Why difference in pressure between blind filter and brewing? - Page 2

Need help with equipment usage or want to share your latest discovery?
User avatar
Dalla Corte USA

#11: Post by Dalla Corte USA »

Wow! I never knew an OPV could be so exciting! Over the weekend I'll shoot some videos of the two valves in question and post the results on youtube. This should clear up any questions of the results and I hope create some clarity.

User avatar
Supporter ★

#12: Post by erics »

Dalla Corte USA wrote:Wow! I never knew an OPV could be so exciting!
That little 3/16" glass ball that serves as the valve "head" in the F-O-T OPV packs a wallop. :) Here's some pump capacity curves to liven things up even more:

As an additional data point, I brought my Silvia out of retirement :) because she is equipped with the F-O-T OPV valve, a fairly accurate brew pressure gage, and the stock Ulka EX pump. A blind filter produced 146 psig (~10.1 bar) and a run with the Scace Thermofilter produced 129 psig (~8.9 bar) for a differential of 1.2 bar.

edit 5/7/09 - changed value in FOT curve @ 9 bar

Eric S.
E-mail: erics at rcn dot com

User avatar

#13: Post by dsc »

Hi guys,

remember that vibe pumps vibrate (as the name indicates) which makes stuff even more complicated, or at least can make it more complicated.

Just for reference the PT on my group in a rotary powered machine shows 8.65bar with a blind filter and 8.5bar during extraction. Obvious that some water gets absorbed by the coffee and it actually goes through it instead of hitting a brick wall in the form of a rubber basket insert.


User avatar
Team HB

#14: Post by another_jim »

It's easier thinking about the imperfections of OPVs in terms of flow, rather than pressure:

-- Suppose that to maintain 9 bar, a pump has to flow at 150 mL/min
-- Call the flow through the puck Fp, and the bypassed flow Fb. Whenever the flow through the puck is less than 150 mL/min, the flow through the bypass has to exactly make it up to maintain 9 bar. In other words, Fp + Fb = 150 mL/min
-- This would mean that in the "ideal" OPV, the flow declines in a straight line as the puck goes from choke to a 30 second, 50 mL or so double (25 mL for the puck). It would also mean that in the ideal OPV, the thing would inject water and make up the lost pressure if the flow gets faster than 150 mL/min.
-- Real world objects don't have discontinuities or negative flows, so Fb goes towards zero asymptotically. This means the total flow will always be too high for 9 bar as the flow through the puck increases. A good OPV has a nearly straight Fb flow line, with the asymptote close and tight at the bottom. A poor OPV has a curved Fb line all the way from Fb max (blind filter) to Fb zero.

The engineering trade off is simple. A long looser spring with a very curved Fb response is simpler to adjust and can be adjusted very precisely, while a short hard spring with a very linear Fb response is very difficult to adjust, and can be adjusted only crudely. The FOT OPV would be fine if everyone had a Scace set to their favorite espresso flow rate to use for adjusting it. But if I were adjusting against a blind filter, I'd like one of the nasty, hard to adjust, tight springed brass ones.
Jim Schulman

User avatar
Dalla Corte USA

#15: Post by Dalla Corte USA »

As you can see in the videos the Fluid-o-Tech pump and OPV have slightly better performance than the Ulka and commercial OPV. The method here is crude but I think it shows the amount of pressure change from flow on the Scace device to static. I will build a variable setup when time allows.

User avatar

#16: Post by gyro »

acquavivaespresso wrote:think of the poor guy who buys a DC Mini and sets 2 bar higher (with terrible brew results) because he read Marshall reporting that Mr Warren stated .....
Hmm, seems this is settled then :wink:

Nice clips, thanks James.


#17: Post by Billc »

At the risk of starting more arguments I have this to offer:

Pressure is dependant upon flow rate. For example, take a garden hose. When on, water flows freely out of the end. The pressure when it exits is atmospheric pressure or 0 gauge pressure. Now there is still water coming out. The pressure behind this is the pressure that the water company provides. This usually remains pretty constant. So in the hose somewhere there is the water supply pressure and it reduces to 0 after it exits (the pressure gradient inside the hose is much more complicated so we stick to the end conditions that are easier). Now as you put your thumb over the outlet there is a decrease in flow the more you close off the outlet (reducing the flow) the harder it is to hold. This is because you are creating a condition where the water supply pressure is getting closer to the end of the hose. If you close it off completely, there is the water supply pressure directly behind your thumb (or very very close). So the amount of pressure directly behind your thumb is directly proportional to how much flow you let out. i.e. as you let more water flow the pressure directly behind your thumb is reduced, but the actual water supply pressure remains the same.

If you relate to the espresso machine, the pump provides the water supply pressure. The coffee is your thumb and the water system is the hose. If the coffee is loosely packed water flow more freely and the pressure directly behind the coffee is less. If the coffee is packed to tightly so no water passes, the pressure behind the coffee is the water supply pressure (or close to it there are some inefficiencies due to orifice pressure drop and such).

Most Manufactures measure pressure in a place that it remains constant, like in the boiler or just after the pump. This gives the illusion that the pressure is remaining constant at the group. However if you measure at the group while water is flowing then you will get a different measurement. Now, using a device at the group head while water is flowing is tricky. You must estimate the flow rate of the coffee exactly to get the correct measurement. If the water flows too fast then you will get a lower measurement than if the water is flowing slow. It is very hard to simulate the dynamic flow rate of the coffee because it is not constant. At first you have quite a bit of water rushing into the basket area filing up and displacing all of the air. Then some of the water soaks into the coffee and the coffee expands. When this happens the coffee then acts as a restrictor to the water and lowers the flow rate.

So if you are measuring pressure at the group head with a portafilter device and then comparing results it is important that each person measuring has the exact flow rate. Just a slightly different flow rate will change the pressure significantly (in the case of espresso machines).

The OPV is very similar. Normally it is closed. As the pressure rises to the set point of the OPV the pressure then exerts enough force on the OPV so that a small flow of water is started. The distance the rubber piece in the OPV moves is almost not noticeable. As soon as flow starts the pressure in the system drops (more flow = less pressure). The rubber piece does not care what force is behind it, spring, thumb, someones screwdriver.... you get the point. Essentially is is just a constant force. The rubber in the OPV does not move enough (less than.15mm .005in.) to noticeably increase the force exerted by the spring (it does increase... F=kx but not noticeably).

Bill C


#18: Post by double_pedro »

Let me see if I have things straight....Let's suppose you have the OPV set at some point, P_opv - i.e., this is the pressure (determined by the spring's kx and the cross-sectional area of the plunger) at which the OPV will just crack open and begin to allow some flow in order to limit pressure. This by-pass flow will cause the pressure to drop but I suppose exactly how much depends on the design of the valve. Now, if I slap-on a blind PF and crank up the pump (assume the output pressure is well above P_opv), the pressure I measure will be the maximum brew pressure possible and may well be higher than P_opv because the flow through the OPV is limited, right?

If so, isn't the point of all of this simply that the blind PF pressure measurement establishes is the maxuimum brew pressure but can't say anything further about the actual brew pressure as this depends on the grind, tamp, dose?

User avatar
Supporter ❤

#19: Post by cafeIKE »

yes and any restrictors in the brew path


#20: Post by Billc »

Exactly! There is only 1 or 2 machines that actually show you the brew pressure (Synesso had one at coffee fest) and the other one will be released in 6 months or so.

This is why the pressure thing is so complicated.