The following is quite a long post about how my Mazzer Super Jolly and Elektra Nino grinders both produced inferior results when used for single dosing.

This post kind of follows on from Is anybody single dosing with a Robur? and some of the comments towards the end of Elektra Nino Grinder as well as A Tamper is Too Heavy for Grinder Anti-Popcorn Remedy.

All of these threads take the virtues of big heavy grinders for granted and involve people who like to drink more than one type of bean/blend during the course of a day and/or want to avoid as much as possible any contamination of their fresh-ground beans by stale grinds that remain in the burr chamber and chute.

Single-dosing for the former is attractive for the obvious reason that you can use one grinder for everything without changing hoppers and purging beans, so it's less fiddly and of course it also takes up less kitchen space and leaves a much smaller dent in the wallet than having several grinders.

For the latter, it's attractive because the only grind retention you need worry about if you have a model that allows you to clean out the chute, are the grinds in the dosing chamber.

By single-dosing I refer to the practice of weighing out your single dose and then placing those beans alone in the grinder throat and grinding until nothing more comes out.

Having got positive feedback from HBers single-dosing with a Robur (big 71mm conical burrs with a massive motor) I thought I'd try it on my Elektra Nino (big 68mm conical burrs with a slightly smaller massive motor) and take my Super Jolly (64mm flat burrs, substantially smaller motor) along for the ride.

For the comparison, I was using beans that require very similar doses and grinds, using a double basket of around 14g in the Elektra (spring lever) and Cremina (manual lever).

At first, the results were encouraging. The Super Jolly took just a couple of dial marker points finer to dial in the grind, obtaining visibly identical extraction and pour. The Nino took a bit more work partly because it uses a grub screw adjuster and partly because it required around 1/8 of a revolution of the burr carrier to dial in for single-dose operation. I was using the technique Jon recommended in the Robur thread, closing off the top of the burr chamber with the hopper flap to stop the beans popcorning all over the place. On the SJ, I just popped a 58mm tamper in the burr chamber throat. This rests on four ledges and does not apply any pressure to the beans.

The experiment, which lasted for the best part of last week, started off using an Ethiopian Wild Bean from the Bale region (roasted nice and light by Londinium Espresso), which is nevertheless very consistent in taste and mouthfeel, unlike some other wild beans.

I was disappointed to note immediate negative differences. Now I want to stress this was not blind testing, but the differences in the cup or rather in my mouth were so obvious that I really don't feel this invalidates my impressions. of course, your mileage may vary

In some ways the result with the Super Jolly was an improvement. The Bale Wild Bean is very smooth and creamy, with some predominantly floral brightness and earthy, almost mushroomy undertones. On the Super Jolly these nuances are less pronounced than on the Nino, and single dosing brought them out slightly more, with the disadvantage that the brightness which had been absent before nevertheless tasted more like an underextracted sourness than a champagne zing.

On the Nino, the taste profile was as distinct as ever but somehow more broken up, less integrated, and the mouthfeel was less luscious than when using the grinder as designed. Although much sweeter than the SJ, there was still some loss of winey zing and a little more acid than usual as well.

Next up (after a few days) was a Guatemalan - Finca la Perla - a very nice, sweet, nutty and chocolatey coffee. I was expecting this to fare quite well in the single-dosing Super Jolly. It proved much fussier than I had imagined - not in terms of getting a decent extraction but in getting that floating veil of sweetness in place over the top of the other tastes both in the "cup" and in the aftertaste. It simply didn't happen on the Super Jolly. Instead there was a hint of acidity, less chocolatey chocolates, a bit of earthy heaviness in the background. It was an altogether flatter experience and the mouthfeel was lighter than usual.

I should emphasise I was only comparing what was happening with my own recollections at this stage and across grinders. Minimal adjustments to both grinders were required between beans and in each case 250g were used. I only went back to hopper user towards the end of each bag of beans.

The Nino fared much better with the Finca LP. The delicate sweetness came through, the earthiness changed into rich chocolate and the cup was smoother. Nevertheless the experience was not as I remembered when running the Nino normally.

On the Nino, I started to experiment at this point with removing the hopper and single-dosing with a tamper resting on a ledge within the dosing chamber. This reduces the space for popcorning without however exerting any pressure on the beans and did not appear to make absolutely any difference to the results in the cup.

Switching back to using both grinders as they were designed - with beans in the hoppers - yielded a massive improvement on both grinders, but on the Nino in particular - where I had least expected it since the results with the Finca LP were acceptable. The improvement in body of the Finca LP, the shot-silk effect of the delicate sweetness both in the cup and aftertaste was staggering, and this, the chocolates and nuts were not only more distinct but equally more integrated. It was very clear that in single-dose mode I'd been missing much of the experience. The same was true of the Super Jolly. The sweetness in this bean only came through with a weight of beans in the hopper, and the integration and body were also improved.

