John for those baskets in the shot did you overfill and level before weighing? Or did you just fill them up as close to the rim as you could? I ask because I overfill and level and the weight (per my scale) doesn't vary as much as the four baskets you show.

Also beginning to wonder if "acceptable" dose variation increases as overall dose increases? For example would the same amount of variation that would destroy a 14g shot hardly impact say a 18g or 19g dose?

TimEggers wrote:Also beginning to wonder if "acceptable" dose variation increases as overall dose increases? For example would the same amount of variation that would destroy a 14g shot hardly impact say a 18g or 19g dose?

This is going to vary by machine. Some machines do not tolerate updosing beyond a fairly limited point (say 16g). In this case, going much above 16g could effect some machines more than others, independent of the percentage of dose variation.

In the case of machines that "don't care much" about updosing, then I think the impact of variation around a given (intended) dose would vary by approximately the percentage of deviation. 0.5g variation in an intended dose of 14g is equal to around 3.6%, by weight and presumably by volume. In the case of an intended 20g dose, 0.5g would represent around 2.5%. This doesn't, on the face of it, sound like much, however it is a variation of around 45% in terms of the "error" in dosing represented.

Dosing is but one of the variables in making espresso. I don't know about anyone else's experiences, but even from early on, I learned that minor variations in dosing had much more impact on shot quality than anything I did with regards to tamping, distribution, and even the grind level.

I seldom weighed coffee when I used to updose. I used to experience a lot of inter shot variation when I updosed, but in all fairness I had less capable grinders than I have now, so my ability to reach a conclusion on this is limited by having too many variables to consider.

ken

TimEggers wrote:John for those baskets in the shot did you overfill and level before weighing? Or did you just fill them up as close to the rim as you could? I ask because I overfill and level and the weight (per my scale) doesn't vary as much as the four baskets you show.

I was trying for variation, as an illustration of the difficulty in accurately eyeballing a dose.

With a basket funnel (either the ever-classy yogurt cup, or one of those tacky Orphan Espresso aluminum derivatives ) and a scale, it's easy to dose a specified weight. For the 18g dose, I tapped downwards once or twice to get the grinds roughly level with the rim. For downdosing, I still find a a WDT tool (aka dissecting needle) useful for fluffing up the grounds (and also for distributing below the rim).

TimEggers wrote:Also beginning to wonder if "acceptable" dose variation increases as overall dose increases? For example would the same amount of variation that would destroy a 14g shot hardly impact say a 18g or 19g dose?

Good question, and I don't think there is a simple answer. But in my experience, changes in flow are not linearly related to dose, especially at extreme doses. In going from 18g to 19g, you might choke the machine. This is worth investigating, and may be dependent on a variety of factors (grouphead geometry, basket size and shape, type of coffee, grind setting).

TimEggers wrote: Also beginning to wonder if "acceptable" dose variation increases as overall dose increases? For example would the same amount of variation that would destroy a 14g shot hardly impact say a 18g or 19g dose?

I will give a clear Yes to this question. This was my epiphany that I mentioned in a prior post. I have been experimenting (with my unsophisticated lowly scale) and confirmed that conclusion over and over since then.

But I will admit to ken and John's point, and allow for mitigating factors.

Ken Fox wrote:..This is going to vary by machine. Some machines do not tolerate updosing beyond a fairly limited point (say 16g). In this case, going much above 16g could effect some machines more than others, independent of the percentage of dose variation.ken

RapidCoffee wrote:...This is worth investigating, and may be dependent on a variety of factors (grouphead geometry, basket size and shape, type of coffee, grind setting).

These variations that "destroy" shots - are we really talking about something beyond the means of intelligent and simple method to eliminate?

For example, I am drinking the tail end of an extremely nice Monsooned Malabar - very light roasted as is Londinium's style - and therefore a bit fussy about dose. Put 7g in a single basket and it will channel enough to give you a sink shot. 8g behaves perfectly, but 8.5 gives you a wonderful, rich, spicy and creamy sensation. Pack the single basket to 9 and the shot cloys to becoming only just about acceptable.

This is not an unusual range between extremes in my experience, so we're talking 2g on a "classic dose" single shot for the sink shots at either end. Not the sort of variations of around 0.2-0.5 that people more typically find on a timed grinder or by following whatever fill the basket and level over the top or with doser lid etc. they prefer......

Do I live in Wonderland or are we getting into the minutiae of accuracy for the sake of it?

michaelbenis wrote:These variations that "destroy" shots - are we really talking about something beyond the means of intelligent and simple method to eliminate?

For example, I am drinking the tail end of an extremely nice Monsooned Malabar - very light roasted as is Londinium's style - and therefore a bit fussy about dose. Put 7g in a single basket and it will channel enough to give you a sink shot. 8g behaves perfectly, but 8.5 gives you a wonderful, rich, spicy and creamy sensation. Pack the single basket to 9 and the shot cloys to becoming only just about acceptable.

