Just stumbled on this thread and read the discussion on misconceptions about pump types. If anyone is interested, here is my spin on this subject. Some of this restates what was already clarified by others, but hopefully I added some new information and pulled it all together. BTW, if anyone disagrees about any of this, please let me know. I'm always happy to discuss and debate, and maybe even learn something new.

First some general pump terminology. Positive displacement (PD) pumps (also called volumetric pumps) work by repeatedly trapping a volume of liquid and moving it through the pump. Reciprocating PD pumps do this by moving a shaft back and forth in the axial direction. Rotary PD pumps do this by spinning a shaft.

With that out of the way, here is some specific terminology used by espresso machine manufacturers (and vendors) and what it really means.

1. VIBRATION PUMP: This is a piston pump that use solenoids to drive the piston. It is a type of reciprocating PD pump.

2. ROTARY (VANE) PUMP: This is actually a sliding vane pump, which is a type of rotary vane pump, which in turn is a type of rotary PD pump.

3. GEAR PUMP: This type of pump uses a set of gears to move the liquid through the pump. Gear pumps are a type of rotary PD pump.

Now on to some common misconceptions.

1. Espresso manufacturers (or at least the vendors) refer to sliding vane pumps as simply "rotary pumps". I believe they are using "rotary pump" as short for "rotary vane pump", which can lead to confusion because "rotary pump" is a general category of pumps that includes vane pumps and gear pumps (among others).

2. There are pumps (such as centrifugal pumps) that use rotating shafts but are not "rotary pumps". I think this (incorrect) association between centrifugal pumps and the word "rotary" may be the source of the misconception that "rotary pumps" used in espresso machines are not PD pumps.

3. Variable speed control can be used to vary flowrate for all three types of pumps commonly used in espresso machines. For vibration pumps, look no further than Decent who varies the solenoid frequency to control the flowrate. For rotary vane pumps and gear pumps, a variable speed dc motor can be used on either (as is evident looking through pump-motor combinations on the Fluid-O-Tech website).

When manufacturers select which type of pump to use, they consider many factors including: cost, size (along with motor and starter capacitor), lifespan, sound level, variable control, etc. For cheaper machines, vibration pumps are an obvious choice because they are cheap and very small. For high end home espresso machines, rotary (vane) pumps are good because they are more quiet than vibration pumps and they are less expensive than gear pumps and generally maintenance free. For commercial machines (especially for variable speed control application) gear pumps are commonly selected. To be honest, I'm not sure of all the reasons why.

If you're interested in a simple reference on types of PD pumps see this webpage on volumetric pumps from Fluid-O-Tech, whose pumps are often used by Italian espresso machine manufacturers.

What makes a pump a positive displacement type is that it moves the fluid by mechanically displacing a volume of liquid. This volume can be constant or variable.

In general industry it's more common to find constant volume pumps, and perhaps that creates a misconception that PD pumps must be constant volume. But since a variable volume pump moves the fluid by mechanically displacing the fluid, it's still considered to be a positive displacement pump.

Case in point, Fluid-O-Tech's Solenoid Pump webpage specifically calls their solenoid pump to be positive displacement type, yet this solenoid pump uses a spring to vary the stroke length based on discharge pressure, making this pump curve.

In other applications, pump manufacturers may use more elaborate means to control the stroke length to get a different pump curve.

Edit to add: Also, this Wikipedia page describes several types of variable displacement technologies, including the use of a spring in a piston pump to vary the displacement in proportion to discharge pressure. (However that Wikipedia page has some mistakes, so take it with a grain of salt.)

I'll just add that the gear pumps displace a known amount of water every rotation based on the gear size, BUT this changes with pressure, soooo it's really only useful to measure flowrate via tachometer readings of the pump up until any pressure occurrs, then you would need a compensation table or a separate flowmeter. You can view the graph on fluid-o-tech's data sheet for their pumps.

They also state that they are "pulsation free" which is very neat. I suppose they mean "nearly pulsation free" since there are tiny gears, but it would make sense that they are the most pulsation free of any common espresso machine pump. Whether this matter in espresso is another issue. They also say "noiseless" which isn't true, haha.

Yes, good point. This reduction in flowrate is due to "slip" which increases with discharge pressure. It's inherent to all PD pumps (as far as I know) but to varying degrees. Gear pumps appear to have more slip than sliding vane pumps, for example.