The final bean used was a wonderful deep chocolate and earth Java from Londinium. This came through as rather heavy and ponderous on the Super Jolly, with a vague loss of body and the addition of some slight sour notes when single dosed. It was fuller, richer and sweeter on the Nino when single-dosed, and much creamier, even sweeter and more chocolately when feed with a column of beans in the hopper.

My simple conclusions are that maybe flat burr grinders are better suited to singe-dose use than big conicals but that neither grinder really showed everything they are capable of when used in this way. Overall I would say the improvements were, however, more subtle on the Super Jolly.... except that....well, that amazing delicious sweet vein that shoots through the Finca FP simply didn't show up in single-dose mode, whereas it magically appeared with beans in the hopper.

With the same beans the Nino went from slightly fuller, sweeter and more detailed than the Super Jolly (single-dosing) into a completely different class of distinct yet integrated flavour and smooth, sweet body. Something good became very much better.

Will these findings apply to other grinders? Frankly I haven't the vaguest idea. I have a suspicion that it may apply to other big conicals. I'd certainly be interested to hear from Robur users who have done a similar singe-dose/hopper comparison. Having now analysed the difference between the SJ and Nino, I don't want another flat burr grinder as a "second" grinder, especially with the Major described as a SJ on steroids but without a vastly different taste profile. On some beans, like the Wild Bean, the Nino is significantly better, but on others, such as the Blawa Java, Finca LP or Londinium's amazing Sidamo, it's in a world apart.

As to why there are these differences in taste profile and mouthfeel, my theory - looking at the grind - is that you get silvery scaly chaff-like grinds that make their way through the burrs at beginning and end of each dose, when the popcorning occurs, and that these underextract. The main bulk of the beans that come through once the burrs have been filled are a very tight grind. They provide the fines required for the shot to pour well, but they overextract. So although you get a shot that pours OK, it is composed of particles that do not extract evenly, some overextract and some underextract, and very few are exactly where you might want them or get them grinding "normally" - resulting in a shot that has less body, more extremes and less integration of tastes. Interestingly, although you can usually tell the grind of the SJ and Nino apart from their appearance, the single-dose grinds were almost indistinguishable in the PF.

The above theorising fits with what I tasted, but I emphasise that I went into this without prejudice, indeed very much hoping that never mind single dosing on a Robur I might just get away with doing the same on my Nino and not having to buy anything new at all.

Lastly, one curiosity on the Nino was a black hole into which around 0.3g of coffee would disappear. I was regularly weighing the beans both before and after grinding. This 0.3 or so must adhere to the burr chamber sweepers. I can't think what else is happening. You grind your shot and it's that 0.2-0.3 under. You can then spin the grinder until you are blue in the face, but nothing further will come out. Your next shot is then likely to be 0.2-0.3 over and it comes out then all in one go. Although the grinds wouldn't always disappear into the black hole, they would always come back out for the following shot. On doubles the amount lost in the black hole wouldn't vary but the phenomenon did occur less regularly.

Lastly, to get the best out of single dosing on the Nino if one really wants to you should of course set a high grinding time - longer than you need, rather than to keep pressing the button and grinding several times, since the popcorning will then be repeated for each timed interval, rather than principally at the beginning and end of the shot. At least that was my thinking. I'm not convinced it made that much practical difference. I suspect the beans were just bouncing around in there all the time.

Others' thoughts and experiences?

Cheers

Mike

Yes, certainly less than ideally neutral conditions, Jim.

I did get up to tricks like reversing the order of which one came first and rinsing the PF under a cold tap and so on to ensure as much possible comparability, including for temperature in the cup. But these comparisons affected my comparative impressions of the grinders, not of the type of grinding, which is more debatable since it went from memory over the space of a day or two and then, perhaps more significantly, the comparison of the last two or so shots either side of the "grinding mode changeover".

It did encourage me that once it became clear I couldn't tell the baskets apart in terms of whether it was a SJ or Nino grind, the differences between them were so noticeable. I put little pieces of paper under the baskets and then twiddled them round till I lost track so that element was almost bind.

The temperature stability of the MCal set to 0.8 -1.00 bar isn't that shabby (comparatively) and the new Cremina set to 0.7 - 0.8 is even better. Not only did I get similar results irrespective of order, but also across the machines, taking their own differences in taste separation and body into account. The matter of which basket came from which grinder was blind, to an extent, as explained above - and I would try not to look at the cone on the Elektra.... which would immediately give the grinder away. On the Cremina I had less choice (about not looking).

None of this is conclusive, but I have absolutely no doubt about what I was tasting - and that I would have been delighted to rationalise myself into a reverse impression, namely that I could merrily single-dose my way to paradise, drinking several beans in the course of a day all ground in the same grinder without "waste". I suppose what I am saying is that to an extent the inconvenience of the finding feels like a validating factor to me, but's that's not really any more scientific.

