jim schulman
2004-06-17 15:21:56 UTC
THE BACKSTORY
Since trying the Elektra lever, I've been jonesing for a lever
machine. Not any old lever machine, but a spring lever machine. The
Pavoni shots I've had have been OK but forgettable, the Elektra's
shots were outstanding, not in their crema and body, but in the taste
they delivered with high quality blends (forget robusta blends, they
taste worse). Apparently, the lever action, with it's
smoother-than-rotary delivery, and declining pressure through the
pull, improves the clarity of the shot's flavor. The same
characteristics also reduce the crema, so it's not the ultimate in
espresso, but it may be the ultimate in coffee.
The problem was home lever machines overheat, including the Elektra,
and are only good for a few shots in a row, while commercial levers
are way too big for my kitchen. So how was I going to make my espresso
blend comparisons on a lever machine?
Enter Roger Barrett and his alluring tales of the La Peppina. This odd
little machine has no boiler, rather an enamel kettle that boils the
water sitting directly atop a cylinder with a spring lever that pushes
up. The compressed flow is valved into a side mounted group. I've
posted lots of pics and a shop drawing on abc.
This arrangement has lots of advantages. The pot can be refilled on
the fly, since there's no pressure, just a simple lid. Steam pressure
isn't required to fill the cylinder like on other lever machines,
gravity does the job, so the water temperature in the kettle can be
anywhere one wants.
The only problem is that the machine hasn't been made since about
1980. They still are found quite often on ebay and sell for around
$190 when in good condition. After missing out on two auctions, I took
the "buy it now" route, and for the aforesaid $190 plus $10 shipping
got a machine. I figured it would be a coin toss whether the thing
would even work, never mind work well.
THE MACHINE
Vince, the seller, is an honest man; and the machine is in mint
condition, inside and out. In fact it's in better shape than he
described. My heartfelt thanks to him. I was pulling shots within an
hour of it arriving via fed-ex.
This is not a commercial machine. The PF and group are chrome plated
aluminum and 49mm in size, the base and kettle lid are plastic, the
drip tray is tiny. It is however very well and simply built, and there
are no mickey-mouse features. Well, none except the second pressure
cooker lid with steam wand. This actually sort of works, similar to a
steam toy steamer, but is not going to bring any joy. However, I
didn't buy the machine for steaming, and the modified steamer lid
detracts from the machine's functionality as a shot puller.
The version I got is the earlier one, drawing 700 watts, as compared
to the later 1100 watt one. It got to a boil in 5 minutes, so this is
not a problem. In fact, it is probably preferable (when not steaming),
since there is no control of any sort and the heat runs at full power.
With the bigger heater, the water would just boil more furiously.
The lever arrangement is decidedly odd. It looks like the handle on a
slot machine. In fact, the whole machine looks somewhat surreal, as if
Marcel Duchamp went to a junkyard, got an old steam pressure espresso
machine group, welded it to a enamelled tea kettle, and mounted it all
on a slot machine chassis. However, the operation is easy -- just pull
the lever down, cocking the spring, hold it for about 10 seconds, then
release to to get another 15 seconds or so of pull. The only trick is
that it's best to cock it to below the level of the machine itself, so
the machine should be pulled to the counter edge to let the lever
clear the edge when pulled down.
The draw on the lever is about a 1/2 ounce. This can be brought up to
just under 1 ounce by pumping slightly while the lever is down;
however, with the single basket's 8 gram capacity, the 1/2 ounce shot
makes for a perfect ristretto. My initial comparisons to the Tea were
promising. Like the Elektra, it produced less crema, but an improved
flavor for the all washed arabica blend I was trying. I got very nice
shots from the get go. Since I sincerely doubt I'm any sort of lever
prodigy, I'm guessing this is a fairly simple machine to use.
ITS PERFORMANCE
I have no way to measure the lever pressure, but I'm guessing it's in
the ballpark. This is because the machine requires a grind only a
shade coarser than the Tea, and well within commercial espresso range.
