single-doser, as you'd lose the benefit of bean pressure. hmm...

Not having any experience with other hopper designs, is this a common feature? If not, then what effect might there be with using half-ground beans as part of your shot?

-s.