www.vanelis.com: top-notch espresso equipment and customer support

Titan Grinder Project - Page 5

Behind the scenes of the site's upcoming equipment reviews.

Link to "Titan Grinder Project"by k7qz on Sun Jun 10, 2007 3:48 pm

cannonfodder wrote:Houston, we have a problem. Never let someone that has only used a doser twice in their life run a commercial grinder. She emptied the entire hopper! A quarter pound of coffee shot to hell. I don't think she was pulling the doser level completely around so the flapper was not grabbing the spur on the hub and rotating the doser. She just stood there flapping in the wind. So I tell her she must not have pulled the doser handle all the way around. Mistake. 'Well I've never used one of those before but I did it just like you told me.' I don't think so hun, let me show you again (second mistake). Well, I never quite got that second round of cupping. I am sitting on the deck in the dark typing on the laptop. Oh bother.


Thanks for the effort expended on behalf of the HB community! An n=1 may have to suffice but IMO don't overlook the fact that you predicted or "called the shot" blind as anticipated.

OK now, let me think.... flowers. That's it, flowers. Flowers always seem to work for me when I find myself banished to the outer darkness of my deck. Perhaps those on HB who would like to see more regarding the Kony would even be willing to contribute to cannonfodder's flower fund... :D

Keep it coming!
k7qz
 
Posts: 217
Joined: Aug 18, 2005
Location: Port Angeles, WA

Link to "Titan Grinder Project"by cannonfodder on Sun Jun 10, 2007 8:11 pm

OK, so I do care that I only had one blind cup test. Lets face it, even a blind squirrel will occasionally find a nut. So this morning I managed to scavenge enough beans from the Super Jolly to give the Kony/SJ another go.

I went about it a little different this time, the wife 1.0 interface was bypassed. I took the spring out of my portafilter so I could use two baskets already loaded with coffee. I took two identical baskets, put a small K on one and an S on the other, then covered them with a small piece of tape to hide the mark. I ground and dosed from the Kony into the basket marked K and did likewise for the Super Jolly. I kept the same 16 gram dose, leveled out the baskets and put them on a table.

This time I had my son bring me one of the two baskets. I dropped it into my spring-less portafilter, tamped, flushed and pulled a shot just as the previous test. Then I sampled the shot and placed the basket on the grinds tray of the grinder I believe it came from.

Again, it was not much of a contest. I pulled the tape off the basket and had picked the grinder correctly. The Kony shot was bright with more acidic and less body in the cup. I could tell the beans had aged another day as the cup was slightly different. The Kony was a little harsher today. The drink grabbed my tongue for lack of a better term. A harder attack on the palate (I believe that is a cup term I have heard in regard to spirits). Again the Super Jolly was mellower but the floral and fruit was more prominent than yesterday. Still lots of body but not as clear and defined as the Kony. This morning I preferred the shot pulled from the Super Jolly.

I will be out of town on business all next week and not returning home until Saturday night so any further testing will have to wait. So far I have been pulling a traditional double just to keep things uniform. I think I will try some ristrettos when I return, which is my normal drink, 1.5-1.75 ounces in 30 or so seconds. The drip cup was also intriguing. I will have to try some traditional cupping and a press pot or two when I return home.
Dave Stephens
User avatar
cannonfodder
 
Posts: 3658
Joined: May 23, 2005
Location: Dayton, Oh

Link to "Titan Grinder Project"by Ken Fox on Tue Jun 12, 2007 8:25 pm

cannonfodder wrote:OK, so I do care that I only had one blind cup test. Lets face it, even a blind squirrel will occasionally find a nut. So this morning I managed to scavenge enough beans from the Super Jolly to give the Kony/SJ another go.

I went about it a little different this time, the wife 1.0 interface was bypassed. I took the spring out of my portafilter so I could use two baskets already loaded with coffee. I took two identical baskets, put a small K on one and an S on the other, then covered them with a small piece of tape to hide the mark. I ground and dosed from the Kony into the basket marked K and did likewise for the Super Jolly. I kept the same 16 gram dose, leveled out the baskets and put them on a table.

This time I had my son bring me one of the two baskets. I dropped it into my spring-less portafilter, tamped, flushed and pulled a shot just as the previous test. Then I sampled the shot and placed the basket on the grinds tray of the grinder I believe it came from.

Again, it was not much of a contest. I pulled the tape off the basket and had picked the grinder correctly. The Kony shot was bright with more acidic and less body in the cup. I could tell the beans had aged another day as the cup was slightly different. The Kony was a little harsher today. The drink grabbed my tongue for lack of a better term. A harder attack on the palate (I believe that is a cup term I have heard in regard to spirits). Again the Super Jolly was mellower but the floral and fruit was more prominent than yesterday. Still lots of body but not as clear and defined as the Kony. This morning I preferred the shot pulled from the Super Jolly.

I will be out of town on business all next week and not returning home until Saturday night so any further testing will have to wait. So far I have been pulling a traditional double just to keep things uniform. I think I will try some ristrettos when I return, which is my normal drink, 1.5-1.75 ounces in 30 or so seconds. The drip cup was also intriguing. I will have to try some traditional cupping and a press pot or two when I return home.


Dave,

Thank you for your efforts in trying to do at least a little blind tasting. I think this is valuable, although too limited in scope to "prove" anything. Perhaps when this project is finished, or at the end of it, a real, statistically valid blind tasting can be done on at least a couple of grinder pairs being studied here. It may be that if the differences in the shots are really so obvious, that the tester should be asked not only to pick a preference on the shots but also to try to identify which grinder produced it.

These kinds of studies are a huge PITA to design, and even worse to execute.

I am not volunteering . . . . .

ken
What, me worry?

Alfred E. Neuman, 1955
Ken Fox
 
Posts: 1057
Joined: Oct 28, 2005
Location: Idaho

Link to "Titan Grinder Project"by cannonfodder on Wed Jun 13, 2007 4:23 pm

The last stop for the kit is Jim's place. I believe he is planning on a much more substantial cupping regime. My attempts are strictly off the cuff amateur samplings and should be taken as such. I will try a few more while I have the Kony/SJ, unfortunately I am out of town this week on business so the Mazzer brothers are sitting idle this week.
Dave Stephens
User avatar
cannonfodder
 
Posts: 3658
Joined: May 23, 2005
Location: Dayton, Oh

Link to "Titan Grinder Project"by altoCalgary on Fri Jun 15, 2007 4:57 pm

Teme wrote:By the way, I also have the Illy book and in it he refers to a grinds distribution where both fines and coarse particles are present in appropriate proportions as being desirable. This is what the conical burrs appear to generate better than flat burrs (at least so it seems in the limited and non-scientific scope of the analysis).

Br,
Teme


Teme,

I believe you are referring to M. Petracco in Percolation, Chapter 7, Espresso Coffee: The Science of Quality, 2nd. Ed. 2005, where he essentially states that multi-size grading provides the most suitable particle size distribution for brewing espresso. If this is true, I would be interested in learning what the best tasting multi-size distributions look like and what grinders are capable of creating such distributions.

On the other hand, Dan Ephraim of Modern Process Equipment Corporation in a PowerPoint Presentation for one of the Roasters Guild Retreats argued that particle size and uniformity are essential characteristics of excellent coffee.

http://www.mpechicago.com/coffee/TechInfo/PDF/Roasters%20Guild%20Retreat%20Presentation.pdf

Which of these two perspectives is correct?


Ron
altoCalgary
 
Posts: 23
Joined: Aug 18, 2006
Location: Calgary

Link to "Titan Grinder Project"by darrylr on Sat Jun 16, 2007 1:20 am

altoCalgary wrote:Which of these two perspectives is correct?


Probably neither. Taste is too subjective. Same reason why it's neither "correct" to say espressos should be from blends or that they should be from single origin coffees.

Darryl
darrylr
 
Posts: 43
Joined: Oct 16, 2005
Location: Seattle

Link to "Titan Grinder Project"by altoCalgary on Sat Jun 16, 2007 12:48 pm

darrylr wrote:Probably neither. Taste is too subjective. Same reason why it's neither "correct" to say espressos should be from blends or that they should be from single origin coffees.

Darryl


Two of the top commercial coffee grinding technology manufacturers, Modern Process Equipment (MPE) in the USA and MAHLKONIG in Europe have developed automated machines that accurately create more than one uniform particle size in a run. This multi-modal processing is being driven by demand from the espresso pod and cartridge manufacturers. http://www.mpechicago.com/coffee/AboutUltrafine/Onsite_Articles/Coffee%20%20Cocoa%204-06.pdf

Teme has noted on his site recently that he is familiar with the Mahlkonig K30ES that has been used in the UK barista championships. In a TITAN-like Project taking place in Europe, the K30ES was one of the machines reviewed. http://forum.coffeed.com/viewtopic.php?p=8410 I found it interesting that one of the criticisms of the K30ES was that its grinds were too uniform!

If commercial grinding technology companies have been spending money to create multi-modal processing machines, where is the research that is informing this development? Can this research also be used to inform the TITAN project?
altoCalgary
 
Posts: 23
Joined: Aug 18, 2006
Location: Calgary

Link to "Titan Grinder Project"by HB on Sun Jun 17, 2007 8:02 pm

Follow-on discussion split to Espresso equipment test methodology....
Dan Kehn
User avatar
HB
 
Posts: 6756
Joined: Apr 29, 2005
Location: Cary, NC

Link to "Titan Grinder Project"by AndyS on Sun Jun 17, 2007 9:33 pm

altoCalgary wrote:Dan Ephraim of Modern Process Equipment Corporation in a PowerPoint Presentation for one of the Roasters Guild Retreats argued that particle size and uniformity are essential characteristics of excellent coffee.Ron


Not exactly. If you go through his presentation (page 37), he specifically echoes Petracco: "The Espresso Exception for Quality Grinding: The Benefit of Plurimodal Grinding."
-AndyS
AndyS
 
Posts: 631
Joined: May 05, 2005
Location: NY

Link to "Titan Grinder Project"by another_jim on Sun Jun 17, 2007 9:52 pm

altoCalgary wrote:Teme,

I believe you are referring to M. Petracco in Percolation, Chapter 7, Espresso Coffee: The Science of Quality, 2nd. Ed. 2005, where he essentially states that multi-size grading provides the most suitable particle size distribution for brewing espresso. If this is true, I would be interested in learning what the best tasting multi-size distributions look like and what grinders are capable of creating such distributions.