Varying the grind finer slows the shot and also reduces volume, why I
don't know. The volume change can be made up by pumping on the lever,
but getting good at this will require more practice. The best tasting
shots had an odd extraction profile -- 10 seconds of lever down
infusion, followed by a 15 second spring pull. Longer lasting pulls
resulted in a taste with the tell-tale overextraction taint -- an
instant coffee taste -- that I only get on the Tea if I really blow a
grind and go about 45 seconds.
There's a screw covering the drilled channel between the piston outlet
and the group. By removing the screw one can measure the water
temperature as it reaches the top of the puck (not while making shots
of course). With the water boiling this settled to a fairly steady
temperature betwen 99C and 99.3C
Inserting a TC into the PF and sealing with silver foil got me
consistent shot temperatures within shots between 97.0 and 98.5. These
temperatures repeated shot after shot with no change. Since I don't
have spare PFs or baskets, I can't set up a Schomer rig to measure
actual shot temps. However, the Schomer measure for the Tea closely
parallels this "tin-foil" measure, so I see no reason why it should be
inaccurate here. The exit temperature directly beneath the basket when
pulling shots was 92 to 92.5. These too remained stable shot after
shot.
While these temperatures are about 2.5C too high, they are quite
stable. They don't vary at all from shot to shot, and less than 2C
within a given shot. Not quite as steller as an E61, Cimbali or LM,
but well within commercial range, and far better than any machine
going for less than $700. The simple fact of the water in the kettle
not being replaced by cold water as the shot is drawn removes the
major cause of temperature instability in pump machines. The fact of
the lever not needing a pressurized boiler way above shot temperature
to fill it removes the major cause of instability in lever machines.
The La Peppina has superb temperature stability because it is too
primitive to be unstable; what a concept! The only source of
instability is the lightweight group, which probably accounts for the
1.5C within-shot variation.
The shot temperature can be brought down by the simple expedient of
pulling shots "off-boil", that is turning the machine off, waiting a
few minutes for the boiling to subside, then pulling the shot. This
brings all the temperatures down by 2C to 4C and maintains the same
within shot stability as before. Providing one turns the machine back
on to boil, then goes to off boil, for each shot, one can maintain
shot to shot stability.
The boil/off boil thing is a pita, and it suggests PIDing the machine
to get the temperatures one wants exactly and repeatably (a stem
thermometer through the lid would probably work too). Before reporting
on the PID results, a short note. Besides overheating, the
"portafilter sneeze" was thought to be a barrier to pulling multiple
shots fast on a home lever machine. This turns out not to be the case,
since there is a dodge: One can pull the lever down to depressurize
the PF, remove the PF immediately, and let the lever up -- carefully
-- to clean off the screen. In this way one can pull a shot every 90
seconds or so.
IS PID + GRAVITY LEVER THE ULTIMATE IN ESPRESSO MACHINE?
All this perfomance data suggests that a PID might make a nice
addendum to the machine. Since my roaster PID is built with plugs for
the power lines and TC, it was simple to plug the Peppina into it, run
a TC into the kettle, and retune the PID to keep the kettle at a
steady, user selected temperature. Until the versalab rudely
interrupted my thoughts, I had speculated that a PIDed, temperature
stable lever machine could be the ultimate shot puller. Now stay tuned
to find out if it could at least be the ultimate $400 shot puller.
After a long wait on tuning (enamelled steel holds temperature very
well), I set the control down to 97.5C and started pulling shots with
a foil stuffed PF. After warming up the PF with a couple of pulls, I
got between 95C and 95.5C pull after pull until the water ran low.
Yes! Close to Schomer territory with a little 1970s lever machine. I'm
puzzled about the better within-shot stability at this temperature
compared to the boiling temperature, and would appreciate suggestions.
My guess is that the still boiling water's variable release of steam
can affect the stability, while the 97.5C water is less prone to this.
