Peppina Marries MiniGaggia - Page 2

[mogogear]

The group cap shown earlier is a retro 'low-tech' design that uses a simple pliable washer as the one-way valve. Here is a more high-tech version of the group cap which has been tapped with holes into which modern one-way check valves (not shown) would be fitted. We want excellent water flow from the piston chamber down to the dispersion screen so there are multiple check valves to increase the total effective orifice diameter.

image: http://www.aimsdata.com/tim/espresso/beast/BeastGroupCapHighTech444.jpg

Tim,
I have to disagree with you on this one. Only because of my recent diassembly of my Silvia ( now sold---alsa...) After preping my Rancilio Sivia for sale, I took the dispersion screen off etc. The area under the mesh dispersion screen hasis an inverted brass mushroom inserted into the main water orfice ( or golf ball marker looking insert that acts to keep the water from being directed straight into the screen and into the puck surface. This mushroom has grooves eminating from the center and radiating out to the edges. Maybe 6 or 8 grooves.
See if you can get a look at this arrangement because the washer would act as the same type of diverter and move the water out to the edges . Here it would flow back to the area under the washer and fill the entire void above the screen. Giving an even flow.
I most likely have described this poorly , but see if a search of a site like wholelattelove on cleaning tips etc shows this. It may have an impact on your design ideas and then again....maybe not so much:oops:

One is just much more clean in design, less servicable parts. In an area as prone to build up etc , the washer would be very self-cleaning. That my vote!!

[mathias]

timo888 and mogogear,

I take the liberty to propose multiple marriage

One could use:

The "one-way valve" of the All-Clad Presso interesting
The type of heating used on the Gaggia domestic pump machines
Open boiler type used on Peppina and others
Piston/cylinder dimensions from a commersial lever machine
Shower screen from the above commersial lever machine
PF gasket, PF and PF holder from a pea-valve Gaggia domestic machine


....B....................Rod...................B
....B....................Rod...................B
....B....................Rod...................B
....B....................Rod...................B
....B....................Rod...................B
....B....................Rod...................B
..HB....................Rod...................BH
..HB....................Rod...................BH
..HB....................Rod...................BH
..HB....................Rod...................BH
..HB....................Rod...................BH
..HBBBB..............Rod..............BBBBH
..HBBBB..............Rod..............BBBBH
..HBBBB..............Rod..............BBBBH
..HBBBB..............Rod..............BBBBH
..HBBBB..............Rod..............BBBBH
BBBBBB_P__i__s__t__o__n_BBBBBB
BBBBBB_P__i__s__t__o__n_BBBBBB
BBBBBB_P__i__s__t__o__n_BBBBBB
BBBBBB__Shower.. screen__BBBBBB
GGGGG....................................GGGGG


B= Body (outer diameter 80mm at the heater)
H= Heater (used in plastic moulding machines)
G= Gaggia PF holder

The PF holder from the Gaggia needs to be modified a little (remove the center). Then it bolts directly to the "body".

As the piston diamater commes from a commersial lever machine the spring data can be copied.

Indications show that constant flow gives a good result which means the piston could be driven by a screw at constant speed (one stroke/25 sec.). That way somekind of preinfusion would take place as the grounds I believe create less resitance when dry.

The design ended up this way as I believed it makes it easy to make with limited resources (all parts can be made with a lathe and a drilling machine) and easy to clean. Here I stopped as I can't find an easy way to avoid ruler flat temperature.

Edit: I removed an O-ring between the Body and the PF-holder as it is not needed.

[timo888]

Tim,
... insert that acts to keep the water from being directed straight into the screen and into the puck surface...
One is just much more clean in design, less servicable parts. In an area as prone to build up etc , the washer would be very self-cleaning. That my vote!!

The group cap is a self-contained assembly that will screw into place, and so we can have it both ways retro-tech washer or today's check valves. We could sell this machine as a build-it-yourself kit like a barebones PC, offering, for example, different wattages on the heating elements. Marketing could steal some ideas from PC makers. 'Powerful twin 700-Watt heating elements, one for brew water, one for steam.'

I agree, we'd want to have an even distribution of water to the dispersion screen. The array of valves and a properly designed dispersion screen should address your valid concerns about not wanting to disturb the puck surface. The circular array of valves should distribute the water nicely to the dispersion screen. My hunch is that above the 'golfball' in the Silvia there is not an array of egress ports but a single egress port from the pump, and hence the need to break up a stream of water.

