Ideally a Non-return valve’s function is to close after the screw plasticizes a selected amount of resin (shot size) so that this plasticized resin can be forced (injected) into the mold. The intent of this paper is to describe how the most widely used non-return valve types work and show the advantages of the All Purpose Valve in operation and in lab trials conducted at Van Dorn Demag.

Non-return valve design originated with the ball check (Fig.1 ) this valve type has been and is still used in many Newtonian fluids and applications. Closure in this valve design is accomplished by reverse flow during injection, created by forward movement of the screw and valve, that moves the ball into a seat that is large enough to accept the recovery flow but small enough to not allow the ball to pass through. This design seals by contact of the ball OD with the receiving surface of the retaining seat. As there is minimal contact surface, any loss of contact due to misalignment or contamination results in leakage. With Newtonian fluids this is less of a problem than with Non-Newtonian fluids with low viscosities and high elasticity. Example 1b - - describes the flow that occurs with a minimal opening of 0.002 in an annulus orifice similar to a ball check seal area.

With this example at a hold pressure of 1033 psi (10:1 inject cylinder to barrel id ratio) leakage of 22.25 pph (0.089 oz/sec) is established. The intent of this example is to show that very small clearances due to incomplete sealing can cause significant cushion penetration.

Proper flow paths through valves are important as any "dead spots" that allow for material hang-up can cause degradation and color streaking. In ball check valves due to the necessary configuration it is difficult to eliminate "dead spots".

The second valve type that is widely used in the Injection molding industry is the ring valve. There are two main construction types, 3- piece and 4- piece. The main difference is in the construction that in the 4- piece allows for a replaceable downstream retainer.

Closure is accomplished primarily by a minimal clearance from the ring OD to the barrel ID. This tight fit creates friction that is attempting to hold the ring in place while the screw rotates and is forced rearward, as the screw develops the necessary pressure to overcome the backpressure resistance.

Wear is created by the friction on the upstream side of the retainer and the downstream side of the ring and also the barrel ID and the ring OD. Another version of ring type valve incorporates an interlocking between the upstream side of the retainer and the downstream side of the ring. This design eliminates frictional wear on the retainer, as the ring must rotate with the screw. This design of course does not help barrel wear and is more prone to breakage as the ring is now required to rotate with the screw as do ball check valves. As the ring is required to rotate with the screw barrel wear is increased, as a minimum ring to barrel clearance is still required to close the valve.

As the barrel wears ring valve types loose the friction required for sealing and become erratic. This is seen in increased cushion variation. To help stabilize this situation more and more decompression (pullback) is used. This helps, as the wear is less as the valve is moved rearward in the barrel.

Velocity is also important to the closure of this type of ring valve and ball check valves. The greater the velocity the greater the initial sealing force. At slow velocities the separating force created by the resin leaking upstream and the increasing pressure on the upstream side of the ring can be great enough to resist complete closure and cause cushion variation or non-closure.

Decompression was intended to prevent mold drool. Normal ring valves require decompression or pullback to close consistently. Greater and greater distances are required as the barrel wears to maintain the normal ring to barrel clearance. In many cases splay is created on the molded part due to excessive pullback. Decompression also assists closure in new barrel and valve situations. This is due to the rearward movement of the screw creating a negative pressure by attempting to pull back the resin downstream of the screw. As the screw flights create a helical component to the resin flow path the decompression is greatest at the area from the end of the flight to the ring. This decompression or reduced pressure creates less resistance to closure as the screw moves forward due to reduced pressure on the upstream side of the ring.

In hydraulics "spool valves" ( Fig. 2 ) are widely used. The closure is a sliding closure where flow is attained by aligning entry and exit openings and closure is attained by sliding either the entry or exit opening to a misaligned position. This type of closure is very positive and if designed properly will not leak.

The All Purpose Valve initially utilizes sliding closure. This sliding closure allows for minimal distance to close. The distance to close determines the time to close, or less distance to close means less time required to close. Distance to close and time to close also are major factors in the amount of leakage prior to closure. Also faster closure means less chance for cushion variation and therefore reduced part weight variation. The APV seals as the ring slides over a circumferential groove, continues to seal as the ring continues upstream over a rear land, and has final sealing as the ring encounters the rear retainer.

In "dead head tests", that is, placement of a plate in front of the sprue bushing that accepts the nozzle tip and seals off flow during injection, with 25 MF propylene the APV forward movement was "0" for 20 sec at 20,000 psi. In this test the ring to barrel clearance was 0.001 on all rings tested. In the 3 & 4 piece valves tested the cushion was continuously penetrated displaying continuous internal leakage.

Ring valves (3 or 4 piece) cannot reduce the distance to close without increasing recovery time and increasing melt temperature. This is due to the increased resistance or pressure drop. Ring valves must have a sealing surface that is not point-to-point contact to minimize leakage during injection. The ring and rear retainer must be properly aligned to seal and the ring thickness must be adequate to resist the inject pressure. If attempts are made to reduce the ring thickness the "hoop strength" can be exceeded causing ring breakage. This thickness and the clearance from ring ID to body OD are the lengths in the pressure drop equation ( Fig.3). The APV design has corners with little or no land length. Therefore the pressure drop is less with less distance to close in the APV. (Example)

APV distance to close (Fig. 4 ), normally .040, provides another important advantage; dispersive mixing. Dispersive mixing consists of an orifice with a clearance that is less than the particle or agglomerate that is attempting to pass through. As the agglomerate is attempting to go through the orifice shear stress is applied deforming it and conductive heat transfer is applied from the metal surfaces and the adjacent melt as it flows by. This and the laminar flow melt the agglomerate and bring the temperature to unity with the adjacent melt.

The APV also incorporates distributive mixing due to the communication of the circumferential groove with the longitudinal grooves. Distributive mixing is dividing of flow and agitation of flow. This is accomplished without "dead spots" and with flow that is self-cleaning. Color dispersion is improved and degradation is eliminated.

As explained earlier wear in a ring valve is created by the minimal barrel ID to ring OD holding the ring in place in the barrel while the screw rotates and rearward screw movement is occurring during recovery. This minimal clearance also causes barrel ID and ring OD wear. The APV normally uses greater barrel ID to Ring OD clearances therefore reducing wear in these areas. The reduced distance to close and the elimination of the force required for sealing on the ring and rear retainer allow for the greater barrel ID to ring OD clearances while accomplishing less closure leakage, less component wear, improved mixing, and therefore less part weight variation.

The trials by Mr. Blevins were conducted on 2/17/98 - 2/19/98 at the Van Dorn Demag lab in Strongsville, Ohio. The tests were all with fifty parts; the machine setup was identical for all tests. The valve size was 50mm with the Zieger (4-piece) valves OD’s 1.9668 (barrel ID 1.968) leaving barrel ID to ring OD clearance of 0.0012. The Dray APV barrel ID to ring OD clearance was 0.006/0.007.

The Zieger valves rear seats were changed to provide different distances to close, these distances are shown at the top of the graphs. The Dray APV valves had different ring lengths and these are also shown at the top of the graphs.

In these trials the APV part weight variation was approximately half of the Zieger valve. Short stroking the Zieger valve to 0.059 did not improve the performance, as the seat type closure requirements are the same. It may be noted that the APV performance was similar in all distances to close due to the sliding closure.

This is a separate test with the standard Zieger valve, the difference from max. to min. was 0.920.

This test (d3) provided the best results although all of the APV tests were similar. The results are; Min. 100.67, Max. 100.88, Diff. 0.21

This test (d2) provided virtually the same results as test (3). The results are; Min. 100.62, Max. 100.86, Diff. 0.24.

This test (d1) provided virtually the same results as test (2). The results are; Min. 100.68, Max. 100.92, Diff. 0.24.

Mr. Scott Knoop at Van Dorn Demag Strongsville engineering laboratory conducted the following trials. The valve size tested was a 45mm. The resins run are listed on the charts as is the valves tested. All tests were with the same injection unit setup.

The Zeiger valve distance to close was 0.120" and the Demag valve distance to close was 0.098". The APV valves tested hade ring lengths of 1.175 and 1.205, this establishes distances to close of 0.039 and 0.009 respectively.

The most impressive data in the weight tests was the all but two last place positions (by wide margins) of the Zeiger valve in all resins and conditions. The Dray APV valves scored first in (12) of the (16) trials. In the zero decompress trials the Dray APV was first in (6) of the (8) trials.

In this graph the pressure drop calculations are shown to be accurate. As expected the 0.009 distances to close APVs had the longest recovery times. The times were longer on the more viscous resins (PE and PC) and less on the Styrene and Nylon. As you will note the recovery times for the 0.039 APV tests were virtually the same as the 0.120 Zeiger and the .098 Demag.

In this series of trials the APV valves were lower in average melt temperature in (14) of (16) runs. It should also be noted that the 0.009 distances to close APV actually scored first in (10) of the (16) runs. The reason for this is the ability of the APV to provide dispersive and distributive mixing to the resins.