OK, crunch time. How does the souped up Peppina fare against my
reigning kitchen champ, the Tea? No way I could do blind shot
comparisons, so you'll have to either trust or dismiss my biased
version. If there's enough interest, I may set up multi-person tests,
or comparisons to other machines. The whole Peppina/Pid setup weighs
about 15lbs and fits in a kit bag, so "have machine, will travel" is
quite feasible.
The Tea took the first round. The single La Peppina shots, at under
1/2 ounce, were too small to savor for taste, and didn't have the
crema or body of the Tea's. The Peppina took the second round. I
switched to the double basket and started pumping the lever for the
preinfuse until I could feel some backpressure (per Roger's advice).
This produced a 3/4 ounce shot from 12 grams, a little less
concentrated than the 1 1/4 ounce 20 gram shots from the Tea, but now,
to my utter surprise, with crema equal to the Tea's and a buttery
seductive mouthfeel that was actually better. The flavors were also a
hair better, less rough edges, and slightly better aromatics. I'd give
it maybe 3 to 4 points more out of 100 than my Tea shots. The third
round was a wash, the Tea's shot had better crema and aroma, the
Peppina's better mouthfeel and taste. I pulled this round into glass
cups, and posted the pics for a visual comparison.
CONCLUSION
Even PID'ed, the Peppina isn't the ultimate espresso machine. But in
terms of shot quality, it can hold it's own against most commercial
machines, even though it has an underweight, undersized group and an
operator with a total of 24 hours lever experience. For a $200 machine
with a jury rigged PID added, that's pretty impressive. As proof of
concept for this gravity fed lever design, it's even more impressive.
The lever pump provides a smoother extraction than the rotary, and the
sealed water path (no cold or overheated water entering) utterly
simplifies the control task. The super impressive Versalab machine
also uses a sealed water path and a new style, super smooth pump that
can operate in near boiling water. It's undoubtedly better than my La
Peppina rig, but it also costs 25 times as much. This test shows that
the design features that make it so good can be had at much lower cost
-- a Peppina style lever machine with a beefier group, more accurate
PID, and a well engineered spring or compressed air driven piston
could perhaps do it.
Since trying the Elektra lever, I've been jonesing for a lever
machine. Not any old lever machine, but a spring lever machine. The
Pavoni shots I've had have been OK but forgettable, the Elektra's
shots were outstanding, not in their crema and body, but in the taste
they delivered with high quality blends (forget robusta blends, they
taste worse). Apparently, the lever action, with it's
smoother-than-rotary delivery, and declining pressure through the
pull, improves the clarity of the shot's flavor. The same
characteristics also reduce the crema, so it's not the ultimate in
espresso, but it may be the ultimate in coffee.
The problem was home lever machines overheat, including the Elektra,
and are only good for a few shots in a row, while commercial levers
are way too big for my kitchen. So how was I going to make my espresso
blend comparisons on a lever machine?
Enter Roger Barrett and his alluring tales of the La Peppina. This odd
little machine has no boiler, rather an enamel kettle that boils the
water sitting directly atop a cylinder with a spring lever that pushes
up. The compressed flow is valved into a side mounted group. I've
posted lots of pics and a shop drawing on abc.
This arrangement has lots of advantages. The pot can be refilled on
the fly, since there's no pressure, just a simple lid. Steam pressure
isn't required to fill the cylinder like on other lever machines,
gravity does the job, so the water temperature in the kettle can be
anywhere one wants.
The only problem is that the machine hasn't been made since about
1980. They still are found quite often on ebay and sell for around
$190 when in good condition. After missing out on two auctions, I took
the "buy it now" route, and for the aforesaid $190 plus $10 shipping
got a machine. I figured it would be a coin toss whether the thing
would even work, never mind work well.
THE MACHINE
Vince, the seller, is an honest man; and the machine is in mint
condition, inside and out. In fact it's in better shape than he
described. My heartfelt thanks to him. I was pulling shots within an
hour of it arriving via fed-ex.
