So as I'm learning espresso, my first major breakthrough was learning to dial in by yield. Get grind pretty close, then use yield to get a good shot. This has been simple, easy, quick. It's incredibly obvious whether you need to pull longer or shorter for more or less extraction.

But where I'm still lost, is how to adjust a parameter that will affect flow rate, such as grind or dose. Is there a way to know where you're at on the "extraction curve" after only pulling a single shot? For example my shot this morning. Breville infuser + niche, 2:1 on a light roast Ethiopian in 40s using full PI mode, first drop at about 15s. Seems reasonable enough, but the shot is under extracted - too acidic, slightly sweet. Just a small touch of astringency long after drinking.

So the question is, how do you know if you need to grind finer or coarser? To my understanding, in theory, you could already be too fine and flow is too slow for proper extraction. Or, on the other hand, the beans in question may do better with even finer grind and longer shot.

The straightforward answer seems to be try a few shots in each direction to see (seeing as I'm using niche, would be easy to go half step coarser or finer). I'm curious if others have a better way that wastes less coffee. Is there something about the taste that tells you whether you need finer grind or faster flow? Is it just experience on your equipment with different bean varieties?

Basically i am trying to develop a better idea for how to pull GREAT shots. Adjusting yield has been very easy to get good shots, but I feel like i probably need a better understanding of dialing in the remaining parameters, and when to do so, to get the best shots possible.

Dose is a last resort if you have a coarsely stepped grinder in my opinion, so let's put that aside.

I generally do the following:
- Pick a higher temp for light roast / lower temp for dark roast as a default start
- Get grind in the ballpark (very roughly, just not a 15s shot and not a 45s shot)
- Adjust based on ratio
- Tweak grind up/down
- If desperate make big temp changes to try to fix something residual

If you're pulling the shot you described - 40s 1:2 light roast and it's sour, pull a higher ratio. That's far more effective in my experience than playing with grind at that point. Practically I might coarsen the grind slightly since that's a relatively slow shot already and you're pushing ratio longer but initially it's likely better to do one parameter at a time, ideally comparing shots back to back.

The up/down is required because different coffees taste different and interact with different grinder characteristics and machine profiles differently too. By the time "adjust based on ratio" is reasonable, the shot is already very good, so it's not "wasting" coffee at all. I also have a good idea of where to start - for the light roasts I drink it's usually around 1:2.2-2.3 as a starting point and then I can adjust from there. With all of this my first try at a coffee is usually perfectly drinkable, but sometimes a sink shot. Second is definitely drinkable and it gets better after that.

Finally, there are some traditional guidelines on grind finer for higher extraction, grind coarser for lower extraction - but they fall apart at some point depending on the bean/roast and the grinder. At some point you grind finer and your extraction goes down due to unevenness/channeling. That's the other reason to play around.

So, I've spent more time reading and i think Matt Perger's articles and dialing in method kind of cleared up my questions for the most part. Basically lock in dose, then yield, then grind. I was doing the first two, and then didn't know where to go from there, thinking grind would impact too many things and cause me to have to start over. But at least according to him, that's not the case.

So I'm not totally clear why you can't adjust grind first instead of resorting to a longer pull.... But for now i will just accept it due to my experience and move on.

lessthanjoey wrote: Finally, there are some traditional guidelines on grind finer for higher extraction, grind coarser for lower extraction - but they fall apart at some point depending on the bean/roast and the grinder. At some point you grind finer and your extraction goes down due to unevenness/channeling. That's the other reason to play around.

This is where i am still not clear (applies to Matt's technique as well. Let's say I've dialed in ratio and achieved the best balance i can. Now I'm looking to increase sweetness or develop better balance.

I was under the impression that you can grind too fine, still have even extraction, and lose extraction not sure to channeling, but instead just simply too slow of flow. That's what I've been taught. Perger seems to entirely ignore flow rate and simply say it comes down to contact time.