On the other hand, Dan Ephraim of Modern Process Equipment Corporation in a PowerPoint Presentation for one of the Roasters Guild Retreats argued that particle size and uniformity are essential characteristics of excellent coffee.

http://www.mpechicago.com/coffee/TechInfo/PDF/Roasters%20Guild%20Retreat%20Presentation.pdf

Which of these two perspectives is correct?


If we can get enough particle sizer time, I'm going to try testing a very simple model:

1. The rate of flow is determined overwhelmingly by the fines -- the more fines, the slower the flow.
2. The extraction of coffee is determined by how fine the average coarse particles are.
3. The quality of grind is determined how narrow the dispersion of the coarse particles is -- the tighter the distribution, the better the taste.

The conclusion from 3 is obvious. The conclusion from 1 & 2 is more subtle. The finer one grinds, the greater the proportion of fines (more breaking of the particles). So a grinder that tends to produce more fines will have a coarser grind for the same shot speed when compared to one which produces less grinds. Therefore, a grinder that produces a higher p[roportion of fines needs to be dosed lower to get the same solids extraction and taste balance.

The model predicts that two grinders that produce equally tight coarse distributions will have the same taste no matter how many fines they produce, but that one will need to adjust the dose to compensate for the proportion of fines.
User avatar
another_jim
 
Posts: 1957
Joined: May 05, 2005
Location: Chicago

Link to "Titan Grinder Project"by IMAWriter on Sun Jun 17, 2007 10:03 pm

cannonfodder wrote:OK, so I do care that I only had one blind cup test. Lets face it, even a blind squirrel will occasionally find a nut. So this morning I managed to scavenge enough beans from the Super Jolly to give the Kony/SJ another go.

I went about it a little different this time, the wife 1.0 interface was bypassed. I took the spring out of my portafilter so I could use two baskets already loaded with coffee. I took two identical baskets, put a small K on one and an S on the other, then covered them with a small piece of tape to hide the mark. I ground and dosed from the Kony into the basket marked K and did likewise for the Super Jolly. I kept the same 16 gram dose, leveled out the baskets and put them on a table.

This time I had my son bring me one of the two baskets. I dropped it into my spring-less portafilter, tamped, flushed and pulled a shot just as the previous test. Then I sampled the shot and placed the basket on the grinds tray of the grinder I believe it came from.

Again, it was not much of a contest. I pulled the tape off the basket and had picked the grinder correctly. The Kony shot was bright with more acidic and less body in the cup. I could tell the beans had aged another day as the cup was slightly different. The Kony was a little harsher today. The drink grabbed my tongue for lack of a better term. A harder attack on the palate (I believe that is a cup term I have heard in regard to spirits). Again the Super Jolly was mellower but the floral and fruit was more prominent than yesterday. Still lots of body but not as clear and defined as the Kony. This morning I preferred the shot pulled from the Super Jolly.

I will be out of town on business all next week and not returning home until Saturday night so any further testing will have to wait. So far I have been pulling a traditional double just to keep things uniform. I think I will try some ristrettos when I return, which is my normal drink, 1.5-1.75 ounces in 30 or so seconds. The drip cup was also intriguing. I will have to try some traditional cupping and a press pot or two when I return home.

Funny, I would think that if anything, the extra day would take a little edge OFF...it does for my blends that have a little Harrar in them...
Also (and please forgive..I'm just asking...lol)...did you pull a shot BEFORE you pulled your first test shot....I know my first shot from my Anita (admittedly not quite in the league of your Electra) is always just a touch off tempwise, no matter how diligently i try to perform my flush....perhaps the first shot onyour 2nd attempt was a touch hot?...hotter than the previous day's attempt?
Thanks for all you're doing...all of you.
User avatar
IMAWriter
 
Posts: 315
Joined: May 09, 2005
Location: Brentwood, TN

Link to "Titan Grinder Project"by altoCalgary on Mon Jun 18, 2007 11:36 am

another_jim wrote:If we can get enough particle sizer time, I'm going to try testing a very simple model:

Jim,

You might find information presented in this link adds to your particle-size investigation.

http://www.mpechicago.com/coffee/TechInfo/tchamburg/index.html

BTW, I have no relationship with MPE.

R
altoCalgary
 
Posts: 23
Joined: Aug 18, 2006
Location: Calgary

Link to "Titan Grinder Project"by altoCalgary on Mon Jun 18, 2007 12:23 pm

another_jim wrote:If we can get enough particle sizer time, I'm going to try testing a very simple model:

1. The rate of flow is determined overwhelmingly by the fines -- the more fines, the slower the flow.


Petracco [referenced in a previous post] offers some insight into the coarse and fine grounds interactions that occur during percolation and the influence of these two on extraction rate, offering support for Jim's first hypothesis.
An indication that some form of modification does occur inside the bed derives from electron microscopy of the bed structure..., which is made up of coarse particles containing fragments of cell walls, called fines. The latter are assumed to migrate to and eventually to concentrate at the bottom of the bed. Indirect confirmation comes from tests where the direction of percolation was reversed, using an extraction chamber that could be turned upside-down... Surprisingly, flow increases when the percolation chamber is inverted... This effect can be explained by assuming that the finest particles, which concentrate in a lower section of the cake, causing an increase in hydraulic resistance, now counter-migrate, with an increase in hydraulic conductivity in the opposite direction, until the system reaches a new steady state.
altoCalgary
 
Posts: 23
Joined: Aug 18, 2006
Location: Calgary

Link to "Titan Grinder Project"by cannonfodder on Mon Jun 18, 2007 2:16 pm

IMAWriter wrote:Funny, I would think that if anything, the extra day would take a little edge OFF...it does for my blends that have a little Harrar in them...
Also (and please forgive..I'm just asking...lol)...did you pull a shot BEFORE you pulled your first test shot....I know my first shot from my Anita (admittedly not quite in the league of your Electra) is always just a touch off tempwise, no matter how diligently i try to perform my flush....perhaps the first shot onyour 2nd attempt was a touch hot?...hotter than the previous day's attempt?
Thanks for all you're doing...all of you.


I have not noticed much difference between first and second shot in regard to temperature on the Elektra. I do quite a bit of flushing to prepare the machine before the first shot. One of the benefits of a plumbed machine, some would call it wasteful use of water, but I have no problem flushing 3-4 oz of water a few times as I prepare things in the morning. I do a long cooling flush first, then gather my things, do another short flush, clean the chute on which ever grinder I am using, do another short flush, grind dose tamp, flush and pull the shot. Yes it is a lot of flushing, but that gets the group to operating temperature essentially simulating 3 shots before I actually pull the first.

The taste difference could have been a byproduct of out-gassing, different blends age differently.

FWIW, I did another basket shuffle this morning with some fresh Rocket espresso and picked the correct match again.
Dave Stephens
User avatar
cannonfodder
 
Posts: 3658
Joined: May 23, 2005
Location: Dayton, Oh

Link to "Titan Grinder Project"by altoCalgary on Mon Jun 18, 2007 6:21 pm

another_jim wrote: The model predicts that two grinders that produce equally tight coarse distributions will have the same taste no matter how many fines they produce, but that one will need to adjust the dose to compensate for the proportion of fines.

I am looking forward to Jim's experiments. In a spirit of friendliness, I would like to offer an opposing point of view. I would like to suggest that Daniel Ephraim's quote from his presentation at the Tea & Coffee World Cup 2005 in Hamburg, Germany below implies to me that differing quantities of fines of two grinders that produce equally tight coarse distributions may have different tastes because of their varying exposed extraction surfaces.

Sometimes, it's advantageous to produce a plurimodal particle size distribution, or grind, where the finer particles enhance the exposed extraction surface (chemical need) and the coarser particles allow for the water flow (physical need). One example of this is in espresso coffee, where a concentration of 20 micron-type particles is essential to producing the ideal cup of espresso.
altoCalgary
 
Posts: 23
Joined: Aug 18, 2006
Location: Calgary

Link to "Titan Grinder Project"by RapidCoffee on Wed Jun 20, 2007 1:13 am

Titan Grinder Impressions

After almost three weeks with the two of the Titans, the Macap MXK and the Mazzer Robur, perhaps it's time to share some thoughts on grinder construction and ergonomics. Most of my first impressions, posted early in the TGP thread, have held up quite well. The baseline for my comparison is the flat burr Mazzer Super Jolly, an excellent grinder in its own right.

Image Image
Macap MXK on the left, Mazzer Robur on the right

The Macap MXK and the Mazzer Robur are impressive grinders that boast large conical burrs (63mm for the MXK and 73mm for the Robur). Both models feature stepless grind adjustment, practically a requirement for great espresso. The MXK grind is adjusted via a precise, easy-to-operate worm drive. Large grind adjustments, however, are tedious. The Robur grind is adjusted by turning a dial against fairly stiff resistance. This is hard to do one-handed, but fine adjustments are still reasonably precise, and large adjustments are much quicker.

Both grinders come with large (>1kg) bean hoppers that betray their commercial design. These hoppers are of little use in the home environment, and only encourage bean staling. Most home users will either dose directly into the grinder throat or rig up a mini hopper replacement.

Likewise, both grinders come with large dosers. Dosers provide the ability to pregrind many shots ahead of time, a dubious benefit that is largely irrelevant at home. Used properly (thwack-thwack-thwack!), dosers can help break up clumps in the grinds. Many people prefer dosers but...

<rant>
Confession time: I really dislike dosers. I removed my Super Jolly doser two years ago, and never felt the slightest inclination to reattach it. The past three weeks have only reinforced my bias. In general, I find dosers to be noisy, inconvenient, and major contributors to stale grinds retention. Now that that's out of my system...
</rant>

The MXK dosing lever pulls much more easily than the Robur. At first this felt good, but as the dosing levers loosened up with use, I found myself preferring the stiffer pull of the Robur. The Robur doser is built more robustly than the MXK, with more metal construction.

Image Image
Dosers with finger guards intact: Macap MXK on left, Mazzer Robur on right

Both dosers sweep cleanly, but an uncomfortably large volume of grinds are retained in the exit chutes from the grinding chambers, especially the wide Macap chute. I removed the large metal coffee outlet guard (finger guard) on the Robur to allow grinds to be brushed out. Both grinders have autogrinding cutoff switches that block access to the chute even further. The plastic autogrinding switch on the MXK snapped off during a routine brushing out of stale grinds (shipping damage?); even so, the attachment block made it difficult to wiggle a brush into the chute. The Robur retains considerably less grinds in the chute, which is good, because access to the chute was even more effectively blocked by the attachment block and the small metal autogrinding cutoff switch. More grinds are retained in the conical burr grinding chambers of the MXK and Robur than the flat burr grinding chamber of my Super Jolly.

Image Image
Dosers with finger guards removed: Macap MXK on left, Mazzer Robur on right

The MXK is operated by an on-off switch, covered by a flexible plastic membrane to keep grinds out of the switch. I disliked the feel of this switch, but at least it worked from the get-go! The Robur is operated by a dial that rotates from 0 (off) to 1 (automatic grinding) to Start (manual grinding). However, a locking microswitch prevents the grinder motor from running unless the enormous bean hopper is in place. This safety feature can be circumvented by inserting a similarly shaped piece of plastic (such as a cut-up credit card) in place of the hopper tab. (C'mon Mazzer, was this really necessary? :roll:) Of course I cannot recommend that a safety feature be disabled... but it is possible to operate the grinder manually with a reasonably sized hopper.

Both grinders operate quietly and smoothly. The conical burrs are easy to dial in, and the grinds produce visually appealing pours.

Image Image
Naked extractions with Rocket Classic Espresso: Macap MXK on left, Mazzer Robur on right

The MXK grinds quite slowly for a large commercial grinder, more than twice as slowly as the Robur or Super Jolly. Of the three grinders, the MXK taste profile was my least favorite. MXK shots are characterized by clear definition of individual flavors and a thin body, highlighting the bitter end of the taste spectrum.

The Robur grinds rapidly, slightly faster than the Super Jolly. The Robur produces a balanced taste profile, characterized by richness and body, and a flavor profile somewhere between the transparency of the MXK and the muted blending of flavors in the Super Jolly.

I've been greatly impressed by the Robur from the very beginning, and it rapidly became my grinder of choice. Do I want one? Hell yes! Do I want one in my kitchen? Hell no! It's far too big, heavy, and costly to be a strong contender in the home barista market. That's too bad, for this is truly a wonderful espresso grinder. :)
John
User avatar
RapidCoffee
 
Posts: 1036
Joined: Dec 11, 2005
Location: Rapid City, SD

Link to "Titan Grinder Project"by RapidCoffee on Wed Jun 20, 2007 1:45 am

Cupping the Titans

I'm embarrassed to admit that my early attempts to cup these grinders failed rather miserably. Although I was convinced that I could taste differences between them, I was unable to consistently quantify this in taste tests. Why? Well, in my defense, it's really hard to cup espresso. With one espresso machine, I can only pull one shot at a time. The shot appearance and taste change dramatically over a period of a minute or two, making it well-nigh impossible to cup espresso in the traditional sense.

In desperation, I tried a more traditional cupping approach. I marked the bottom of three Bodum Pavina glasses with different colored dots for the MXK, Robur and SJ grinds. I brewed 7g of coffee (espresso grind) in 5oz of 200F water. Due to the fine grind, I used a 1 minute infusion, filtered through a Swissgold filter before tasting. The conicals both yielded a brighter, more interesting cup than my flat burr Super Jolly. In blind taste tests, however, I typically picked only one grinder correctly, and mixed up the other two. Conclusion: do not quit your day job to become a coffee cupper! :oops:

Image
More traditional cupping of Robur, Super Jolly, and MXK grinds

Dave (cannonfodder) has suggested a basket swap protocol: prepare pucks from two grinders in two identical filter baskets, and ask an accomplice to select one at random. Pull the shot and guess which grinder was used. I tried the random basket swap this morning with encouraging results. After preparing baskets from the Robur and MXK, I had no trouble identifying the first shot as the Robur grinds, from its rich flavor and body. The second shot was equally easy to identify as originating from the MXK grinder. I was also able to distinguish between the MXK and Super Jolly grinds.

However, this should not be taken as definitive evidence. As the only home barista in the house, I have to prepare and pull the shots. The visual appearance of the pours is often enough to bias me, despite use of a spouted portafilter:

ImageImage
Can you distinguish between the MXK and Super Jolly shots?

In the future, I'll prepare the baskets, but ask my wife to pull the shots. (She calls them, anyway. :lol: )

As the owner of an HX machine, brew temperature is another issue that becomes hard to control. I'm eagerly awaiting delivery of Eric Svendson's E61 thermometer adaptor kit, which should help with brew temperature reproducibiity issues.

I believe that the CBS (Cannonfodder Basket Swap) protocol is likely to yield meaningful cupping results, especially if tracked over time. But I'm certainly open to other suggestions. :idea:
John
User avatar
RapidCoffee
 
Posts: 1036
Joined: Dec 11, 2005
Location: Rapid City, SD

Link to "Titan Grinder Project"by RapidCoffee on Wed Jun 20, 2007 2:34 am

Miscellaneous Observations and Speculations

In an attempt to correlate taste with quantifiable measurements, I offer the following:

Espresso Shot Density
I pulled a 17g 50ml shot on each grinder, then weighed the shot and measured its volume after the crema subsided. All three samples yielded a shot density of about 1.1g/ml, roughly 10% higher than water. The initial volume (with crema) was roughly 1.8-2.0X the final volume (after the crema subsided). Little difference was observed between the grinders, but the volume measurements were not very precise. Has anyone else measured espresso shot density? Perhaps a Brix meter or TDS measurement would yield an interesting comparison as well.