As a point of comparison, in the Peppina, there's a canal leading from the piston cylinder to the group head. The canal has a diameter of ~5mm. The canal leads to two (2) small egress holes on the group above the dispersion screen which are 2mm diameter each. So the Peppina's dispersion screen is fed by an effective orifice diameter of only 4mm. The effective orifice diameter would need to be larger for a 58mm screen, but I feel confident that the circular array of check valves would provide ample egress for water from the piston chamber to the group without the kind of concentrated streaming that requires a pre-dispersion before the water hits the dispersion screen.

The centrally placed check valve might need to become a post for the piston head to abut against.

Regards
Timo

[timo888]

Mathias,

In the design I have in mind, the one-way valve assembly (I call it the 'group cap') is a threaded cylinder with a plate which contains the array of egress ports and the valve(s). The group cap connects the bottom of the piston cylinder to the top of the group cylinder, which protrudes into the boiler. The group-cap is a fairly extensible design and it could accommodate other kinds of one-way valves. It's not clear to me where the one-way valve fits in your diagram and I'm not familiar with the All-Clad one-way valve. Could you describe it?

I am certainly a proponent of simplicity. The group should be very simple. It needs a cylinder that projects up into the boiler, and collar to bolt it to the underside of the boiler, and a cylinder on the bottom into which the PF locks. The water will be at the correct temperature and so we don't need anything fancy in the group. No thermosyphons etc etc. I would like to be able to make use of readily available OEM portafilters.

Regards
Timo

[timo888]

Mathias again,
Please bring me up to speed on the Heating device you're talking about. I am not familiar with it. At first glance it strikes me as an unnecessary complication to have a heating device on the piston, but maybe I don't understand it?

In the design that has evolved so far, the piston is inside the boiler within a cylinder that extends from base of boiler (top of group) up to the lid of the boiler. The water in the boiler is heated by either an immersible element (like Peppina) or by an element that sits beneath the boiler (like Arrarex Caravel).

Regards
Timo

[mathias]

Mathias again,
Please bring me up to speed on the Heating device you're talking about. I am not familiar with it. At first glance it strikes me as an unnecessary complication to have a heating device on the piston, but maybe I don't understand it?

You can see one here http://makeashorterlink.com/?B2E91233D

The heater is not on the piston but on the outside of the boiler (B=Body). I tried to show the idea with the "drawing". The boiler and piston cylinder is made in one piece. The PF holder could be made within the piece as well bu to me it seemed a lot easier to take one from an old Gaggia, "cut" out the center and just bolt it to the "Body".

[mathias]

Mathias,

In the design I have in mind, the one-way valve assembly (I call it the 'group cap') is a threaded cylinder with a plate which contains the array of egress ports and the valve(s). The group cap connects the bottom of the piston cylinder to the top of the group cylinder, which protrudes into the boiler. The group-cap is a fairly extensible design and it could accommodate other kinds of one-way valves. It's not clear to me where the one-way valve fits in your diagram and I'm not familiar with the All-Clad one-way valve. Could you describe it?

I am certainly a proponent of simplicity. The group should be very simple. It needs a cylinder that projects up into the boiler, and collar to bolt it to the underside of the boiler, and a cylinder on the bottom into which the PF locks. The water will be at the correct temperature and so we don't need anything fancy in the group. No thermosyphons etc etc. I would like to be able to make use of readily available OEM portafilters.

Regards
Timo

The All-clad can be seen here http://www.home-barista.com/forum...-machine-t210.html

[timo888]

You can see one here http://www.home-barista.com/forum...-machine-t210.html

The heater is not on the piston but on the outside of the boiler (B=Body). I tried to show the idea with the "drawing". The boiler and piston cylinder is made in one piece. The PF holder could be made within the piece as well bu to me it seemed a lot easier to take one from an old Gaggia, "cut" out the center and just bolt it to the "Body".

OK, I think I see what you mean. The piston is immersed in the boiler water? When the piston retracts, does it displace water? Where are the seals? Or is the piston in its own cylinder which is inside the cylindrical boiler?

One of the design goals is to have a minimum of 400ml of water at stable brew temperature. When the 500ml boiler is down to 100ml remaining, fresh (i.e. cold) water is pumped into it from the reservoir -- in the auto-fill model. There's a less expensive manual refill model. Everything is a-la-carte. Barebones Beast.

