Sixteen shots made it across our tongues; the rest went straight into the sink. Then the long awaited Cimbali Junior Vibe versus Cimbali Junior Rotary test was over, and the two proud machines on Ken's counter got a rest. Two nearly identical one group commercial machines, differing primarily in the type of pump they contained, allowed direct comparison of espressos prepared with either a vibratory or a rotary pump.
We compared results. Of the four paired blind-tasted shots I had prepared, and Ken had tasted, two pairs had tied, the rotary won one, and the vibe won the other. Of the four paired blind shots Ken had prepared for me; one pair was tied and virtually indistinguishable, on another pair the vibe shot was better in all respects. On two pairs, I got better body and crema on one shot, better taste on the other; but lest there be any hint of system, the vibe and rotary switched positions. Ken didn't rate his shots; I rated all eight of mine between 85 and 90. We discussed our ratings and identified the shots after we had finished all the tasting. We both agreed that the difference between the machines was too small to distinguish from the shot to shot variability of either one. There may still be a difference in long term consistency; but there are no big differences in taste.
These are not entirely idle results, since we spent the morning adjusting the vibe machine, a pourover, dating from late 1995 to duplicate the basic characteristics of the rotary, one year old and plumbed in, as closely as possible.
The boiler size and configuration, as well as the heat exchanger and group are identical on both machines.
The rotary showed 9.25 bar on the pressure guage pf. After getting the OPV (overpressure valve) on the vibe up to speed with a hose back to the tank, and a little spring loosening, we were able to adjust it so it oscillated around 9 bar +/- .125 or so. Originally, it had oscillated wildly between 12 and 14 bar, with the OPV not bypassing any water. The more water the OPV bypassed, the smoother the pressure reading became.
The 250 mL water debit from the two machines: rotary -- 23.5 seconds, vibe -- 26.2 seconds.
The temperature differences between the two machines with the guages showing the same boiler pressure were large, around 95C for the vibe, 92.5C for the rotary. After adjusting the pstat of the vibe machine down from 1.1 to about 0.95, and adjusting the water flush, we got the first-shot-after-idle on the vibe down to around 92 - 92.5 as well. I rated shots on sour/bitter balance, and the results were random between machines, so I'm guesing the temperatures were roughly right.
The water used was identical, softened water, from a cation exchange softener and charcoal filter. The coffee used was Supreme Bean Abruzzo blend that had been roasted 3 days prior, and was ground in the same Cimbali Junior grinder to the same grind setting.
All in all, these results support the thesis that once a vibe pump is adjusted to operate at the same pressure as a rotary, the taste differences due to 60 Hz pressure wave are either non-existent or very small.
Ken's prior test comparing these two machines showed a preference for the product of the rotary over the vibe machine. Today's test varied in the more rigorous attempt to control variables such as temperature (regardless of front panel pressure gauge reading) and the pump pressures. Once these variables were equalized there was no reliable difference between the output of the two machines.
In prior postings Ken has advocated the use of rotary machines over vibe machines, and suggested the utility of converting old commercial vibe machines over to rotary pumps. After the completion of this "experiment," (not to mention some recent repair hassles), he no longer feels that such modifications are worth doing, at least if results in the cup are the intended purpose. Given some unpleasant experiences with modern electronically controlled commercial machines, his advice to alties seeking the ultimate one group commercial machine for use in the home, would be to seek out a very well maintained older commercial machine in good condition, with manual (electric rather than electronic) controls, with sightglass and no autofill. This would be preferable in most respects to any of the examples of the current crop of electronically controlled machines; the results in the cup are equal and the hassles in home use are far fewer with machines of older vintage.
The only real advantage of a rotary pump over vibe is in a multi group situation. There are of course exceptions, especially in non plumbed units. A rotary pump can draw it's own water from a bottle allowing even autofill functions to work.
"Ken Fox" <morceaudemerdeSNIPt...@hotmail.com> wrote in message
> Sixteen shots made it across our tongues; the rest went straight into the > sink. Then the long awaited Cimbali Junior Vibe versus Cimbali Junior Rotary > test was over, and the two proud machines on Ken's counter got a rest. Two > nearly identical one group commercial machines, differing primarily in the > type of pump they contained, allowed direct comparison of espressos prepared > with either a vibratory or a rotary pump.