Brew Ratios
My brew ratios tend to fall squarely in the normale double range, even when I deliberately downdose. I dosed by volume into my "14g" ridgeless double basket, weighed the dose, pulled the shot, weighed the shot, and computed the brew ratio. Shot volume is listed as well.

Image

Note that 14g is a misnomer. These are 18g baskets. The 15g MXK dose fell well below the rim of the basket.

Grinds Packing
When dosing by volume, the conical burr grinders tend to dose heavier than my flat burr Super Jolly. After preparing the pucks, I weighed the dry grounds as follows:

1) Grinders were adjusted to yield a normale double pour in 25-30 seconds when dosing by volume.
2) Using the same ridgeless "14g" basket throughout, whole beans were premeasured by volume to the rim of the basket (~20g of whole beans).
3) Grinds were dosed into basket with a WDT yogurt funnel. I did my best to brush all the grinds out of the dosers.
4) After WDT stirring, the yogurt funnel was removed, and the grinds were leveled to the basket rim with the dissecting needle handle.
5) Puck was tamped and weighed.

Puck weights (average of three measurements):
MXK: 18.8g
Robur: 18.5g
SJ: 17.4g

The flat burr grinder doses about 1 gram less. Why?

Speculations
In the SEM thread, I speculated that conical grinders, because of their longer grinding paths, produce more spherical particles. The flat burr Super Jolly grinder appeared to produce more irregularly-shaped grinds. This could explain the dosing differences. Irregularly-shaped particles should pack poorly, with more space in between; spherical particles should pack more tightly.

Following this line of fantasy: irregularly-shaped particles have a large surface area per unit volume. This should cause irregularly-shaped grinds to extract faster than spherical particles, which have the smallest possible ratio of surface area to volume. All other things being equal (grind size, flow rate, etc. etc.), conical burr grinds should underextract relative to flat burr grinds.

Perhaps these observations can partially account for the taste differences that Dave and I have noted for the conical grinders. First, regularly-shaped conical burr grinder particles pack better and tend to overdose the basket, compared to irregularly-shaped flat burr particles. Second, regularly-shaped conical burr grinds tend to underextract relative to the irregularly-shaped flat burr particles. Overdosing plus underextraction should yield a different cup.

Obviously these are not the only factors affecting taste. Particle size distribution undoubtedly plays a major role. The large fines peak in the MXK distribution may explain the bitter taste profile. The MXK distribution also features a wide spread of large particle sizes, perhaps contributing to the clarity and definition of individual flavors. The Super Jolly grinder, with the narrowest large particle peak, is characterized by the greatest blending/muting of flavors.

One final speculation: Jim has noted WDT-related taste changes that he associates with lowered solids extraction. Stirring the grinds should allow them to pack better, similar to shaking a box of marbles. So if you dose by volume, stirring may lead to updosing and lower extraction ratios. If you dose by weight, then stirring may create a more compressed puck that offers greater resistance to flow, which would lead to a coarser grind being selected, again favoring underextraction.

Pure speculation, but interesting grounds for thought... :wink:
John
User avatar
RapidCoffee
 
Posts: 1036
Joined: Dec 11, 2005
Location: Rapid City, SD

Link to "Titan Grinder Project"by RapidCoffee on Wed Jun 20, 2007 2:56 am

One final post before I pack it in for the night. When I got home today, I found a wooden box awaiting me from H-B's newest sponsor:
Image

Inside was a bag of PT's Bella Vita, a northern Italian style espresso blend, and wow! 12oz of Ethiopia Yergacheffe Biloya. :D
Image
Shirt logo: "Humans get a lot done, not because we're smart, but because we have thumbs so we can make coffee."

Many thanks to Jeff and the rest of the crew at PT's Coffee Roasting for their contribution to the TGP!
John
User avatar
RapidCoffee
 
Posts: 1036
Joined: Dec 11, 2005
Location: Rapid City, SD

Cimbali Max Grinder vs. the Cimbali Junior and Cadet Grinders; with Blind Tasting Comparisons

Link to "Titan Grinder Project"by Ken Fox on Wed Jun 20, 2007 3:05 am

Knowing that I owned a couple of new Cimbali Max grinders, plus had a couple of Cimbali planar burr grinders lying around, Dan Kehn asked me if I would do a comparison of these grinders for the Titan Grinder Project. This was a couple of weeks ago when I was up in Vancouver, Canada, on vacation, but we made arrangements with Larry of Rocket Coffee Roasters to send me a supply of his excellent coffee, for use in the comparison, which was performed last weekend. The first few paragraphs of this post deal with my own personal history with Cimbali grinders, and if you find the writing tedious, you can skip down a few paragraphs and pick up the story at that point.

Perhaps a little background would be useful before I get into the actual tasting procedure and results. My own home espresso adventures began around 15 years ago, when in short succession I owned several cheap steam and pump operated machines, none of which lasted very long. I was about to give up on home espresso when I was discussing this experience with a my favorite restauranteur in Vancouver, John Bishop, on a prior trip about 13 years ago. John suggested I buy a Cimbali Junior, and that if I bought it, the machine would last the rest of my life. On my way back home I stopped off at the now-long defunct, by appointment only, location of the Boise Cimbali dealer. The salesman there tried to talk me out of buying a Junior, saying it was overkill for a home (this was 1995); he insisted that I take a Domus and a small grinder home, and if I didn't want those after a few weeks, he'd take them back and order me a Junior. So I took the Domus home and a week later called the salesman and said I wanted the Junior and a matched grinder. Since I wanted the equipment in brass, I had to wait several months before it arrived. The brass Junior is the same (now-heavily modified) vibe pourover I still own today. Although I had ordered a Junior grinder, they sent a Cadet instead, which I got at the same price as I would have had to pay for the Junior since it was their error.