In the gravitational model I have in mind, the boiler water is immediately above the group. Straight down to the puck. One of the drawbacks of the band heating element you've suggested (at least with the specific band element at the link provided) is that with an 80mm maximum diameter one needs a fairly tall boiler to get 500ml of brew water, especially if the boiler houses a piston/piston rod and perhaps also a piston cylinder.

But if there are band elements of greater than 80mm diameter, that type of element would indeed present an alternative to the immersible element and the Arrarex-style element-below-boiler.

Cylinder's Volume = Pi * Radius squared * height (minus whatever volume is displaced by stuff inside the boiler, piston etc). We're talking 12-14cm or more in height, depending upon what's inside the boiler cylinder that takes up space. Not a tall boiler for a pressurized machine, but for a gravity-based design with the boiler directly above the group, too tall. Remember, it has a lever on top, so height and center of gravity are primary concerns.


It's positioned to be a home machine. I have been aiming for something with the MiniGaggia's form-factor. No too tall and imposing. A home machine with some wife-appeal. My wife thought the Rancilio Silvia was huge and institutional-looking, for example. We don't have the budget for focus groups, so she will have to suffice as Everywoman.

I have to leave for work but will check out the one-way valve of the All-Clad later today.

Regards
Timo

[timo888]

For the spring-style lever, the downstroke compresses the spring and causes the piston to retract. In the design I've set out here, piston would retract upwards. When the piston retracts, we want to prevent water traveling from the puck up into the piston chamber but allow water to flow into the piston chamber from the boiler. I think my group cap with one-way valve(s) between dispersion screen and piston addresses this issue better than the All-Clad piston's rolling gasket design would. Going to stick with the simple and extensible group-cap

Regards
Timo

[mathias]

OK, I think I see what you mean. The piston is immersed in the boiler water? When the piston retracts, does it displace water? Where are the seals? Or is the piston in its own cylinder which is inside the cylindrical boiler?

Yes / very little (if retraction is slow enough) / one O-ring on the piston, see All-Clad (doesn't show on my drawing") / piston cylinder is below cylindrical boiler (same piece of metal though).

One of the design goals is to have a minimum of 400ml of water at stable brew temperature. When the 500ml boiler is down to 100ml remaining, fresh (i.e. cold) water is pumped into it from the reservoir -- in the auto-fill model. There's a less expensive manual refill model. Everything is a-la-carte. Barebones Beast.

My goal was 6-8 doubles (225-300 ml).Manual refill.

In the gravitational model I have in mind, the boiler water is immediately above the group. Straight down to the puck. One of the drawbacks of the band heating element you've suggested (at least with the specific band element at the link provided) is that with an 80mm maximum diameter one needs a fairly tall boiler to get 500ml of brew water, especially if the boiler houses a piston/piston rod and perhaps also a piston cylinder.

Same model. I've seen 100mm band heaters though I was going for the 80mm. 300 ml, inner diameter 70mm gives a length of 128 mm. With a 100 mm heater you could have 90 mm inside diameter. 500 ml would then give a length of 127 mm. The actual length need to compensate for the piston rod (and the spring if it is placed within the boiler)

Remember, it has a lever on top, so height and center of gravity are primary concerns.

With a 47 mm piston diameter ( http://makeashorterlink.com/?O56E4233D ) and 75 ml/pull the stroke would be 43,2 mm (the Gaggia doesn't make a double/pull) . My idea was to have the spring and lever mechanism behind the group. That way the lever axis could be at the base a la Peppina and allow for a longer lever. The total height would then be 220mm (appr.) + PF + cup height + drip tray. I think I ended up around 500 mm. 8,5 bar requires almost 1500N with a 47 mm piston. 150N pull force gives a 1:10 ratio. A 43mm piston stroke then requires a 430mm long lever.

It's positioned to be a home machine. I have been aiming for something with the MiniGaggia's form-factor. No too tall and imposing. A home machine with some wife-appeal. My wife thought the Rancilio Silvia was huge and institutional-looking, for example. We don't have the budget for focus groups, so she will have to suffice as Everywoman.

Women are powerful aren't they
Would this be a machine for the market? Mine was just intended for me, therefore a design that could be made with little shop capacity.