> We compared results. Of the four paired blind-tasted shots I had prepared, > and Ken had tasted, two pairs had tied, the rotary won one, and the vibe won > the other. Of the four paired blind shots Ken had prepared for me; one pair > was tied and virtually indistinguishable, on another pair the vibe shot was > better in all respects. On two pairs, I got better body and crema on one > shot, better taste on the other; but lest there be any hint of system, the > vibe and rotary switched positions. Ken didn't rate his shots; I rated all > eight of mine between 85 and 90. We discussed our ratings and identified the > shots after we had finished all the tasting. We both agreed that the > difference between the machines was too small to distinguish from the shot > to shot variability of either one. There may still be a difference in long > term consistency; but there are no big differences in taste.
> These are not entirely idle results, since we spent the morning adjusting > the vibe machine, a pourover, dating from late 1995 to duplicate the basic > characteristics of the rotary, one year old and plumbed in, as closely as > possible.
> The boiler size and configuration, as well as the heat exchanger and group > are identical on both machines.
> The rotary showed 9.25 bar on the pressure guage pf. After getting the OPV > (overpressure valve) on the vibe up to speed with a hose back to the tank, > and a little spring loosening, we were able to adjust it so it oscillated > around 9 bar +/- .125 or so. Originally, it had oscillated wildly between 12 > and 14 bar, with the OPV not bypassing any water. The more water the OPV > bypassed, the smoother the pressure reading became.
> The 250 mL water debit from the two machines: rotary -- 23.5 seconds, > vibe -- 26.2 seconds.
> The temperature differences between the two machines with the guages showing > the same boiler pressure were large, around 95C for the vibe, 92.5C for the > rotary. After adjusting the pstat of the vibe machine down from 1.1 to about > 0.95, and adjusting the water flush, we got the first-shot-after-idle on the > vibe down to around 92 - 92.5 as well. I rated shots on sour/bitter balance, > and the results were random between machines, so I'm guesing the > temperatures were roughly right.
> The water used was identical, softened water, from a cation exchange > softener and charcoal filter. The coffee used was Supreme Bean Abruzzo > blend that had been roasted 3 days prior, and was ground in the same Cimbali > Junior grinder to the same grind setting.
> All in all, these results support the thesis that once a vibe pump is > adjusted to operate at the same pressure as a rotary, the taste differences > due to 60 Hz pressure wave are either non-existent or very small.
> Ken's prior test comparing these two machines showed a preference for the > product of the rotary over the vibe machine. Today's test varied in the > more rigorous attempt to control variables such as temperature (regardless > of front panel pressure gauge reading) and the pump pressures. Once these > variables were equalized there was no reliable difference between the output > of the two machines.
> In prior postings Ken has advocated the use of rotary machines over vibe > machines, and suggested the utility of converting old commercial vibe > machines over to rotary pumps. After the completion of this "experiment," > (not to mention some recent repair hassles), he no longer feels that such > modifications are worth doing, at least if results in the cup are the > intended purpose. Given some unpleasant experiences with modern > electronically controlled commercial machines, his advice to alties seeking > the ultimate one group commercial machine for use in the home, would be to > seek out a very well maintained older commercial machine in good condition, > with manual (electric rather than electronic) controls, with sightglass and > no autofill. This would be preferable in most respects to any of the > examples of the current crop of electronically controlled machines; the > results in the cup are equal and the hassles in home use are far fewer with > machines of older vintage.
In article <c4aiin$2falt...@ID-192261.news.uni-berlin.de>, morceaudemerdeSNIPt...@hotmail.com says...
>interesting post snipped<
Thanks, Ken and Jim. A well conceived and executed comparison.
One question about the setup on the vibe pump. Is the bypass on the pourover Junior the standard fitting? I seem to recall Ken mentioning it last year, but IIRC it didn't seem to work properly at that time.
Also, do you have any sense of what the water debit is under pressure with the bypass in place? I wondered if the bypass volume made the debits of the 2 machines more equal, or if the debit was lower on the pourover machine?
> In article <c4aiin$2falt...@ID-192261.news.uni-berlin.de>, > morceaudemerdeSNIPt...@hotmail.com says...
> >interesting post snipped<
> Thanks, Ken and Jim. A well conceived and executed comparison.