I guess the theory is irrelevant. What I'm practically asking is, if i think i need to adjust what "extraction curve" I'm on by grinding coarser or finer... How do i know if i can go finer or not? How do i know I'm already going over the cliff?

If the answer is channeling, then that's easy. Channeling is very easy to detect and i suppose i can assume i can grind finer until i start to notice hints of channeling. But if that's not true... Then i suppose that's what I'm still wondering.


I guess it's simple to go back to my shot today. My next course if action should be too increase yield to get the most extraction before bitterness creeps in. Taking the "channeling" theory as true, i can either grind finer if i feel i need more sweetness, or if there's always a touch of channeling detectable, then instead i need to grind a touch coarser (or pay extra attention to prep). Is that pretty much it?

cgibsong002 wrote:So, I've spent more time reading and i think Matt Perger's articles and dialing in method kind of cleared up my questions for the most part. Basically lock in dose, then yield, then grind. I was doing the first two, and then didn't know where to go from there, thinking grind would impact too many things and cause me to have to start over. But at least according to him, that's not the case.

So I'm not totally clear why you can't adjust grind first instead of resorting to a longer pull.... But for now i will just accept it due to my experience and move on.



This is where i am still not clear (applies to Matt's technique as well. Let's say I've dialed in ratio and achieved the best balance i can. Now I'm looking to increase sweetness or develop better balance.

I was under the impression that you can grind too fine, still have even extraction, and lose extraction not sure to channeling, but instead just simply too slow of flow. That's what I've been taught. Perger seems to entirely ignore flow rate and simply say it comes down to contact time.

I guess the theory is irrelevant. What I'm practically asking is, if i think i need to adjust what "extraction curve" I'm on by grinding coarser or finer... How do i know if i can go finer or not? How do i know I'm already going over the cliff?

If the answer is channeling, then that's easy. Channeling is very easy to detect and i suppose i can assume i can grind finer until i start to notice hints of channeling. But if that's not true... Then i suppose that's what I'm still wondering.


I guess it's simple to go back to my shot today. My next course if action should be too increase yield to get the most extraction before bitterness creeps in. Taking the "channeling" theory as true, i can either grind finer if i feel i need more sweetness, or if there's always a touch of channeling detectable, then instead i need to grind a touch coarser (or pay extra attention to prep). Is that pretty much it?

You can adjust grind before yield, but it's just a weaker knob (within reason) so it doesn't really make sense to do so. Some exceptions might be if you're targeting a specific yield for known reasons and you know you can make it work then fine, just start at that yield and move right onto grind.

For example, I pull 1:4-1:5 flat 4ml/s shots (Rao Allonge style) sometimes. I know I want a yield in this range, I've usually played with the coffee first so have a rough idea of whether I want more like 1:4 or 1:5, and then I'm dialing grind to get the right peak pressure. Another example would be if you really wanted to make something work at a known low ratio, like 1:1.5. You could setup your shot at that ratio and play with grind to see if you could make it work.

But, all of the above are kinda exceptions to a normal flow and your goal isn't "trying to find a good balance for a medium-size drink" - it's specifically pushing one end or the other of ratio first, and then trying to make that work.

Flow rate ~ contact time/ml at higher ratios. At lower ratios? All the water gets dumped in the puck first, so yeah, shot time becomes kinda contact time, but the first drops out get a lot less contact time than the later ones? It gets complicated. Using a DE1, I can see flow rate through a shot. First order, for basic shot profiles (not a long PI bloom, not a super short ratio, etc), I think that flow rate better indicates the extraction than time.

Re: too fine? if you're into lighter stuff I think too long contact also muddies flavors, reduces clarity, so that'd be one line. I just find I prefer faster shots with my gear and beans. Assuming that's not an issue, look for black areas on the bottom of the puck indicating underextracted areas. If you're seeing those you're definitely grinding too fine and you need to coarsen it up to regain evenness. Beyond that? Taste...

