Physics of tamping an uneven bed

wilburpan

Good tamping practice says that you should level the pile of ground coffee in the basket before tamping, in order to get a homogenous puck.

It’s also said that tamping is only good up to a certain pressure, and beyond that, the coffee is not compressible any more, so tamping harder has no effect.

So today I tried something for grins. I ground coffee for my espresso, deliberately put it into the basket so that it was piled higher on one side. Then I tamped as hard as I could, resulting in an even puck. Here’s a sketch to show what I’m talking about.

Here’s my question. Given that the coffee is incompressible after a certain point, and the bed is even, why wouldn’t this provide a good cup of espresso? If the bed is even, tamping the previously uneven pile of coffee must have moved coffee particles from the higher side to the lower side. And since I used enough pressure with my tamp, the coffee must be packed evenly.

I thought one possibility is that I was wrong about the incompressibility issue, and that it was possible to have a situation where the bed was even after the tamp, but the coffee was still unevenly distributed. If this was the case, then there should be less densely packed areas of the puck. But probing around with a toothpick (aka my excellent WDT tool 😆) didn’t find any soft spots.

The only thing I didn’t do was to make espresso with this puck. I scrambled the puck with my toothpick, leveled the ground coffee, and tamped. My espresso was yummy. But I probably will pull a shot while tamping a deliberately uneven bed soon to see what happens. In the meantime, I was wondering what you all might have to say.

hornbyben

wilburpan if they started uneven but were level after tamping then the density on the side that was higher must be more than the side where it was lower. Once the higher side is fully compressed the tamper is prevented from going deeper, but that doesn’t mean the lower side has reached this point.

The higher density side provides more resistance to flow than the lower, so more water will go through the lower side, giving a shot like tompoland described.

tompoland

Sorry to say I got lost trying to understand the question. Not your bad, just my feeble brain.

What I can tell you is this little story (gather round the campfire kids)…

I was asked to visit the home of a long time espresso enthusiast to see if I could help with getting a reasonable espresso out of a new machine he’d bought.

First I asked him to show me his routine which including tamping with a standard tamper. The tamp was very uneven and I pointed that out to him. He protested with something like “I’ve been making espresso like this for 20 years!”. That annoyed me. I mean, I travelled 35 minutes to get to his place and here he was being ultra defensive.

So being the impatient person that I am I informed him “in that case you’ve been drinking sxxx espresso for 20 years”

I explained to him that when he poured the shot he’d see dark espresso on one side of the naked basket and light on the other and that he would consequently have both bitter and sour flavors in the cup. And of course that’s exactly what happened.

Nothing surprising there I’m sure.

And that’s why I use a self levelling tamper: to minimize the likelihood of an uneven tamp.

I’d love to understand your question though so I’ll stay tuned for someone smarter posting.

wilburpan

tompoland First I asked him to show me his routine which including tamping with a standard tamper. The tamp was very uneven and I pointed that out to him. He protested with something like “I’ve been making espresso like this for 20 years!”. That annoyed me. I mean, I travelled 35 minutes to get to his place and here he was being ultra defensive.

So being the impatient person that I am I informed him “in that case you’ve been drinking sxxx espresso for 20 years”

Maybe what I’m seeing is because I make sxxx espresso and don’t know it. 😆 I admit that could be a real possibility.

But I think my case is a bit different because I start with an uneven bed, but tamp so that when I’m done, the bed is even.

Ernie1

If I try and tamp an uneven bed of coffee I get a flat surface but one that’s slanted. If I use a deep basket with a ridge, it’s easier to see how slanted it is.

It’s rare this will actually affect my output enough to be like ‘ugh I can really taste this uneven tamp’ but for consistency I use a leveller before tamping.

Interestingly the other day I used the leveller and forgot to tamp and the coffee came out great…

MediumRoastSteam

Ernie1 Interestingly the other day I used the leveller and forgot to tamp and the coffee came out great…

That never happened to me. I once levelled and forgot to tamp, and the coffee came gushing out. An unquestionable sink shot, even for my standards.

I remember reading on various forums back in the day some people only did a very gentle and light tamp. That never worked for me.

wilburpan

hornbyben if they started uneven but were level after tamping then the density on the side that was higher must be more than the side where it was lower. Once the higher side is fully compressed the tamper is prevented from going deeper, but that doesn’t mean the lower side has reached this point.

The higher density side provides more resistance to flow than the lower, so more water will go through the lower side, giving a shot like tompoland described.

That’s what I thought, except for three things.

First, given enough force, any differential in density should go away, given that coffee is not compressible. This is basic physics.

Second, as I mentioned, I deliberately left the bed uneven, and tamped by pushing down really hard on the ground coffee so that the bed was level after tamping. I poked around with a toothpick, and didn’t find any soft spots.

Finally, I tried this out yesterday. I made a shot like I normally do, leveling the bed, and doing WDT so that things were even before tamping. The espresso was great. I gave that one to my wife, because…

For the next shot I deliberately left the coffee uneven in the basket by quite a lot. I then tamped with enough pressure to make the bed level. No WDT, obviously, because that would have leveled out the bed. I pulled a shot for myself. (That’s why I gave the first shot to my wife, in case things turned out badly.)

The shot wasn’t perfect, but was still very good. I’d put the shot with the uneven bed well within the range of espressos I’m used to getting with my usual technique. The interesting thing is that the extraction time was actually a little longer than what I normally get. This indicates that there’s no soft spot or less dense area that the water could get through more easily. (Point of explanation — I have a 9Barista, so there’s not any manipulation of the extraction once it happens. The length of the extraction depends on the dose and grind of the coffee, and also on how well-packed the puck is after prepping and tamping.)

I’m going to keep leveling the bed and using WDT, but these findings were interesting to me.