The Cadet grinder is basically a Junior grinder with autogrind and a larger bean hopper, the same grinder motor, the same 64mm planar grinder burrs, and the same grind path. In home use it is a real nuisance as the autogrind switch is less convenient to use for small volume, shot to shot, grinding than is the proper on-off switch one finds on the Junior. Because the Junior and the Cadet both grind rapidly, the autogrind and on-off rocker switch of the Cadet are a real liability, and results in a lot of coffee waste. In retrospect I wish I had refused the Cadet and told them to re-order the Junior I had originally requested, but nonetheless the Cadet served me well for about 8 years as my sole grinder.

After about 8 years, I ordered a Mazzer Mini which I kept for 3 months, for use as a 2nd grinder; I found the Mazzer doser to be annoying and ended up selling the almost new grinder to an alt.coffee friend. After selling the Mazzer, I bought the first of two Cimbali Junior grinders, and shortly thereafter the 2nd one, retiring the Cadet to the basement for use as a spare, which never happened. I did, however, replace the old burrs on the Cadet with new ones several months ago, although did not use the grinder afterwards other than to test it.

I had heard about the Cimbali Max grinder, which has a hybrid Conical/Planar burr set, but read some conflicting things on the internet and doubted whether the burrs were as described. Further discussions here and elsewhere convinced me that it was a real product with both sets of burrs, similar to the highly regarded DRM belt driven grinders that Schomer and others used to rave about. The outer planar burrs on the Max have an external diameter of 64mm (like the planar burrs on the Junior, although they are of a different design), whereas the DRM planar burrs measured 68mm. I contacted T.J. Tarateta of Ammirati Imports, a Cimbali importer located in NY, who participates in these forums, and he graciously offered to obtain a couple of Max grinders for me to purchase, even though the Max grinder is not a grinder that they normally import. I received these grinders in late March, about 10 days before a long trip to France that ended in Long Beach at the SCAA convention.

Since obtaining the Max grinders, I sold one of my Junior grinders to a friend but kept the other one, and also had the Cadet grinder, which had not been used after having its burrs changed for new ones 6 months ago. All of the above is to simply illustrate that I have a lot of experience with Cimbali 64mm planar grinders, including the Junior and the Cadet, and that this experience spans 12 years, plus now about 5 weeks experience using the new Max grinders I received in March.

The Max "suffers" from the same major flaw as the Cadet in home use, i.e. it is an autofill grinder that is designed to fill the doser with coffee after which the grinder will shut off. For home use this is obviously not suitable, so one is forced to use the rocker "on-off" power switch in lieu of a proper on-off switch in order to grind per shot. A related flaw is the "flapper" sensor in front of the grinder chute leading into the doser, which makes cleaning the chute a bit of a chore. Finally, like the Cadet, the bean hopper is larger than the bean hopper of the Junior, and unlike the Junior, the Max is a couple of inches too high to fit under most kitchen cabinets. I should temper these negative comments by saying that the Max grinds approximately 1/2 as fast as the Junior or the Cadet, and as a result it is possible to learn how to use the power switch effectively so as not to grind too much coffee and end up with a big coffee waste problem. I never learned how to do this with the Cadet, as it simply grinds too fast for control of the power switch to work, but with the Max I quickly was able to grind per shot with no more bean wastage than with the Junior.

In actual use, the first thing I noticed with my new Max grinders was that it seemed that grind adjustments were easier, and needed less often. By this I mean, that changing in between different coffees, or adapting to changes in coffee as it aged, or dealing with weather and humidity changes -- all of which normally require grinder setting adjustments on the Junior and Cadet grinders, seemed to require less adjusting or in some cases no adjusting in order to use the Max to get good espresso shots with proper time/volume relationships. In addition, the "quality" of the grinds are different; from the Junior, they are like grains of sand, but from the Max, even with robust doser thwacking, one gets small clumps. These are not "bad clumps," however, and the basket dose is easily distributed and packed, with no added channeling evident on a bottomless portafilter. I also noticed a small and subjective improvement in shot quality with the Max, but of course that sort of conclusion is pretty easy to talk yourself into, especially as a means of justifying a recent (and expensive) purchase.

Shortly after returning home from my Vancouver trip, I received a Priority Mail package containing 5lbs of coffee, Rocket Coffee Roaster's Classic Espresso Blend to be precise.

Image

That is one nice blend and I do thank Larry for kindly sending it! This coffee made the following blind tasting a pleasure rather than a chore.

I decided I better get serious about this grinder comparison, and quick, before the coffee got too far along in its evolution; I did freeze about half of it for future testing (or drinking!) since I didn't think we could possibly use up 5lbs of it with the testing I had envisioned, at least at this point. So, I got the Cadet and Junior grinders out of the basement, and lined them up with one of the new Maxs for a "family portrait:"

Image

I had previously decided to use the Cadet grinder in the head to head comparisons with the Max grinder, because the Cadet, like the Max, had new grinder burrs. My remaining Junior grinder has 3 year old burrs, which although not anywhere near ready for replacement, might introduce an additional undesired variable into the testing. Given the identical planar burr set, grind path, and doser, I felt comfortable using the Cadet for this comparison rather than the Junior.