I can take a clean grinder, run 16 grams through it, sweeping all of the grinds into the filter basket and get a poor tasting shot. Take the same grinder and coffee, run say 20 grams through with no sweeping of the grinds into the basket and get a superior shot. Maybe coffee's contact with old coffee oil or metal as opposed to the coffee just passing over fresh coffee makes a difference. Can't say I've tested this enough to make an assertion, but it might make an interesting study.

Very interesting test. I have only one flat burr grinder - a Macap M4 stepless doser. I have been grinding single dose for convenience (and for an Elektra Semiautomatica single basket/espresso); but after your test I may do some of my own experiments and "fill the hopper" and compare...taste.

Howard

Interesting point, Doug.

Neither grinder was cleaned with Grindz or anything before I did my little week-long experiment and I carried on using the same coffee.

In terms of coffee in the grinds path (or not) the Nino was operating more or less as normal, perhaps with very slightly less coffee in the burr chamber (which holds very little) after each single-dose shot.

The cleaning of the Super Jolly chute and doser was exactly the same whether or not I was single dosing and just involved brushing out not vacuuming or anything more thorough, so those variables were were comparable.

Cheers

Mike

Great post, thanks for sharing your findings. I'm not surprised at the results, which corroborate my own experience. This includes the flat vs. conical burr results: conicals do seem to popcorn more.

As I've stated in other threads, I'd much prefer to use my grinders in single dose mode. But after running a few tests during the TGP, I began to shy away from single dosing, at least for espresso. This was partly related to taste, and partly to logic. Given the gross changes in pour times when single dosing, and the fact that commercial grinders are designed to be run with a large bean mass in the hopper, the burden of proof is on those who want to use these grinders in single dose mode, not vice versa. Sadly, I've not been able to convince myself that single dosing works as well. My big Mazzer grinders seem to function better with at least a small load of beans (several doses worth) in the hopper.

(Response from a question asked in the Elektra Nino thread)
Elektra Nino Grinder

I have not done any comparison testing for single dose vs. loaded hopper on my Robur. I have a K30 Vario sitting next to the Robur and that grinder I (obviously) run with beans in the hopper. I love the convenience of the K30 but I hate the waste of having to purge grounds every session

Your posts about this subject are interesting, but I would prefer to see some blind (or double blind) testing using a pair of equivalent grinders loaded both ways so the shots could be tasted together.

Without blind A/B tasting it's difficult for me to buy into any assertions made so far about the subtle differences in taste. Call me skeptical But I am intrigued so I am considering giving loaded-doser another go...one of these weeks in the not-too-distant future I'll have to roast a few extra pounds of espresso beans so I can do some comparitive testing for myself (maybe even get together with someone who has another large conical to use for side-by-side shots).


But I've been getting fairly stellar results lately, both from my home roasts and from the pro roasts I've had recently (it's been a bit too cold outside for me to roast on a couple of weekends this season). I'm not boasting, just indicating why I'm not in any hurry to run some pretty involved testing centered around altering a usage pattern that has been working very well for me.

Thanks for the feedback, Jon. It makes perfect sense to me that one wouldn't want to spent too much time futzing around when you're already getting great results.

And of course it could be that for some reason the Robur is more tolerant to single dosing than the Nino.

That said the results were far from subtle for me - it was like trying a whole new grinder. I think I took my eye of the ballas far as taste and texture were concerned when I obsessing over the whole business of dialling the grinders in for single-dose use. I didn't expect to have to go on a whole odyssey with the little worm gear knob (I much prefer the Mazzers in this respect) and so concentrated on pour and extraction almost to the exclusion of all else. The contrasts between the Nino and SJ also influence in distracting me from how the Nino was performing in the cup compared to itself in "normal" operation.

The first time I shifted back was a jaw-dropping moment....

The fact that there were similar albeit less dramatic improvements with the SJ led me to believe this may not just be a Nino phenomenon.

Nevertheless I completely agree with you and Jim that the ideal thing would be to run two grinders side by side and do some proper blind testing.

Thanks for all your input on my extravagant second grinder ramblings.

Cheers

Mike

michaelbenis wrote:The first time I shifted back was a jaw-dropping moment....

Sorry for my skepticism, but is this because of inherent grinder attributes or could it possibly be influenced by your preconceptions or perhaps even by your facility (practice or lack of practice) using a method you have clearly preferred in the past against something new and unfamiliar?

If I was to attempt such a test by myself, I feel there is a strong possibility that I might have a difficult time adjusting my style and technique to having a hopper full (in a short amount of time to practice) and so I could imagine seeing a similar result in the opposite direction from your "jaw dropping moment".


Now I am thinking a little more seriously about working out a side-by-side double blind test.