[mathias]

For the spring-style lever, the downstroke compresses the spring and causes the piston to retract. In the design I've set out here, piston would retract upwards. When the piston retracts, we want to prevent water traveling from the puck up into the piston chamber but allow water to flow into the piston chamber from the boiler. I think my group cap with one-way valve(s) between dispersion screen and piston addresses this issue better than the All-Clad piston's rolling gasket design would. Going to stick with the simple and extensible group-cap

Regards
Timo

I understand but when you pull the lever there is no water in PF, just freshly grounded coffee. The Presso one-way valve will prevent air to be sucked through the grounds if the piston is raised no faster then the water can fill the cylinder (gravity, no pressure). If my drawing is not clear enough just look at the Presso. My design is the same but without the legs and lever. The transparent part is made of metal and is higher with a larger inner diameter where it is higher to form a hot water reservoir. Clamp a heater around, voila. I find the Presso solution very clever. A standard O-ring works as the piston sealing and a one-way valve at the same time.

[timo888]

I understand but when you pull the lever there is no water in PF, just freshly grounded coffee. The Presso one-way valve will prevent air to be sucked through the grounds if the piston is raised no faster then the water can fill the cylinder (gravity, no pressure). If my drawing is not clear enough just look at the Presso. My design is the same but without the legs and lever. The transparent part is made of metal and is higher with a larger inner diameter where it is higher to form a hot water reservoir. Clamp a heater around, voila. I find the Presso solution very clever. A standard O-ring works as the piston sealing and a one-way valve at the same time.

But the puck is wet when pulling a double. Also, a certain preinfusion style involves gently pushing some water to the puck and then lifting the lever to fetch more water and then pulling the shot. For these reasons, the one-way valve has to prevent coffee-oil-water from being drawn up into the piston chamber when the piston retracts. Group-cap with one-way valve(s) addresses this issue well. The doppio is a fact of life your design must reckon with. The double-pull (or the pull-and-a-half) also has implications for your piston design, and your piston-to-group interface, as follows.

The challenge for my group cap design is to keep the cap as short in height as possible, so that at its maximum extension the piston is able to push almost all of the water into the puck.

GOOD:

{ E X T E N D E D P I S T O N }
=====================
=====================
group cap with one-way valve(s)
=====================
=====================
D I S P E R S I O N S C R E E N


BAD:

{ E X T E N D E D P I S T O N }
=====================
=====================
group cap
cap
with (water here if valves are tall)
one-way
valve(s)
=====================
=====================
D I S P E R S I O N S C R E E N


We don't want a large amount of water remaining above the puck when the piston is fully extended. The pliable-washer is superior to the modern check valve in this regard. Maybe there is a very short and squat modern one-way check valve, ~10mm long, with an effective orifice of 2mm?

Now, also in regard to having excess water remaining above the puck, the Peppina-type piston which you have proposed does have an advantage over the MiniGaggia-type piston, which sits immediately above the puck, and has undergone the modification I propose, of having a group-cap inserted between dispersion screen and piston head. (But a check-valve 10mm in length with a 2mm effective orifice would greatly reduce the amount of water sitting above the puck after the pull.)

The tradeoff is that the Peppina-type piston makes the group-interface more complicated: there must be a canal drilled connecting piston chamber and group, and so there then must be two sets of one-way valves, 1) a valve to prevent the piston from pushing water back into the boiler and 2) a valve to prevent puck-water from being drawn back into the canal that connects piston-chamber to group-head, above the puck, when the lever is raised after the puck has become wet.

Again, the double-pull complicates the design. The original MiniGaggia design acted as though the double-pull did not happen, and you have probably seen pictures of the inside of the MiniGaggia boiler. Here's one I nicked



Regards
Timo
P.S. MiniGaggia inside dirty boiler photo credit to member "HV" on TooMuchCoffee.com

[timo888]

What I had in mind for the high-tech backflow-prevention valves, inserted into threaded holes in the plate of the group cap, was a slightly modified version of the ones in the picture below.

These valves, for high-purity low-pressure applications, feature no metal components and are zero maintenance. Check against backflow pressure up to 100 psi. A cracking pressure of 1 psi or less. They are available in inlet and outlet configurations.

The male thread of the valve (available from the manufacturer in lengths as short as 8.5mm) would screw into a threaded hole in the group-cap plate. (There would be several of them arranged in a circle.) The Female Port Depth can approach zero because no female thread is needed.

However, the effective orifice of the valve is only .75mm. Six them would thus yield a 4.5mm effective orifice. I think we need a wider effective orifice for a 58mm diameter basket.