This is not a commercial machine. The PF and group are chrome plated
aluminum and 49mm in size, the base and kettle lid are plastic, the
drip tray is tiny. It is however very well and simply built, and there
are no mickey-mouse features. Well, none except the second pressure
cooker lid with steam wand. This actually sort of works, similar to a
steam toy steamer, but is not going to bring any joy. However, I
didn't buy the machine for steaming, and the modified steamer lid
detracts from the machine's functionality as a shot puller.
The version I got is the earlier one, drawing 700 watts, as compared
to the later 1100 watt one. It got to a boil in 5 minutes, so this is
not a problem. In fact, it is probably preferable (when not steaming),
since there is no control of any sort and the heat runs at full power.
With the bigger heater, the water would just boil more furiously.
The lever arrangement is decidedly odd. It looks like the handle on a
slot machine. In fact, the whole machine looks somewhat surreal, as if
Marcel Duchamp went to a junkyard, got an old steam pressure espresso
machine group, welded it to a enamelled tea kettle, and mounted it all
on a slot machine chassis. However, the operation is easy -- just pull
the lever down, cocking the spring, hold it for about 10 seconds, then
release to to get another 15 seconds or so of pull. The only trick is
that it's best to cock it to below the level of the machine itself, so
the machine should be pulled to the counter edge to let the lever
clear the edge when pulled down.
The draw on the lever is about a 1/2 ounce. This can be brought up to
just under 1 ounce by pumping slightly while the lever is down;
however, with the single basket's 8 gram capacity, the 1/2 ounce shot
makes for a perfect ristretto. My initial comparisons to the Tea were
promising. Like the Elektra, it produced less crema, but an improved
flavor for the all washed arabica blend I was trying. I got very nice
shots from the get go. Since I sincerely doubt I'm any sort of lever
prodigy, I'm guessing this is a fairly simple machine to use.
ITS PERFORMANCE
I have no way to measure the lever pressure, but I'm guessing it's in
the ballpark. This is because the machine requires a grind only a
shade coarser than the Tea, and well within commercial espresso range.
Varying the grind finer slows the shot and also reduces volume, why I
don't know. The volume change can be made up by pumping on the lever,
but getting good at this will require more practice. The best tasting
shots had an odd extraction profile -- 10 seconds of lever down
infusion, followed by a 15 second spring pull. Longer lasting pulls
resulted in a taste with the tell-tale overextraction taint -- an
instant coffee taste -- that I only get on the Tea if I really blow a
grind and go about 45 seconds.
There's a screw covering the drilled channel between the piston outlet
and the group. By removing the screw one can measure the water
temperature as it reaches the top of the puck (not while making shots
of course). With the water boiling this settled to a fairly steady
temperature betwen 99C and 99.3C
Inserting a TC into the PF and sealing with silver foil got me
consistent shot temperatures within shots between 97.0 and 98.5. These
temperatures repeated shot after shot with no change. Since I don't
have spare PFs or baskets, I can't set up a Schomer rig to measure
actual shot temps. However, the Schomer measure for the Tea closely
parallels this "tin-foil" measure, so I see no reason why it should be
inaccurate here. The exit temperature directly beneath the basket when
pulling shots was 92 to 92.5. These too remained stable shot after
shot.
While these temperatures are about 2.5C too high, they are quite
stable. They don't vary at all from shot to shot, and less than 2C
within a given shot. Not quite as steller as an E61, Cimbali or LM,
but well within commercial range, and far better than any machine
going for less than $700. The simple fact of the water in the kettle
not being replaced by cold water as the shot is drawn removes the
major cause of temperature instability in pump machines. The fact of
the lever not needing a pressurized boiler way above shot temperature
to fill it removes the major cause of instability in lever machines.
The La Peppina has superb temperature stability because it is too
primitive to be unstable; what a concept! The only source of
instability is the lightweight group, which probably accounts for the
1.5C within-shot variation.
The shot temperature can be brought down by the simple expedient of
pulling shots "off-boil", that is turning the machine off, waiting a
few minutes for the boiling to subside, then pulling the shot. This
brings all the temperatures down by 2C to 4C and maintains the same
within shot stability as before. Providing one turns the machine back
on to boil, then goes to off boil, for each shot, one can maintain
shot to shot stability.