But yeah, your final sentence is pretty much there too. When you have the time (which I'm sure isn't nearly all the time) it's really instructive to pull back to back shots with a small tweak and compare them as they cool. I think my good shots are far better merely warm than hot, and that makes it easy to compare two of them side by side and understand what's changing.

Finally - if you're making small changes, it's worth making sure you can pull roughly the same shot a few times in a row when trying to do so. If your random shot variation exceeds your deliberate shot variation then all the experimentation is worthless.

I guess part of my issue with grind is it seems i keep being to go 1:3 to get a balanced shot. Maybe that's just the case for the beans I'm using. I'm just paranoid I'm needing to pull longer to compensate for always grinding too fine. I'm pretty much always 35+seconds, but the ramp to full pressure on the infuser is 10+ seconds, so that seems reasonable. On the coffee in question i went coarser and pulled identical shots, with a noticable increase in acidity. So i think that shows me i wasn't yet over the cliff.

cgibsong002 wrote:I guess part of my issue with grind is it seems i keep being to go 1:3 to get a balanced shot. Maybe that's just the case for the beans I'm using. I'm just paranoid I'm needing to pull longer to compensate for always grinding too fine. I'm pretty much always 35+seconds, but the ramp to full pressure on the infuser is 10+ seconds, so that seems reasonable. On the coffee in question i went coarser and pulled identical shots, with a noticable increase in acidity. So i think that shows me i wasn't yet over the cliff.

What's the coffee?

cgibsong002 wrote:I guess part of my issue with grind is it seems i keep being to go 1:3 to get a balanced shot. Maybe that's just the case for the beans I'm using. I'm just paranoid I'm needing to pull longer to compensate for always grinding too fine. I'm pretty much always 35+seconds, but the ramp to full pressure on the infuser is 10+ seconds, so that seems reasonable. On the coffee in question i went coarser and pulled identical shots, with a noticable increase in acidity. So i think that shows me i wasn't yet over the cliff.

When you ground coarser, did you decrease the preinfusion time? I would think you'd have to do that to slow down the flow to get the same shot time.

You may not be grinding fine enough. Or you may be grinding fine when you don't need to. I realize this sounds contradictory but read on.

On the shots where you have to go 1:3, you must be grinding pretty fine because 10+ seconds of preinfusion isn't giving you a gusher. Remember, the whole point of long, slow preinfusion is to allow you to grind a lot finer, hopefully to extract more fully. The long, slow preinfusion allows the puck to open up so the fine grind doesn't choke the machine. But you're not running the shot particularly long, at least not by the parameters I use with light roasts, so it could be that you're not grinding fine enough and not preinfusing long enough. I grind very, very fine and set the preinfusion flow rate to about 120 ml/min, which translates to about 20 seconds of preinfusion. Total shot time can range from 50-70 seconds, depending on the roast.

If the roast is light or very light, try grinding finer, increasing preinfusion time and total shot time, aiming for a 1:2 ratio. If it's a medium roast, you might consider going the other way and pulling a shot with "standard" preinfusion. Grind much coarser and use the natural preinfusion ramp time of the machine. For most machines that's 2-3 seconds.

Currently Heart Ethiopia Beriti. Its light roast Ethiopian (so 1:3 seems to makes sense), but it's quite well developed with low fruit and acidity. But I've also done a few other beans so far ranging from light to medium and i just don't seem to be getting good results on anything around 1:2.

cgibsong002 wrote:Currently Heart Ethiopia Beriti. Its light roast Ethiopian (so 1:3 seems to makes sense), but it's quite well developed with low fruit and acidity. But I've also done a few other beans so far ranging from light to medium and i just don't seem to be getting good results on anything around 1:2.

That all sounds perfectly reasonable.

I'm not sure if the infuser's preinfusion can really be defined as such. Really what it does it's just slowly ramp the pump. So using that mode, it might not build to full pressure until 8-12s. But to my knowledge there's no way to extend the preinfusion time.