Pressure Profiling and Flow Profiling - Page 2

Need help with equipment usage or want to share your latest discovery?
User avatar
CarefreeBuzzBuzz
Supporter

#11: Post by CarefreeBuzzBuzz »

AssafL wrote:
BTW - EP and FP are why I am waiting with enthusiasm for the Decent machine to make its way to owners. I really want to see if indeed and which EP and FP profiles make sense.
What is the FP feature of the decent espresso machine. How will it compare to the GS3 (current version ) and Slayer?
CarefreeBuzzBuzz
___
Artisan Quick Start Guide
http://bit.ly/ArtisanQuickStart

User avatar
AssafL

#12: Post by AssafL »

Flow profiling uses a propeller to measure and control flow.

Obviously both flow and pressure depend on the ability of the pump and puck so only work within a compliance range.
Scraping away (slowly) at the tyranny of biases and dogma.

grassroothuihui

#13: Post by grassroothuihui »

There is a youtube video http://www.youtube.com/watch?v=Ud7a1znoo2w
The screen is showing pressure when the water is passing through an empty portafilter.
Does that mean it is measuring the pump pressure?

grassroothuihui

#14: Post by grassroothuihui »

Also, for FP and PP, there should be different inside the coffee punk.

From surface of the punk to the bottom of the punk, during "pre-infuse"
The pressure dropped from 9 bar to air pressure(1bar) in FP
The pressure dropped from 3 bar to air pressure in PP

The flowrate difference caused pressure drop rate to be different, given the resistance created by the coffee punk.

When we look at the drops from the portafilter, they look similar for both FP and PP, but what happens inside is different.

Which one is better? Maybe only taste can tell?

zfeldman

#15: Post by zfeldman »

grassroothuihui wrote:There is a youtube video http://www.youtube.com/watch?v=Ud7a1znoo2w
The screen is showing pressure when the water is passing through an empty portafilter.
Does that mean it is measuring the pump pressure?
Is it possible that readout is just indicating the position of the lever? Move the lever to 7 bar and pump cranks up to a pre-calibrated level, but nothing is actually "measuring" 7 bar?

ShelbiRyan

#16: Post by ShelbiRyan »

AssafL wrote:What we don't know is whether it measures and profiles the pressure or the pump voltage. It isn't the same. For Strada EP it would need a pressure sensor. For MP nah.
If per say the R9 does measure pressures and profiles from pump voltage, would you say it could be in accurate? Or less repeatable? You seem to have spent a lot of time modifying your GS3 and I would like to know more of your thoughts on that subject. I have also noticed the R9 was mentioned to have a flow meter, I wonder if or how this could possibly make pressure profiles more repeatable or if it has anything to do with it at all. I would assume it could be beneficial if there is an auto shot shut off feature. I would also be interested to know if the pressure profiles work off of flow rate or time. I would assume it would be time, but thats just speculation. I don't have a lot of experience with pressure or flow profiling machine and the R9 has caught my eye for possibly my next machine. I have asked some dealers questions about the R9 but not many know much more then the previous attached article and the couple of videos on YouTube.

User avatar
Jake_G
Team HB

#17: Post by Jake_G »

grassroothuihui wrote:There is a youtube video http://www.youtube.com/watch?v=Ud7a1znoo2w
The screen is showing pressure when the water is passing through an empty portafilter.
Does that mean it is measuring the pump pressure?
Most machines (even "profiling" machines except Slayer and conical valve LM and a few others) have the gauge installed before the gicleur for convenience of plumbing. This isn't ideal, but not that big of a deal as profiling the pressure ahead of the gicleur results in a directionally similar profile in the puck.

Of interesting note, and touched on by a few in this topic, is that pressure and flow are completely interrelated in a given shot. What you choose to attempt to control tends to dictate the nomenclature. One of the reasons I am also excited to see Decent Espresso get off the ground and into shops and homes is that they understand this relationship implicitly. Their modes of operation are "pressure priority" and "flow priority" since you cannot control one without changing the other. In general though, a pressure profiling machine measures pressure and tries to control pump output to hit a user-influenced (if not user-defined) pressure profile, while a flow profiling machine does the same thing, but in an effort to follow a flow profile. Any control methodology can be used for either, the question is what you are measuring and trying to control...

How interrelated are pressure and flow? Even test results from Assaf using the FLB to restrict the flow on Chimera for the entire duration of the shot resulted in skewed data from the internal flow meter because the tiny restriction changed the dynamics within the flow meter, even though accuracy is totally maintained when simply slowing the pump speed to achieve the same flow rate/brew boiler pressure. Why? Because this stuff gets tricky and weird things happen when you get into the nitty gritty...

