Kruve: Further Analysis on the State of the Art

Sieve technology is not a hot topic, but in espresso making, it should be. Sifting makes a huge difference and levels the playing field amongst grinders. Sieve technology is very old, but to be useful for coffee (especially espresso), the particle size needs to be tightly specified. Kruve’s sifter is an interesting first for coffee, so let’s take another look particularly with respect to alternative sieve technology (my initial look is here).

Laboratory Sieves

https://www.coleparmer.com/i/cole-parmer-sieve-shaker-accepts-up-to-7-full-height-3-sieves-115v/

Scientific sieves have been in use for a long time. There are a handful of companies selling them, and they typically cost $15 to $50 per sieve. Kruve costs $15/sieve, and they also sell sets. Some lab sieves can be attached to vibration tables to making sifting automatic. This is not cheap and not convenient in terms of space. The larger ones don’t work well for coffee, but the smaller test sieves might.

Then the question is accuracy. Current lab sieves are governed by ASTM standards. These standards are based on sieves with square openings as they are made by overlaying two wire meshes. This means even if the sieves were made in perfect spec, a variety of particles up to the sqrt(2)*width of the hole could make it through.

So even with their faults, Kruve still produces the best sifter on the market for the price. However, that doesn’t mean we should have a look deeper.

Sieve Analysis Continued

My previous analysis had some unanswered questions:

  1. How accurate are image based measurements in determining the hole distribution?
  2. How repeatable is this method with respect to minor changes in the camera angle?

Accuracy

Let’s first look at the accuracy for measuring hole distribution. For the 200um sieve, the holes are so small that they average about 4 pixels total. So, let’s put these sieves on an iPad Mini instead giving around 12 pixels for hole area.

As a sanity check, let’s zoom in on a region of holes for the 200um (image above). So the boxplot for this one is not the real distribution in terms of real world coordinates, but merely to show there is a spread in the data. I could take better measurements to estimate the whole size for that image, but it was a quick sanity check only.

The center’s for these sieves are off a bit, but my real concern is the distribution spread. So let’s examine the 800um sieve and take images further away to help understand the accuracy at 200um. We could zoom out enough to where each hole for 800um is only a few pixels in total volume similar to the 200um on an iPad Mini scale.

On this plot, the um/pixel is shown so smaller means the image of the screen is smaller in size.

This shows different levels of zooming out, and the center location can be skewed, but there is still a distribution around it.

Let’s look at one of the images showing hole size on the original image. The below distribution of hole size shows an interesting pattern that is persistent no matter how the image is taken. In this case, the image is angled which could give the notion that the larger holes are on the bottom of the filter when it is an optical issue.

Repeatable

Let’s take a look at multiple different images with slight variations in camera angle on the 800um screen. Again, while there is an offset in the center of the distributions, they all have a spread. Some are distributed more to the right or to the left, and that was caused by having a large angle with respect to the flat surface.

Comparison to Other Sieves

The real question to answer is: Does the Kruve sieve sift better than what’s currently available at a higher price for lab testing?

Turns out, they do pretty well. I haven’t put these sifters through the same analysis, but according to their spec sheet and design, they are limited due to design and allow larger particles to get through than Kruve.


I don’t think Kruve will ultimately be the optimal design for a sifter with respect to coffee. After using it daily for a month, it adds time to making espresso, so a better design would make the process automatic and faster. I have a few thoughts on the topic.

I believe Kruve is only an introduction to a new way to do coffee (not just espresso), and I’m positive the best design is already coming to life. We will one day see the sieve become standard at quality coffee shops.


If you like, follow me on Twitter and YouTube where I post videos espresso shots on different machines and espresso related stuff. You can also find me on LinkedIn

Further coffee readings of mine:

My coffee setup

Artisan coffee is overprice

The Tale of the Stolen Espresso Machine

Affordable Coffee Grinders: a Comparison

Espresso : Grouphead Temperature Analysis

Espresso filter analysis

Portable Espresso: A Guide

Kruve Sifter: An Analysis

Overthinking Life

Thinking too much on Philosophy, Math, Science, Politics, Work, and Life

Robert McKeon Aloe

Written by

I’m in love with my wife, my kids, espresso, tomatoes, cooking, engineering, talking, family, Paris, and Italy, not necessarily in that order.

Overthinking Life

Thinking too much on Philosophy, Math, Science, Politics, Work, and Life

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade