is part of the Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Extrusion basics: Pumps and lumps, or life between the screw(s) and die

Gear pump for extrusion
The added cost of gear pumps can be justified by savings of material because aim thickness can be lower. Another benefit is reducing the load on the extruder, which may include pushing the melt through a static mixer to improve appearance, properties and maybe production rate.

There are no toxic plastics. None of them. I’ve been starting my monthly column that way for a while, so I hope you’re already spreading this gospel to your families, friends and neighbors. To do this you have to know it, not just believe in it. The plastophobes believe the opposite; we have science on our side, but that scares people, too.

Demo-scale pump with clear plastic housing. Image courtesy Allan Griff.

Now to extrusion technology. The pumps I’m talking about are gear pumps, devices that pick up the melt as it leaves the extruder after it has gone through screens and push it forward through the rest of the system—adapter and die, and maybe a static mixer (more on that later). 

They are called gear pumps because they consist of two intermeshing gears; typically the width across the gears is the same as the diameter. They turn in a tightly fitting chamber so that the entering melt gets caught in the teeth of the gears and is carried around between gears and chamber walls to the other side. Nothing goes through the nip of the gears. That nip is virtually sealed because one gear drives the other. Controlling the speed gives precise volumetric displacement, so that whatever the variation in entry, the output is constant in axial flow rate. No surging, no pulsing (from extruder sources). That is the main reason for their use. Their added cost can be justified by savings of material because aim thickness can be lower if the precision is better. Another benefit is reducing the load on the extruder, allowing high-resistance dies without high pressure and consequent high heat development in the screw(s).

The industry often refers to them as melt pumps (melt in, melt out) but I prefer the term gear pump, because an extruder is a melt pump, too, even though it’s pellets in, melt out. 

In operation, the inlet pressure to the pump is preset (e.g., 800 psi) and the screw speed in the extruder is controlled to maintain that pressure lest the pump be starved, which would damage the pump as its drive shaft is lubricated by the melt itself.

Buying a gear pump requires some idea of the pressures expected, as they are sold to match these needs: Light, medium or heavy, referring to the expected differentials across the pump and from inside to outside air. They are sized to match expected output, and the maker should therefore provide the free volume of the pump (cubic centimeters per turn for both gears). From this volume, we can calculate the weight per turn if we have a melt density.

That is the theoretical maximum output. In practice it is slightly less, as some melt is recirculating as shaft lubricant, and may be further reduced by backflow over adjacent teeth and along the gear sides if walls are too thin for the pressure differential (clamshelling).

As a rough rule, if you are getting 90% or more of the theoretical (free volume x rpm) that’s very good—go look for another problem to solve. If you are between 80 and 90%, that’s normal, but in the low 80s you aren’t getting the full benefits. If the ratio of production to theoretical is in the 70s, think about rebuilding or repairing, but first, find out why it’s so low. And if you are below 70%, you may be better off taking the pump out now.

Lumps in your product? The gear pump evens out the axial flow, but doesn’t do any mixing. In fact, if it lowers the pressure inside the extruder, the mixing may actually be worse than without the pump! That’s why pump makers also offer static mixers installed after the pump—after, so the pump pushes it through the mixer, not the extruder. Static mixers are tubes with baffled or otherwise twisted innards, so the flowing melt is split and resplit again and again. They are usually horizontal, and typically one extruder size below the diameter (e.g., 3.5 inches for a 4.5-inch extruder). In retrofitting blown film or any line where neither takeoff nor extruder can be moved easily, they can be inserted vertically, changing height but not horizontal length. 

It is useful but not common to have pressure gauges at both entry and exit. You can also calculate this pressure drop if you have good viscosity data for your material as a function of shear rate and melt temperature.

Lastly, if you have a mixing problem, consider altering materials as well as a mechanical remedy like a static mixer or water inside the screw. Changing a filler type or even just its particle size may help, or adding an internal lubricant, or (for mixtures of melts) getting the viscosities of the components in proportion to the proportions. 

If you are going to NPE/ANTEC in Orlando on May 7 to 11 and want to talk with me in person about extrusion or plastophobia (or anything else, for that matter), I’ll be on the Neutrex/Purgex stand (S20035) giving half-hour talks on "The Ten (11) Principles of Extrusion" on Monday and Wednesday at 2 PM and Thursday at 10 AM, and on "Plastics Chemistry for NonChemists" (including my cartoons, one of which is pictured here) on Tuesday and Thursday at 2 PM. You can register for the sessions on the Purgex website.

Allan Griff is a veteran extrusion engineer, starting out in tech service for a major resin supplier, and working on his own now for many years as a consultant, expert witness in law cases and especially as an educator via webinars and seminars, both public and in-house. He wrote the first practical extrusion book back in the 1960s as well as the Plastics Extrusion Operating Manual, updated almost every year, and available in Spanish and French as well as English. Find out more on his website,, or e-mail him at [email protected].

Griff will present a live seminar in Amherst, MA, on May 22. Seminars in your plant are also available. If you can’t attend his live events, he offers a Virtual Seminar, which can be seen at any time, anywhere. E-mail Griff at the address listed above for more information.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.