So I'd like to find a valve like this, with male thread ~4mm long and nominal female port depth (~1mm) and a larger effective orifice. That would permit a very short group cap, thereby minimizing the amount of unexpressed water sitting above the puck.

Regards
Timo

[mathias]

But the puck is wet when pulling a double. Also, a certain preinfusion style involves gently pushing some water to the puck and then lifting the lever to fetch more water and then pulling the shot. For these reasons, the one-way valve has to prevent coffee-oil-water from being drawn up into the piston chamber when the piston retracts. Group-cap with one-way valve(s) addresses this issue well. The doppio is a fact of life your design must reckon with. The double-pull (or the pull-and-a-half) also has implications for your piston design, and your piston-to-group interface, as follows.


We don't want a large amount of water remaining above the puck when the piston is fully extended. The pliable-washer is superior to the modern check valve in this regard. Maybe there is a very short and squat modern one-way check valve, ~10mm long, with an effective orifice of 2mm?

Again, the double-pull complicates the design. The original MiniGaggia design acted as though the double-pull did not happen, and you have probably seen pictures of the inside of the MiniGaggia boiler.

I agree with you that a second pull complicates the design and in my opinion will not improve shot quality. That is why I limited my machine to only make a double shot per pull (75ml/pull as stated earlier). The machine I had in mind is based on alreay existing and working solutions, nothing new. The Gaggia commersial lever machine doesn't suffer from water above the puck as far as I know. The Presso (with modified piston) has the one-way valve just above the shower screen when pull is finished. Preinfusion I've mentioned earlier in connection with using screw/motor to create the pressure. If instead a spring is used the force can be reduced by simply not let go of the lever completly at the beginning of the pull.

Now, also in regard to having excess water remaining above the puck, the Peppina-type piston which you have proposed does have an advantage over the MiniGaggia-type piston, which sits immediately above the puck, and has undergone the modification I propose, of having a group-cap inserted between dispersion screen and piston head. (But a check-valve 10mm in length with a 2mm effective orifice would greatly reduce the amount of water sitting above the puck after the pull.)

The tradeoff is that the Peppina-type piston makes the group-interface more complicated: there must be a canal drilled connecting piston chamber and group, and so there then must be two sets of one-way valves, 1) a valve to prevent the piston from pushing water back into the boiler and 2) a valve to prevent puck-water from being drawn back into the canal that connects piston-chamber to group-head, above the puck, when the lever is raised after the puck has become wet.

I assume you mean Presso-type piston, not Peppina-type, because that is what I proposed. From my point of view the design is not complicated. There is no need for drilled canal. The boiler water is always in contact with the upper surface of the piston. On the upward stroke of the piston the Presso one-way vavle lets the water enter the piston chamber. If the one-way valve is correctly designed water will enter quick enough so no air will be sucked through the coffe grounds.

[timo888]

I assume you mean Presso-type piston, not Peppina-type, because that is what I proposed. From my point of view the design is not complicated. There is no need for drilled canal. The boiler water is always in contact with the upper surface of the piston. On the upward stroke of the piston the Presso one-way vavle lets the water enter the piston chamber. If the one-way valve is correctly designed water will enter quick enough so no air will be sucked through the coffe grounds.

I was referring to this statement you made:

That way the lever axis could be at the base a la Peppina and allow for a longer lever.

With the Peppina, as the lever is pressed the spring compresses and the piston moves downward. Then water flows from the kettle above into the area vacated by the piston. When the spring expands, driving the piston upwards, the one-way valve between piston-chamber and kettle closes and water is forced out to the group through its one-way valve to the puck. So, I am having difficulty visualizing how in your design "the boiler water is always in contact with the upper surface of the piston". Maybe we are meaning different things by "upper surface" since you are thinking Presso and I was thinking you were thinking Peppina

With the machine I envision, a single pull of the lever will draw a gross quantity of water sufficient to net a single shot. If one wants a double from it, one will have to pull twice. I expect people will be pulling twice and so my dream machine will be designed to prevent backflow from the puck and sucking up air. The group cap accomplishes both goals.

Now, a machine that produces only a double-shot seems rather uncivilized Recent experiences with the tiny basket of the Arrarex Caravel make me think the trend towards the triple-basket is the espresso world's counterpart to the Big Gulp. Do you know this behemoth soft-drink in Europe?