> One question about the setup on the vibe pump. Is the bypass on the > pourover Junior the standard fitting? I seem to recall Ken > mentioning it last year, but IIRC it didn't seem to work properly at > that time.
this is the standard fitting, however apparently in Europe they actually use it but in the machines shipped to the USA there is no hose attached and hence the valve is shut and not used. When carefully adjusted the water that recirculates back to the tank is a small fraction of what is used in shotmaking. My fault in looking at it before was that I made no attempt to adjust it nor measure the output. The same sort of Cimbali silicone hose that serves the water supply from tank to the vibe pump, works in this application.
> Also, do you have any sense of what the water debit is under pressure > with the bypass in place? I wondered if the bypass volume made the > debits of the 2 machines more equal, or if the debit was lower on the > pourover machine?
The stated water debits in the post above compare the rotary machine to the vibe machine with the bypass valve operational and adjusted, with the tubing in place. The measured amounts are essentially equivalent.
> > Thanks, Ken and Jim. A well conceived and executed comparison.
> > One question about the setup on the vibe pump. Is the bypass on the > > pourover Junior the standard fitting? I seem to recall Ken > > mentioning it last year, but IIRC it didn't seem to work properly at > > that time.
> this is the standard fitting, however apparently in Europe they actually use > it but in the machines shipped to the USA there is no hose attached and > hence the valve is shut and not used. When carefully adjusted the water > that recirculates back to the tank is a small fraction of what is used in > shotmaking. My fault in looking at it before was that I made no attempt to > adjust it nor measure the output. The same sort of Cimbali silicone hose > that serves the water supply from tank to the vibe pump, works in this > application.
> > Also, do you have any sense of what the water debit is under pressure > > with the bypass in place? I wondered if the bypass volume made the > > debits of the 2 machines more equal, or if the debit was lower on the > > pourover machine?
> The stated water debits in the post above compare the rotary machine to the > vibe machine with the bypass valve operational and adjusted, with the tubing > in place. The measured amounts are essentially equivalent.
Nice test and very significant result. I haven't had the opportunity to do this sort of test and I'm very glad to see you two perform it. The results don't surprise me at all. Another myth properly shot down by alt.coffee researchers.
I think autodosing is more trouble than it is worth in a home machine. However, autofill is a worthwhile feature to have and not especially trouble prone as the electronics are not that complex. I wouldn't avoid a machine just 'cause it has autofill and machines with sight glasses are becoming increasingly rare. And just as in used cars, generally the newer machine the better and limiting yourself to machines w.o. autofill will restrict you to an aging pool of equipment.
"Ken Fox" <morceaudemerdeSNIPt...@hotmail.com> wrote in message
> electronically controlled commercial machines, his advice to alties seeking > the ultimate one group commercial machine for use in the home, would be to > seek out a very well maintained older commercial machine in good condition, > with manual (electric rather than electronic) controls, with sightglass and > no autofill. This would be preferable in most respects to any of the > examples of the current crop of electronically controlled machines; the > results in the cup are equal and the hassles in home use are far fewer with > machines of older vintage.
> I think autodosing is more trouble than it is worth in a home machine. > However, autofill is a worthwhile feature to have and not especially trouble > prone as the electronics are not that complex. I wouldn't avoid a machine > just 'cause it has autofill and machines with sight glasses are becoming > increasingly rare. And just as in used cars, generally the newer machine the > better and limiting yourself to machines w.o. autofill will restrict you to > an aging pool of equipment.
Hi Jack.
Still, if it was me and if I could find a non-autofill non-autodose, completely non-electronic one group commercial of good quality in good condition, I would prefer it to a machine with any of those electronics. I don't think there has been a significant improvement that matters to the home enthusiast in this type of basic commercial machine, in the last 20 years. Autofill is completely useless and potentially detrimental in home use as long as you don't use your boiler water to make drinks with (sorta like drinking out of the sewer in my opinion in any event).
A basic espresso machine has few parts that are apt to need replacing in normal use with good maintenance; we are talking about a few seals, basically, maybe a new vibe pump, the odd pstat or solenoid, and these are cheap. So, the benefits of getting a more modern electronics-laden machine in my view are outweighed by the detriments of the troublesome electronic stuff itself.
> "Ken Fox" <morceaudemerdeSNIPt...@hotmail.com> wrote in message > news:c4aiin$2falth$1@ID-192261.news.uni-berlin.de... > snip > Given some unpleasant experiences with modern > > electronically controlled commercial machines, his advice to alties > seeking > > the ultimate one group commercial machine for use in the home, would be to > > seek out a very well maintained older commercial machine in good > condition, > > with manual (electric rather than electronic) controls, with sightglass > and > > no autofill. This would be preferable in most respects to any of the > > examples of the current crop of electronically controlled machines; the > > results in the cup are equal and the hassles in home use are far fewer > with > > machines of older vintage.