I hear what you're saying about "pulsation free" and "noiseless". All manufacturers have sales departments who write that kind of stuff. The purchaser's engineer will completely ignore that and go straight to the data sheet, which (not surprisingly) says nothing about pulsation and noise. But a good engineer will already know that gear pumps (and rotary vane pumps for that matter) are low enough pulsation that they won't cause vibration issues and will be relatively quiet compared to reciprocating pumps. If ultra quiet operation is required, the purchaser's engineer will specify the sound level specification in the purchase request and get the manufacturer to measure and confirm they meet that spec.

On the subject of pulsation, certain applications of reciprocating pumps require a pulsation dampener to manage the vibration. But in espresso machines, I theorize that there is some inherent pulsation dampening due to the small amount of air that inevitably gets trapped when the grouphead is flooded.

bostonbuzz wrote:I'll just add that the gear pumps displace a known amount of water every rotation based on the gear size, BUT this changes with pressure, soooo it's really only useful to measure flowrate via tachometer readings of the pump up until any pressure occurrs, then you would need a compensation table or a separate flowmeter. You can view the graph on fluid-o-tech's data sheet for their pumps.

After my earlier reply, I re-read your post and now I'm unsure why you would want to know the exact pump flowrate vs pressure.

Here's where my mind is at. Assuming a constant speed motor and a rotary vane pump, the typical design is to use the pump's internal relief valve as a pressure regulator set at roughly 9 bar. The flowrate is controlled by adjusting the puck resistance through grind and dose.

For a gear pump, I would expect the same thing except that an external valve will be required to regulate the pressure by recycling water back to the pump suction line. As far as I have seen, gear pump manufacturers don't seem to offer optional internal relief valves, probably due to the layout of pump internals.

And if a variable speed motor is used for flow control, it would probably be a good idea to measure the flow since slip will inevitably increase with valve wear.

Let me know. I would be interested to know if I am missing something.

La Marzocco sets the pump pressure on their gear pump machines using a pressure transducer at the puck and controlling the pump speed on the fly per their manual, below. There is no recirculation although Fluid-o-tech references some gear pumps coming with relief valves.

When the STRADA follows a set profile, the CPU reads a pressure that is very close to that in the coffee puck, therefore it is very susceptible to minor changes in grinding, dosing and tamping. If, for instance, the operator tamps slightly weaker, or packs in the basket a smaller quantity of ground coffee, the water will encounter less resistance in its path, therefore the pump will have to spin faster (consequently delivering more water) in order to reach the set pressure.

It seems like a plain old OPV/expansion valve would work for a gear pump to limit the max pressure for safety reasons.

My application is pressure profiling. Set the pump max speed so it can hit your highest desired pressure, say 9 bars on the fastest shot you would want to pull. There you have a useable range. Adjust it as you like during a shot with a potentiometer.

And if a variable speed motor is used for flow control, it would probably be a good idea to measure the flow since slip will inevitably increase with valve wear.

Yes, you're right.

Here is a DIY approach example of profiling with a gear pump (not me).

So you're going to DIY a speed control? Very cool!

Depending on the variable speed motor you're using, a standard OPV may not have sufficient flow capacity for an overspeed condition, and you could reach dangerous pressures if things go wrong. It would be best if you could use a motor with a max speed that is only slightly higher than you think you might need. Otherwise you will need to find a non-standard solution to overpressure protection.

Just because, gear pumps do not slip. They leak due to the clearances necessary to allow operation. In theory they leak very slowly, slowly enough that it can be ignored for our needs. The clearances may increase over time but probably not enough it matters. But, if you're running very slow profiles, mostly for pour over or brewed coffee, it may become an issue. I use a profile that is set to flow .5ml/second for 90 seconds and when I first installed the profile it was set to .1ml/second.

Hi. I think part of your first sentence got lost, but if I understand correctly you are saying that "slip" and "leakages due to the clearances" are two different things. Perhaps I am misunderstanding you, but it seems that you are assuming that "slip" is referring to slippage between the gear and the shaft?

Slip is an industry term that is defined as leakage due to clearances. You can see lots of results confirming this with a Google search for "gear pump slip". For example, this primer from a pump manufacturer defines slip as follows. "The Slip is the quantity of fluid which leaks through internal clearances of a pump per unit of time. It is dependent upon the internal clearances, the differential pressure, the characteristics of the fluid handled, and, in some cases, on the speed."

This slip results in noticeably lower flow rates when pumping water at higher pressures. For example, see this data sheet for a Fluid-O-Tech gear pump, noting that water would be curve "A".

I agree that for home use, wear is probably not an issue. But even a brand new pump is going to have a flow rate that noticeably varies with discharge pressure.