Regards
Timo

[mathias]

With the Peppina, as the lever is pressed the spring compresses and the piston moves downward. Then water flows from the kettle above into the area vacated by the piston. When the spring expands, driving the piston upwards, the one-way valve between piston-chamber and kettle closes and water is forced out to the group through its one-way valve to the puck. So, I am having difficulty visualizing how in your design "the boiler water is always in contact with the upper surface of the piston". Maybe we are meaning different things by "upper surface" since you are thinking Presso and I was thinking you were thinking Peppina

With the machine I envision, a single pull of the lever will draw a gross quantity of water sufficient to net a single shot. If one wants a double from it, one will have to pull twice. I expect people will be pulling twice and so my dream machine will be designed to prevent backflow from the puck and sucking up air. The group cap accomplishes both goals.

Now, a machine that produces only a double-shot seems rather uncivilized Recent experiences with the tiny basket of the Arrarex Caravel make me think the trend towards the triple-basket is the espresso world's counterpart to the Big Gulp. Do you know this behemoth soft-drink in Europe?

Regards
Timo

Talking about designs/function is difficult on internet

It is only the lever that is at the base, the piston and cylinder is just under the tiny boiler. Me uncivilized? Does that have something to do with my swedish origins When I was designing this machine there were not a lot of talk about the Caravel but rather how difficult it was to use a single basket.

To visualize the function of the machine just look at the Presso. Remove the legs, keep the rest but imagine it is in the shape of a cylinder with the same diameter through its lenght, add a small boiler betwen the lever arms and the piston chamber (transparent on the Presso, all metal on my machine) and clamp a heater around it.

[timo888]

Me uncivilized??? Does that have something to do with my swedish origins When I was designing this machine there were not a lot of talk about the Caravel but rather how difficult it was to use a single basket.

No, Mathias, you're quite civilized. You're nearly Norwegian! It's the trend towards the mega-dose that is going in the opposite direction of civilization. The Cremina's single basket and the Peppina's single basket both produce an excellent espresso. (I had a very difficult time with Silvia's shallow single basket, on the other hand.) The espresso that comes from the Caravel's little basket is smooth and sweet.

In Europe they sell delicious little ice-cream cones, ice-cream about the size of a golf-ball. Here in the US, we strive for several baseballs of ice-cream piled on top of each other. It's a circus. Circus Maximus.
Regards
Timo

[timo888]

Since we're on the subject of piston, let's take a look at the MiniGaggia piston :

--Photo credit to HV from TooMuchCoffee.Com

The MiniGaggia piston would appear to hang down like a stalactite from the boiler lid and the spring would appear to be immersed in the boiler water (I haven't seen one of these machines in person yet.) A single o-ring seal is located at the head of the piston.

This design could be improved first with a cylindrical piston housing that extends from the boiler lid (where it fits into a circular collar on the lid) down to the group cap into which it is threaded. The group cap is threaded onto the top of the group. This housing would keep the spring and piston shaft out of the water where calcification might eventually interfere with its smooth operation; and it would serve to reinforce the boiler structure.

The piston head should be made to be longer than the MiniGaggia's and another o-ring should be placed on the piston head such that o-ring1 is located above the ingress ports when the spring is fully extended and o-ring2 is above the ingress ports when the spring is fully compressed:


When the piston reaches its fully retracted position, o-ring2 is above the ingress ports and water flows into the chamber. Thus, water enters the piston chamber only when the lever is fully cocked.

Regards
Timo

[mathias]

Timo, do you know how to design the o-ring seating (diameter and tolerances)? I never found information on that.

[timo888]

The drawing I just posted exaggerates the o-rings; they would have a much more modest profile on the real piston, and would be almost flush with the outer wall of the piston shaft. Judging from diameters on the o-rings on the Peppina, the Caravel, and the Cremina, I think there's a range of acceptable configurations. Seat-of-the-pants estimate: tube diameter of the o-ring would be 5mm.

I am still leaning towards a smaller basket. Given the gravity-fed boiler-integral-with-group design, a 58mm group is not necessary for heat stability. The Peppina's featherweight small-diameter portafilter produces excellent espresso. I'd also like the machine to be capable of producing a good single shot. A single shot is hard to do when the basket has a 58mm diameter!

The diameter of Peppina's piston is 51mm. The Pavoni portafilters and baskets are fairly easy to come by. So maybe this machine would be best served by a 51mm group, and there'd be a ready supply of tampers and replacement baskets and portafilters. I don't suppose there's a patent in effect that would prevent a group from accepting Pavoni portafilters?

Regards
Timo