> The rotary showed 9.25 bar on the pressure guage pf. After getting the OPV > (overpressure valve) on the vibe up to speed with a hose back to the tank, > and a little spring loosening, we were able to adjust it so it oscillated > around 9 bar +/- .125 or so. Originally, it had oscillated wildly between 12 > and 14 bar, with the OPV not bypassing any water. The more water the OPV > bypassed, the smoother the pressure reading became.
Ken and Jim,
Did You happen to take the time to test the difference in taste and shot quality on the vibe after the pressure adjustment. I never noticed significant difference when I did the pressure mod with my Silvia. Would it be worth a look on the Junior?
> > The rotary showed 9.25 bar on the pressure guage pf. After getting the OPV > > (overpressure valve) on the vibe up to speed with a hose back to the tank, > > and a little spring loosening, we were able to adjust it so it oscillated > > around 9 bar +/- .125 or so. Originally, it had oscillated wildly between > 12 > > and 14 bar, with the OPV not bypassing any water. The more water the OPV > > bypassed, the smoother the pressure reading became.
> Ken and Jim,
> Did You happen to take the time to test the difference in taste and shot > quality on the vibe after the pressure adjustment. I never noticed > significant difference when I did the pressure mod with my Silvia. Would it > be worth a look on the Junior?
> Thanks, > Steve in Boise
Hi Steve,
Jim and I did not test the taste difference between vibe with pressure adjustment and without. We did not do it, for a couple of reasons. One is that we considered it a more or less settled issue.
In the original Jr. vs. Jr. post I did last year, I used a non-coffee-freak friend as my test subject and using canned Illy beans, my friend consistently preferred the product of the rotary machine in a blind taste test. This test was done prior to the overflow pressure valve modification on the vibe machine. My own personal, non-blinded, testing done at the same time showed the same result. Jim has done an overpressure valve pressure adjustment on his own Isomac Tea, and has found benefit from it in his own repetitive testing. Also, I had previously found most of my near-choking shots (grind too fine) from the vibe machine to be undrinkable, whereas such shots from the rotary were good ristrettos. When we did our new revised vibe vs. rotary test a couple of days ago, neither of us could tell the difference consistently between vibe and rotary produced shots, which would have been obvious before the pressure mod was done on the vibe machine.
On the pourover Jr. the "modification" is simply taking appropriately sized silicone tubing and inserting it on the overpressure valve "nipple" and threading the tubing around so that the bypassed water ends up back in the pourover tank; it took about 5 minutes to do. Adjusting the overpressure valve itself is a bit time consuming as you need to have a PF manometer and it takes repetitive testing over time to show that the adjustment has taken.
> Nice test and very significant result. I haven't had the opportunity > to do this sort of test and I'm very glad to see you two perform it. > The results don't surprise me at all. Another myth properly shot down > by alt.coffee researchers.
> Hooray for you!
> -Greg
Don't agree at all. Would need to do many more repetitions before statistical significance would be reached to support any conclusion at all.
> > Nice test and very significant result. I haven't had the opportunity > > to do this sort of test and I'm very glad to see you two perform it. > > The results don't surprise me at all. Another myth properly shot down > > by alt.coffee researchers.
> > Hooray for you!
> > -Greg
> Don't agree at all. Would need to do many more repetitions before > statistical significance would be reached to support any conclusion at > all.
> Ray
Greg is a scientist, which I am not, however my own pre-professional training included several publications in scientific journals, in which the scientific method was rigourously followed and the data was subjected to appropriate statistical testing. In his posting I'm sure Greg did not mean "significant" in the sense of this data "has met the criteria for statistical significance at the 0.01 percent level." Rather, what he meant was that this was something he found interesting and difficult to test ("one would need the right machines in the same place and people who can taste espresso critically available and to set up the test bed appropriately in order to make relevant observations").
I believe that our results are valid, or at least as valid as say, Robert Parker's evaluations of the 2002 vintage of Red Bordeaux wine. This is, however, as you point out, not research that would meet the criteria for publication in a major scientific journal. I would submit that it is probably nearly impossible to conduct an experiment on this subject that would rise to that level, so like many subjects one must take the best information one has available and try to use it, because the alternative is no data at all.