Dusk

MediumRoastSteam

I wonder if this could be due to how “gently” the shower screen distributes water. Perhaps a good tamp can make up for water that isn’t evenly distributed over the coffee that would otherwise immediately find the path of least resistance and cause channeling??

coffeeb

There appear to be 2 discussions happening here:

An uneven level of tamped coffee in the basket which would logically mean different path length for water going through it, hence differences in the coffee - I don’t think this is what the OP is talking about here as they are saying it is level
An apparent even level of tamped coffee (i.e. top surface is parallel to the bottom of the basket) but which was achieved by force

For the latter, if you apply sufficient force in the right axis, then you should be able to provide a level surface as the coffee will move around. Without resorting to actual physics of particulate beds which are well outside my understanding (even before you add liquids), this doesn’t necessarily mean that it will be packed more densely on one side, it will depend on the ‘flow’ of the coffee when you press it.

In effect you are moving some of the coffee from one side of the basket to the other rather than just squashing it more. Although obviously all of this would depend on the levels of force you apply.

CoyoteOldMan

Ernie1 Pressure from the water will definitely have a compacting effect, but there’s a lot of stuff going on beyond simple pressure once you wet the puck: bits get dissolved, bits get dragged by the water flow, bits swell as they absorb the liquid and bits stick to each other as water acts as a (weak) glue.

I think coffeeb is totally right, though, that the fact that coffee is “incompressible” does not mean it won’t move. A very easy visual proof is to get a glass with caster (or icing) sugar or flour inside it, and to use a tamper (or something else) to squish it. Even if there is no air and the powder is as compressed as can be, it will move around like a very viscous liquid.

Ernie1

I know next to nothing about physics but if a recommended tamp is about 8lbs per square inch, and a 9-bar pump is 130lbs per square inch,

Factoring headroom in, even though it’s a liquid that will find the path of least resistance, could water have somewhat of a compacting effect also?

I’m already confused writing this.

Cuprajake

on all my levers i have been able to use a spinner and just pull a shot, i think it depends how deep the spinner goes,

Ernie1

Also do different machine types have different ramp-up times? I should know this having owned more than one machine… but I can’t remember.

The LM seems to literally blast water the very second the pump is activated which I’d imagine has a more compacting effect than one that starts less aggressively.

DavecUK

Ernie1 Also do different machine types have different ramp-up times? I should know this having owned more than one machine

They most definitely do…and this makes a big difference.

coffeeb

From a general perspective there’s all sorts of confusing things going on when you pressurise things, particularly in this case. The pressure will push on the coffee grounds from top to bottom as though trying to push the coffee through a pipe, but the portafilter then restricts the flow which means pressure builds up in there. This means you have a pressure forcing stuff through the holes, but also a pressure (hydrostatic-ish) pushing equally across the whole surface of every single coffee ground. This forces water into the porous coffee ground, displaces air and dissolves things.

This is a pretty interesting video with a transparent portafilter showing what goes on:

Tije’s Transparent Portafilter clip3 - YouTube

You can see that as it fills up, some of the top layer gets mixed up into a suspension - I presume the more aggressively the water is pumped in, the more of this material gets disturbed, so you have less material in the ‘solid’ puck…

MWJB

wilburpan First, given enough force, any differential in density should go away, given that coffee is not compressible. This is basic physics.

If this were true, tamping flat, without distributing would be as efficient as any other method. I think quite a few people have tested and found that distributing before tamping increases efficiency. I was in conversation with friend who supplied with data (for his machine & grinder) that showed hard tamping definitety lowered extraction…so being flat isn’t all there is to it.

hornbyben

wilburpan interesting. I’ll have to get my non-levelling tamper out and see what I get from a lopsided distribution

CoyoteOldMan

MWJB If this were true, tamping flat, without distributing would be as efficient as any other method.

Hm - no. The amount of force/pressure needed to shift the grounds against their mutual friction may be well above the amount of force/pressure that allows maximal (or optimal - not necessarily the same) extraction. It remains a physical fact that, given enough force, the differential in density will go away.

With all respect (and I mean this), you keep confusing statements that are about the basic physics of things with statements about what is optimal for coffee making. See our prolonged discussion on the weight of burrs.

MWJB being flat isn’t all there is to it.

Definitely not, as far as having a good coffee is concerned. However, it seems that having a level puck pre-tamping isn’t all there is to it either. 😁

wilburpan

MWJB wilburpan First, given enough force, any differential in density should go away, given that coffee is not compressible. This is basic physics.

If this were true, tamping flat, without distributing would be as efficient as any other method. I think quite a few people have tested and found that distributing before tamping increases efficiency. I was in conversation with friend who supplied with data (for his machine & grinder) that showed hard tamping definitety lowered extraction…so being flat isn’t all there is to it.

Just to clarify, it did take more time and attention to make the uneven bed flat with just the tamper, so I’m not saying that there’s no benefit for leveling, nor am I saying that it’s just as efficient. Like I said, I’m going to keep leveling and doing WDT.

MWJB

CoyoteOldMan With all respect (and I mean this), you keep confusing statements that are about the basic physics of things with statements about what is optimal for coffee making.

It’s not so much confusing between physics & optimal coffee making, more focusing on the coffee making aspect, because that is the end result we are aiming for.

CoyoteOldMan

MWJB because that is the end result we are aiming for.

In which case - and again, with the greatest of respect - I think you incorrectly attribute to others the same singularity of purpose you have. Plus, some of us - even if we share that purpose - need to reason some things through step-by-step; either because we are less intelligent, or because we have less of an encyclopedic knowledge of coffee making and are thus not able to make the same ‘leaps’ in reasoning that you make.

MWJB

CoyoteOldMan I doubt that I am any more intelligent that most here, probably less so than plenty of members.