This is not an unusual range between extremes in my experience, so we're talking 2g on a "classic dose" single shot for the sink shots at either end. Not the sort of variations of around 0.2-0.5 that people more typically find on a timed grinder or by following whatever fill the basket and level over the top or with doser lid etc. they prefer......

Do I live in Wonderland or are we getting into the minutiae of accuracy for the sake of it?

You are describing the area (17g) that we, or at least I, have been describing as updosing. From the description you are giving; I am understanding that at 8.5g the shot is near perfect but at 9.0 it is borderline acceptable and at 8g you say nothing of taste only behavior. This is a variation of +/-0.5g so I could assume that +/- 0.3g, or even perhaps +/- 0.2g, would be a safer limit to get good shots reliably. Of course this is for a single shot, for which I have very little experimental knowledge to compare with.

This whole thread started as a discussion on the possibility to get good shots without the full grooming procedure. For my part, I have become convinced that I can get away with it using my machine and grinder, but in doing so I am cornered into dosing precisely.

I will now start doing a few experiment where I will vary the dose a lot (more than +/-0.5g using a scale) but will groom the coffee with more discipline and see what that gives me. Never tried that. Hey I am a knewbie at this level of control and mastery (which I have none)!

michaelbenis wrote:These variations that "destroy" shots - are we really talking about something beyond the means of intelligent and simple method to eliminate?
...
Do I live in Wonderland or are we getting into the minutiae of accuracy for the sake of it?

This strikes to the heart of the matter for me. No, we're far beyond discussion of complete meltdowns. No, we're not discussing minutiae. Small changes in dose can have a profound impact on taste in the cup. And yes, there is a simple way to manipulate dose accurately: use a scale.

I tell you three times: I am not lobbying for weighing every dose. But I dislike seeing those who do, being cast as refugees from Planet OCD. Dose matters.

michaelbenis wrote:These variations that "destroy" shots - are we really talking about something beyond the means of intelligent and simple method to eliminate?

For example, I am drinking the tail end of an extremely nice Monsooned Malabar - very light roasted as is Londinium's style - and therefore a bit fussy about dose. Put 7g in a single basket and it will channel enough to give you a sink shot. 8g behaves perfectly, but 8.5 gives you a wonderful, rich, spicy and creamy sensation. Pack the single basket to 9 and the shot cloys to becoming only just about acceptable.

This is not an unusual range between extremes in my experience, so we're talking 2g on a "classic dose" single shot for the sink shots at either end. Not the sort of variations of around 0.2-0.5 that people more typically find on a timed grinder or by following whatever fill the basket and level over the top or with doser lid etc. they prefer......

Do I live in Wonderland or are we getting into the minutiae of accuracy for the sake of it?

Not meaning to get too far off into the weeds, however, Monsooned Malabar is probably not considered, by most SO fans, to be a good or representative coffee for use as an SO espresso. In addition, the grind level required and the way it behaves when used by itself to make an espresso, are probably not representative of how the great majority (maybe ALL) non-aged coffees will behave in a PF when making espresso.

ken

Of course dose matters. And hopefully no one is casting anyone as refuges from anywhere.

So, let's get down to the heart of it, having pretty much all agreed that a scale is a very useful marker when setting things up, but not necessarily something that needs to form part of the preparation of every shot.

The heart of it, in my opinion, is to keep things practical. Including what I think could potentially be a very useful discussion.... quite simply about how to keep the cup we want sufficiently repeatable to enjoy it every time.

So, the question is, simply "how small a variation can we get away with?"

This is obviously going to depend not only on the dose as mentioned above, since that's not an isolated preference (e.g. for a short single or a voluminous double). It also depends on the bean, especially if we are drinking single origins. I challenge anyone to dose a Sidamo the same way they would a Brazilian yellow bourbon.... and enjoy it.

So we take our choice on the dose we prefer for our favourite bean and then we think about how we are going to set that up with sufficient repeatability.

Most people posting here mentioned getting variations of largely within 0.5g if they are not weighing every dose. I'd say that was close to the cusp, but not over the edge, whereas variations of a couple of tenths of a gramme here or there are neither here nor there.

Would you agree with that? And if some of the beans you use are more sensitive to dose, it would be interesting to know what beans they are: interesting as a marker for people to look out for.... Even with a Yirgacheffe, which I tend to dose at barely over 7g, my Nino gives enough consistency on the timer for all the shots to be good,

I don't see this as a matter of drawing lines in the sand on method so much as comparing experiences.....

Cheers

Mike

Ken Fox wrote:Not meaning to get too far off into the weeds, however, Monsooned Malabar is probably not considered, by most SO fans, to be a good or representative coffee for use as an SO espresso. In addition, the grind level required and the way it behaves when used by itself to make an espresso, are probably not representative of how the great majority (maybe ALL) non-aged coffees will behave in a PF when making espresso.

ken

I'll concede much of that, but it meant I could quote figures without relying on my memory, which is considerably less accurate than any pair of scales

And Londinium's light roast makes it less atypical than it often is, since there is much less oiliness to bind the grounds.

I'm now waiting to see whether you tell me that Yirg isn't a good example of a SO one would generally want use for an espresso either....