I used to practice medicine and that was the sort of thing we had to do every day, to make the best of the data you had available and to try to infer the rest; you would only know the whole picture when the autopsy report was in.
>I believe that our results are valid, or at least as valid as say, Robert >Parker's evaluations of the 2002 vintage of Red Bordeaux wine. This is, >however, as you point out, not research that would meet the criteria for >publication in a major scientific journal. I would submit that it is >probably nearly impossible to conduct an experiment on this subject that >would rise to that level, so like many subjects one must take the best >information one has available and try to use it, because the alternative is >no data at all. >
i believe the appropriate phrase is: "merits further study."
Secondly -- let me make a heretical suggestion (just for fun). Of course, I know subjective taste is the main endpoint here, but maybe there are some objective measurements of espresso from these excellent experiments of yours that would be interesting to try.
I haven't come across any posts using quantitative measures of espresso (like, say, viscosity or specific gravity; or performing organic chemistry to get fractions of oils in the shots) to try to avoid perception biases. There must be materials engineering / orgo folks here who could suggest how to measure these things for espresso. Like, for viscocity -- how much of the shot clings to the side of a sluice of a certain material, held at a certain angle. Or, using a hydrometer/refractometer to measure specific gravity (i.e., sugar content) of a diluted shot.
"Ken Fox" <morceaudemerdeSNIPt...@hotmail.com> wrote in message <news:c4aiin$2falth$1@ID-192261.news.uni-berlin.de>... > Originally, it had oscillated wildly between 12 > and 14 bar, with the OPV not bypassing any water.
Not sure if I get this (I own a lever machine anyway...). Was that a fault of the machine, or something normal for vibe pumps? And if the latter is the case, would you recommend doing similar modifications on vibe machines to get a smoother output?
> > Originally, it had oscillated wildly between 12 > > and 14 bar, with the OPV not bypassing any water.
> Not sure if I get this (I own a lever machine anyway...). Was that a > fault of the machine, or something normal for vibe pumps? And if the > latter is the case, would you recommend doing similar modifications on > vibe machines to get a smoother output?
normal behavior of a vibe pump unless modified somehow in the machine. I think it depends on the machine; in most cases you probably have to live with it.
Imagine a sine wave going up and down between 12 and 14 bar (this is the diaphragm of the vibe pump swinging back and forth at 50/ 60 cycles). Then imagine relieving any pressure that is in excess of 9 bar and drawing a steady line at 9.(that's the relief valve) - any pressure that is over 9 bars, whether it is 10 or 12 or 14 or 20, fluctuating or steady , blows out of the relief valve and is not perceived at the gauge or brew head.
Theoretically, the pressure of the vibe pump drops to zero on each return stroke of the pump. However, the system as a whole remains pressurized, but the pressure starts to bleed off slightly in the short interval before the next pressure stroke begins. Think of pumping up a bicycle tire, one with a pinhole leak in it, with a hand pump. In between each stroke, the pressure would start to fall off, but not to zero.
"CoffeeGreek" <coffeegr...@fastmail.gr> wrote in message
> > Originally, it had oscillated wildly between 12 > > and 14 bar, with the OPV not bypassing any water.
> Not sure if I get this (I own a lever machine anyway...). Was that a > fault of the machine, or something normal for vibe pumps? And if the > latter is the case, would you recommend doing similar modifications on > vibe machines to get a smoother output?
If you notice the sales drivel for a lot of vibe pump machines, particularly home machines, it'll say something like "pump produces up to 16 bars of pressure", and act luck it's a selling feature. Well it's laziness and cheapness on the part of the manufacturer. Those pressure ratings are what the vibe pump will produce if it's pushing against a blind filter, if there is no provision for pressure regulation, something not usually supplied. I found it useful to implement such regulation on my vibe pump machine. You can google search on Silvia pressure regulation and read quite a bit about it. The approach used on the Silvias is quite applicable to any number of vibe pump equipped machines.
> Not sure if I get this (I own a lever machine anyway...). Was that a > fault of the machine, or something normal for vibe pumps? And if the > latter is the case, would you recommend doing similar modifications on > vibe machines to get a smoother output?
Ken Fox wrote: > Two nearly identical one group commercial > machines, differing primarily in the type of pump they contained, > allowed direct comparison of espressos prepared with either a > vibratory or a rotary pump. <snip> > We both agreed that the > difference between the machines was too small to distinguish from the > shot to shot variability of either one. > These are not entirely idle results, since we spent the morning > adjusting the vibe machine, a pourover, dating from late 1995 to > duplicate the basic characteristics of the rotary, one year old and > plumbed in, as closely as possible.