The long and short of it is that if you want to control to a constant pressure, you will have an increasing flow rate as the coffee puck erodes. Likewise, if you want a control to a constant flow, you will have a decreasing pressure ramp as the coffee puck erodes. You cannot attain constant flow with constant pressure, and controlling either pressure or flow to a certain time-based and dynamic profile will result in a rather complex (and likely inverted) profile of the other.

All that really matters at the end of the day is what's in the cup and the ways to get to your subjective preference are increasing daily at a seemingly exponential rate. While hardware and control loops may be better or worse among the plethora of profiling machines available today, I don't think one is heads and shoulders above the rest; it's all down to personal preference of taste, aesthetics, ease of use and the customer focus of the brand. That said, I would be very surprised if the R9 used an open control loop, where the paddle just changes the voltage to the pump. They have the hardware in place to have a sophisticated closed loop control system that continuously adjusts the pump output (likely PWM or % duty cycle) to hit the pressure target, which is adjusted by the paddle...

Just my $.02 :P

Cheers!

- Jake

User avatar
AssafL

#18: Post by AssafL »

It would be somewhat hard to get 7 bar when there is no coffee in the puck. 9 bar would be even harder but I don't know if the 9 bar was actually shown on the screen or if Andrew merely read out what the profile said - and the machine did its best to try to keep up (if the latter, it means they would use a pressure transducer).

Obviously, the gicleur does allow pressure buildup inside the boiler, but the flow needed to maintain 9 bar is usually too high for standard pumps (and thus lacks "compliance"). Obviously Rocket may have chosen a tiny gicleur (<0.5mm) that could induce a high pressure drop or a very high throughput pump, but the former is prone to clogs while the latter is expensive (and unnecessary).

Jake - open loop is possible - the Slayer doesn't have a pressure transducer in it. It is completely open loop (control wise; it really isn't open loop: the coffee puck self regulates the flow...)... Also, all the MP are open loop with the human closing the loop.

Cool machine. Lots of new stuff coming out. I do hope they did their homework and their rubber surface turns out to be of good quality. I don't want to imagine what happens if the plastic turns out to deteriorate, release the plasticiser and becomes sticky. I try to avoid soft plastic textures because of the sticky plastic situation.
Scraping away (slowly) at the tyranny of biases and dogma.

User avatar
Jake_G
Team HB

#19: Post by Jake_G »

AssafL wrote:It would be somewhat hard to get 7 bar when there is no coffee in the puck. 9 bar would be even harder but I don't know if the 9 bar was actually shown on the screen or if Andrew merely read out what the profile said - and the machine did its best to try to keep up (if the latter, it means they would use a pressure transducer).

Obviously, the gicleur does allow pressure buildup inside the boiler, but the flow needed to maintain 9 bar is usually too high for standard pumps (and thus lacks "compliance"). Obviously Rocket may have chosen a tiny gicleur (<0.5mm) that could induce a high pressure drop or a very high throughput pump, but the former is prone to clogs while the latter is expensive (and unnecessary).
I think you're absolutely correct here, but with my plumbed rotary at 200 LPH, I can (and do) bounce off the bypass at 9 bar when flushing the group with no gliceur at all... no doubt at all that a tank-fed gear pump would not do this, but it is totally possible and not at all "hard" to do. :wink:
AssafL wrote: Jake - open loop is possible - the Slayer doesn't have a pressure transducer in it. It is completely open loop (control wise; it really isn't open loop: the coffee puck self regulates the flow...)... Also, all the MP are open loop with the human closing the loop.
Again I agree wholeheartedly but I'm not sure that Slayer - in all its glory - even qualifies as a control loop at all. The pump turns on and a solenoid is either in position "a" or position "b". The human control loop is curiously one of the most (if not unequivocally THE MOST) sophisticated control systems available. Not nearly as fast of a reaction time as a PID, but we do unexpected things, and that leads to discovery :P .


My point in commentary is not so much that open loop is bad, but that at a certain price point and for a machine touting technology, I would be bit surprised if the R9 didn't have a pressure transducer and a closed loop control scheme but not necessarily disappointed, if that makes sense...

I don't think one will be finding it's way into my home any time soon, but I wouldn't mind if one did, that's for sure. Unless the plastic got sticky. That would be an unpleasant thing for sure.

User avatar
BaristaBoy E61

#20: Post by BaristaBoy E61 »

Jake_G wrote:The human control loop is curiously one of the most (if not unequivocally THE MOST) sophisticated control systems available. Not nearly as fast of a reaction time as a PID, but we do unexpected things, and that leads to discovery :P .
Brilliant - This is why I thoroughly enjoy reading what you guys have to say!

SR
"You didn't buy an Espresso Machine - You bought a Chemistry Set!"