I enrolled a friend, Bob, also a home espresso enthusiast who owns an Andreja Premium (and now, my former Junior grinder as well!), as a fellow taster. He had also served admirably during the frozen coffee study previously published on this website. Bob had one requirement, which was that he did not want to become "buzzed" like he did after the two days of 8-shot pairs he experienced on the freezing trial. What I designed was a brief paired shot blind tasting trial utilizing my two Cimbali Junior espresso machines, in much the same way they were used in the freezing trial.

In brief, we had the same Rocket Coffee Roasters coffee in both the Cadet and the Max grinders, and for half of the test period each grinder was paired with one of the espresso machines, and for the other half, with the other machine, in a balanced design. There were a total of 10 paired, blind tasted shots, of which I tasted 6 pairs and Bob tasted 4 pairs. For the first 5 paired shots the older vibratory Pourover Junior was paired with the Cadet grinder, and the newer Rotary Junior espresso machine was paired with the Max, then for the last five the grinder and espresso machines paired were the opposite. Simultaneous shots with equivalent time/volume parameters were pulled on the two machines and then presented simultaneously in a blinded fashion to the taster as "right" and "left" cups. These shots were made with bottomless portafilters filled to be flat with the top of the PF basket with about 17g of coffee, tamped lightly, and then pulled as 1.25-1.5 oz double shots over about 25-30 seconds. They were poured directly into white Revol porcelain espresso cups located directly beneath the bottomless PFs.

The taster's "job" was to compare the two shots according to 4 parameters and to determine (1) the intensity of the aroma and taste; (2) the quality of the crema; (3) the mouth feel; and, (4) the overall preference. The identities of what was in each cup in each pair were determined by the "server" of the coffees with coin flips prior to testing; these were not revealed to the tasters until the entire tasting session had been completed.

Prior to starting the testing, the espresso machines were adjusted with their PID electronic temperature controls and a Scace Thermofilter, to produce equivalent shot temperatures during shot series mimicking the frequency of shot pulls that was performed during this blind tasting. The brew temperature was approximately 199 to 200F for this study. Extraction pressures were measured to be approximately 9 bar, with each machine having its own sort of preinfusion, either from the innate behavior of vibe pumps, or the preinfusion modification I have performed on my rotary machine with a solid state delay timer, previously described on this website in several postings and prior experiments.

After a couple of hours we had test results, and after looking at them casually it was my impression that the results were random and not indicating any preference between the grinders. I then scanned the tasting sheets and sent them off to Jim Schulman, our resident statistics expert, who "tortured the data until it confessed," yielding real results that I had not expected; the Max actually scored higher than the Cadet, in all four comparison categories given above.

I'm going to quote from a couple of paragraphs from Jim's analysis without going into any of the numerical underpinings, which I am going to let Jim explain for himself in a subsequent post (largely because I'd prefer not to make too big of a fool of myself, not having actually done any statistical calculations myself in the last two decades):

"The Max did a better job in all the categories . . . . .In essence, the Max spanked the Cadet when it was paired with the rotary, and did slightly worse when paired with the vibe. Bob liked the Max a bit more than Ken did. . . . . The outcome is consistent across all the taste categories. This consistency adds to the significance of the result, since if the result were just chance, it would be like flipping 4 heads in a row."

"My conclusion is that the Max grinds an immediately perceivable better shot on the rotary Cimbali than the Cadet or Junior grinder. On the Vibe machine, the edge disappears completely. Whether this conclusion about the Max applies to all rotary and vibe machines is unknown. However, the results warn that one needs to check for an interaction between conical grind quality and the pump being used."


Personally speaking, I find these results interesting however from my perspective, after using the Max daily for more than a month, I think the most persuasive reason to buy one is the increased certainty of getting a good shot without excessive fiddling with grinder settings. Although I had made the observation about the setting being less critical on the Max before I set up this tasting trial, the proof of that observation came as I used the grinders in this testing. Before my friend Bob came over for the tasting, I dialed the grinders in with the Rocket Coffee Roasters Classic Espresso blend. It took SIX double shots, 5 of which were sink shots, to get the Cadet dialed in. With the Max, there was one sink shot, one mediocre shot, and beautiful shot for the 3rd. Midway through the actual tasting when we switched the grinder machine pairs, I had to toss one set of shots because the Cadet needed to be adjusted to go from the vibe machine to the rotary. Not only that, but I needed to make small adjustments on the Cadet a couple of other times during the 10 shot trial. I never adjusted the Max, not even once, from start to finish of the testing including the switchover from pairing it with the rotary to the vibe.

T.J. Tarateta has posted here recently that they will be bringing in a modified version of the Max for the N. American market, having the attributes of the Max with the form factor of the Junior. I believe that this modified Max will lack autogrind and have a smaller, Junior sized hopper. It will therefore be very "home kitchen friendly," fitting under counters and with the appearance of a Junior which will look good in most kitchens. You can't say that about the huge pure Conical grinders being tested in this grinder project, and this is a point very important to also consider.

ken
What, me worry?

Alfred E. Neuman, 1955
Ken Fox
 
Posts: 1057
Joined: Oct 28, 2005
Location: Idaho

PreviousNext

Return to The Bench