This is quite a significant result, it's great that you did such careful testing. It's really not all that surprising, considering: (1) you got the peak pressure pretty much standardized (2) the vibe pump and the rotary pump both produce an oscillating pressure wave, although the frequency and amplitude differ somewhat
> We compared results. Of the four paired blind-tasted shots I had > prepared, and Ken had tasted, two pairs had tied, the rotary won one, > and the vibe won the other.
I can understand how you could run this experiment using "blind-tasting" techniques. But the pumps make such different sounds that you'd have to "deaf-taste" the shots, too. So how did you do that?
> The 250 mL water debit from the two machines: rotary -- 23.5 seconds, > vibe -- 26.2 seconds.
Just curious, why did you measure it at 250 ml? Al's standard water debit measuring technique (ml delivered in 10 seconds), or something similar (say, seconds to 100ml) would be more useful. IMO, it's only in the first few seconds of the extraction that water debit is significant (that's when the puck is "set up" for extraction). After that the water debit is nearly irrelevant, unless it's way way off. Measuring it at 250 ml could result in similar numbers from machines that deliver water very differently in the first few seconds.
I doubt this particular nuance affected the results of your test one way or another, however.
> All in all, these results support the thesis that once a vibe pump is > adjusted to operate at the same pressure as a rotary, the taste > differences due to 60 Hz pressure wave are either non-existent or > very small.
Again, a Procon (or similar) pump running on 60 Hz electric produces a 115 Hz pressure wave (1725 rpm / 60 sec x 4 vanes per revolution). The amplitude of the pressure wave depends on the source pressure, but it's probably always less than that of the vibe pump.
> Ken's prior test comparing these two machines showed a preference for > the product of the rotary over the vibe machine. Today's test varied > in the more rigorous attempt to control variables such as temperature > (regardless of front panel pressure gauge reading) and the pump > pressures. Once these variables were equalized there was no reliable > difference between the output of the two machines.
I don't doubt your results, but the sample size was so small that the experiment bears repetition. If I ever find the time, I will hook up the original vibe pump and a pressure regulator in parallel with the rotary pump on my machine. Switching from one pump to the other on the same machine would be an even more direct way of testing. Of course, given my history of grand plans this may never happen. Meanwhile, thanks for doing such a thorough and interesting experiment. --
> I can understand how you could run this experiment using "blind-tasting" > techniques. But the pumps make such different sounds that you'd have to > "deaf-taste" the shots, too. So how did you do that?
Both machines were set up. We found it was possible to get equivalent shots from both machines with the same grinder setting. The taster was at the dining room table, facing away from the espresso machine area, and adjacent to the kitchen. The guy preparing the shots would grind enough coffee to fill two PFs (they were identical double baskets installed in single spout PFs), then fill and tamp the baskets, then flush the groups more or less simultaneously, then pull the shots more or less simultaneously. The machines were located across from each other, one on the main kitchen counter and the other on the kitchen island 3 feet away. What I did was start one shot, then maybe 5 seconds later started the second, darting back and forth to keep an eye on both shots (which is a bit of a challenge!). I'm sure Jim did something very similar, judging by the noises I heard. I have some Italian porcelain cups that are identical, from a specific designer's series but with 2 different patterns. One is predominently white and the other predominently black. The taster got presented two shots, one in black cup and one in a white cup, at the same time. Neither shot was more than 5 seconds older than the other one, and the order of preparation was random. The taster did not know which was which, and it varied from shot to shot (I did it arbitrarily as I am sure that Jim did also). The shot puller would write down which machine produced which colored cup's shot for each round. The taster wrote down his observations, comparing each shot to each other, black cup vs. white one, in real time. The "identity" of each shot, e.g. which machine pulled each shot, was not revealed until we discussed them at the end of the test session.
> > The 250 mL water debit from the two machines: rotary -- 23.5 seconds, > > vibe -- 26.2 seconds.
> Just curious, why did you measure it at 250 ml? Al's standard water debit > measuring technique (ml delivered in 10 seconds), or something similar (say, > seconds to 100ml) would be more useful. IMO, it's only in the first few seconds > of the extraction that water debit is significant (that's when the puck is "set > up" for extraction). After that the water debit is nearly irrelevant, unless > it's way way off. Measuring it at 250 ml could result in similar numbers from > machines that deliver water very differently in the first few seconds.