The boil/off boil thing is a pita, and it suggests PIDing the machine
to get the temperatures one wants exactly and repeatably (a stem
thermometer through the lid would probably work too). Before reporting
on the PID results, a short note. Besides overheating, the
"portafilter sneeze" was thought to be a barrier to pulling multiple
shots fast on a home lever machine. This turns out not to be the case,
since there is a dodge: One can pull the lever down to depressurize
the PF, remove the PF immediately, and let the lever up -- carefully
-- to clean off the screen. In this way one can pull a shot every 90
seconds or so.
IS PID + GRAVITY LEVER THE ULTIMATE IN ESPRESSO MACHINE?
All this perfomance data suggests that a PID might make a nice
addendum to the machine. Since my roaster PID is built with plugs for
the power lines and TC, it was simple to plug the Peppina into it, run
a TC into the kettle, and retune the PID to keep the kettle at a
steady, user selected temperature. Until the versalab rudely
interrupted my thoughts, I had speculated that a PIDed, temperature
stable lever machine could be the ultimate shot puller. Now stay tuned
to find out if it could at least be the ultimate $400 shot puller.
After a long wait on tuning (enamelled steel holds temperature very
well), I set the control down to 97.5C and started pulling shots with
a foil stuffed PF. After warming up the PF with a couple of pulls, I
got between 95C and 95.5C pull after pull until the water ran low.
Yes! Close to Schomer territory with a little 1970s lever machine. I'm
puzzled about the better within-shot stability at this temperature
compared to the boiling temperature, and would appreciate suggestions.
My guess is that the still boiling water's variable release of steam
can affect the stability, while the 97.5C water is less prone to this.
OK, crunch time. How does the souped up Peppina fare against my
reigning kitchen champ, the Tea? No way I could do blind shot
comparisons, so you'll have to either trust or dismiss my biased
version. If there's enough interest, I may set up multi-person tests,
or comparisons to other machines. The whole Peppina/Pid setup weighs
about 15lbs and fits in a kit bag, so "have machine, will travel" is
quite feasible.
The Tea took the first round. The single La Peppina shots, at under
1/2 ounce, were too small to savor for taste, and didn't have the
crema or body of the Tea's. The Peppina took the second round. I
switched to the double basket and started pumping the lever for the
preinfuse until I could feel some backpressure (per Roger's advice).
This produced a 3/4 ounce shot from 12 grams, a little less
concentrated than the 1 1/4 ounce 20 gram shots from the Tea, but now,
to my utter surprise, with crema equal to the Tea's and a buttery
seductive mouthfeel that was actually better. The flavors were also a
hair better, less rough edges, and slightly better aromatics. I'd give
it maybe 3 to 4 points more out of 100 than my Tea shots. The third
round was a wash, the Tea's shot had better crema and aroma, the
Peppina's better mouthfeel and taste. I pulled this round into glass
cups, and posted the pics for a visual comparison.
CONCLUSION
Even PID'ed, the Peppina isn't the ultimate espresso machine. But in
terms of shot quality, it can hold it's own against most commercial
machines, even though it has an underweight, undersized group and an
operator with a total of 24 hours lever experience. For a $200 machine
with a jury rigged PID added, that's pretty impressive. As proof of
concept for this gravity fed lever design, it's even more impressive.
The lever pump provides a smoother extraction than the rotary, and the
sealed water path (no cold or overheated water entering) utterly
simplifies the control task. The super impressive Versalab machine
also uses a sealed water path and a new style, super smooth pump that
can operate in near boiling water. It's undoubtedly better than my La
Peppina rig, but it also costs 25 times as much. This test shows that
the design features that make it so good can be had at much lower cost
-- a Peppina style lever machine with a beefier group, more accurate
PID, and a well engineered spring or compressed air driven piston
could perhaps do it.
--
Jim
(***@ameritech.net)
Jim
(***@ameritech.net)