> I doubt this particular nuance affected the results of your test one way or > another, however.
ask Jim; he designed this test and he's smarter than I am.
> > All in all, these results support the thesis that once a vibe pump is > > adjusted to operate at the same pressure as a rotary, the taste > > differences due to 60 Hz pressure wave are either non-existent or > > very small.
> Again, a Procon (or similar) pump running on 60 Hz electric produces a 115 Hz > pressure wave (1725 rpm / 60 sec x 4 vanes per revolution). The amplitude of the > pressure wave depends on the source pressure, but it's probably always less than > that of the vibe pump.
Ask Jim; I'm not smart enough to have an opinion on stuff like this.
> > Ken's prior test comparing these two machines showed a preference for > > the product of the rotary over the vibe machine. Today's test varied > > in the more rigorous attempt to control variables such as temperature > > (regardless of front panel pressure gauge reading) and the pump > > pressures. Once these variables were equalized there was no reliable > > difference between the output of the two machines.
> I don't doubt your results, but the sample size was so small that the experiment > bears repetition. If I ever find the time, I will hook up the original vibe pump > and a pressure regulator in parallel with the rotary pump on my machine. > Switching from one pump to the other on the same machine would be an even more > direct way of testing. Of course, given my history of grand plans this may never > happen. Meanwhile, thanks for doing such a thorough and interesting
experiment.
Please do repeat the test; I think and I'm sure Jim thinks that this would provide useful information. We would certainly have thought about doing a few additional rounds of testing (as time consuming as it had already been notwithstanding) if we had had an inkling that it might matter. As it turned out there were several rounds where the two cups presented were nearly identical in our evaluation. Given that reality, and the fact that the remaining rounds varied back and forth between machines, we felt that no matter how many rounds we had done, the results would be equivocal.
I guess that what we are saying is that there are lots of variables in espresso production, only some of which we know of and fewer of which we can control. Given a competent barista and good coffee, the effect of the pump type, rotary vs. vibe (corrected for overpressure), was too small to pick out of that mix and hence was a relatively minor factor. Perhaps a very large sample size would show that one type of pump actually produces better shots. Even if this is the case it will likely still be a small effect, overwhelmed by such others as coffee quality and freshness, appropriate grind, and espressomaking skills.
> Ken Fox wrote: > <snip deaf and blind, but not dumb, methodology>
> cool. > --
> -Andy S. "how's that for a barry-like response?"
Of course when you do this and when you have two people alternating doing it, who don't dose and tamp exactly the same, there is the risk that one of the shots you are pulling will channel or choke the machine. If either shot looked bad the shot puller had the option of cancelling that particular round before it got tasted. This happened at least twice that I can recall, and those shots not counted added not only the time necessary to grind, pull and dismiss them, but also the time before and after that we allowed in between rounds for machine and tastebud recovery which was about half an hour each round. Getting 8 simultaneous, acceptable, double shots pulled and tasted took a lot longer than you might think (like a whole afternoon, actually).
<schec...@rochester.rr.com> wrote: >Just curious, why did you measure it at 250 ml? Al's standard water debit >measuring technique (ml delivered in 10 seconds), or something similar (say, >seconds to 100ml) would be more useful.
The 250 mL measuring cups were convenient, and I did it this way. The flow is rather turbulent for the first 10 seconds or so, since the water is boiling, so it's not that easy to measure volume (although timing would work fine).
It's easy enough to convert the measures: 106 mL and 95 mL for the rotary and vibe in 10 seconds.
<ba...@rileys-coffee.com> wrote: >i believe the appropriate phrase is: "merits further study."
>:)
You're right, but our tastebuds gave out after eight shots, even with about an hour's pause between flights.
My honest opinion is that "espresso cupping" of any sort is a rather blunt instrument compared to coffee cupping, at least for me, since my tastebuds get overwhelmed. I can pick up large and consistent differences, but anything smaller and more haphazard gets lost in the noise, so to speak, even if it is a real effect. I've been thinking about this for a long while, and haven't come up with any bright ideas for a general solution; although there's sometimes a work around.
>My honest opinion is that "espresso cupping" of any sort is a rather >blunt instrument compared to coffee cupping, at least for me, since my >tastebuds get overwhelmed.
have you tried diluting the shots and then "cupping" them normally?
