Hi all,

I first became involved in working on Glock pistols back in 1998. The one thing that I notice after firing the G24,22,19 and 17 pistols, was that although the guns functioned, ejection patterns were wildly inconsistent. Brass was hitting my head, going completely vertical, off to the left, and many just off my right shoulder at about 5:30 if you viewed everything from above the gun (muzzle being the 12:00). I was told that the problem was caused by "limp wristing", projectile dysfunction, low testosterone and a myriad of other reasons that did not sit well with my mind.

In talking with Jeff Gonzales many moons ago, the phenomenon called the horizontal stovepipe also seemed more common in the Glock pistol than other service pistols that we were both familiar with. After having experiencing a few firsthand in G22s and a couple of 9mms, the data was stored away for some 8 or so years as I had sold my Glocks in search of greener pastures. 

For the past month, I have been scanning various forums and youtube videos documenting such issues, and people's attempts at resolving the problems they have been experiencing. I was looking for some logical pattern in the "If I change part A, the benefit was always X" Ditto with parts B and C. What I have concluded is that while changing the RSA helped for some, it does not help in all cases. The same is true with replacing just the extractor or just the ejector. In almost all of the videos I watched, the single part replacement provided no improvement (unless the problem was failure to extract, in which case replacing the extractor usually helped).

So what is the root cause of the problem Mr. Randy Fancypants Know-it -all? To put it quite simply, the ejection port horizontal wall is too tall, and the cases are hitting the vertical inside wall instead of being guided out by the exit ramp cut. I will try to explain what I am seeing a bit further down in this post.

This is why just changing the RSA or Extractor or Ejector by themselves only helps in some cases, but not in others. Each gun is different and each shooter is different. So results can vary based upon how an individual absorbs recoil forces as well as manufacturing tolerances for the particular gun. Below are some rules of thumb that can help with diagnosing and isolating a particular problem:

Recoil Spring Assembly(RSA)
As long as the slide will lock to the rear while shooting with your non-dominant hand with moderate to light grip pressure, the RSA is not the cause of erratic ejection, stovepipes or failures to extract. I say this because if there is enough energy to lock the slide to the rear once the last round is fired, there is enough force generated to reliably kick the spent case out of a properly set up ejection port and feed another round when a full magazine is used. This of course implies that the loads being shot are of sufficient quality and power to reliably cycle the pistol. Faster slide velocity rearwards can help ejection by causing a more forceful impact of the case against the ejector tip. But in many of the guns I have inspected, it just meant the brass hits your face harder.

Extractor
The Gen 4 extractors I have seen have a secondary angle on the hook that is not present on the older extractors. My assumption (I have not spoken to any of the engineers at Glock) is that the secondary angle allows for less resistance against the feeding case as it slides up underneath the extractor hook. As the cartridge begins its entrance into the chamber throat, the back of the case must change from a point contact (12:00 position of the case head) to flush and parallel with the breechface. The new extractor angle helps with this aspect, but can lead to other problems during the extraction phase. 

As the barrel unlocks and ramps downward it impacts the locking insert in the frame. The barrel essentially bounces. Add this force to the torque forces imparted by the rifling and there can be enough transmitted shock to bump the extractor claw as it tries to pull the spent case from the chamber. The secondary angle on the Gen 4 extractor while making it easier for the case to slip up and under the hook during feeding, now applies force to the case rim by pushing it down and away from the optimal contact point making it easier for the case to slip away from the claw.

The earlier extractor versions and I assume aftermarket extractors have claws that are parallel with the side wall of the breech face. This means that the ejector is nearly always applying a force that is perpendicular to the breech side wall, and at the 3:00 position of the case no matter how low the case travels down the breech face during extraction and ejection.

Ejector
Until recently, the Gen 4 9mms were shipping out with the #336 ejector. I believe the instructors at the Rogers Shooting School found that by altering the ejector position and possibly length, they were able to minimize many of the stoppages. I have heard of others attempting to do the same with mixed results.
The new ejectors that Glock is installing as an in-house service essentially alters the exit angle of the case trajectory. By changing the location and dimensions of the ejector tip, the contact point is now at about 7:00. This makes the ejection angle steeper and will hopefully clear the inner sidewall of the slide that is vertical. If executed correctly, the brass flight path should be roughly a high arc at about 3:00 to the shooter (I don't have one of the guns with the new ejector, so I cannot confirm this).

What is happening in my stock Gen 4???
For the sake of argument, let's say have just pressed the trigger on my Gen 4 box stock new model 19. I am using Winchester White box ammunition. The gun discharges and the bullet leaves the barrel. The barrel unlocks from the slide and bounces against the locking insert. Now, because I am using ammo that I know has a shallower extractor cut, the Gen 4 extractor overrides the case rim for reasons discussed above. Failure to extract leaves me with a double feed. I clear the stoppage and get back to shooting as usual.

I press the trigger again, and the gun discharges. This time the bounce of the barrel does not cause the extractor claw to override the case rim. The case extracts from the chamber and hits the #336 ejector tip. As the slide moves rearward, the front of the case impacts the inner vertical sidewall of the slide. Because of the less that optimal dimensions of the brass, and the secondary angle cut on the extractor, the case is released from the extractor prematurely and is floating in space as the slide moves rearward, and the next round begins to rise in the magazine. The upcoming cartridge contacts the free floating brass and lifts it vertically in the ejection port. The slide has already stopped rearward travel and is now moving forward. Without any other forces acting on the spent case, the slide closes on the hovering brass. The result is a horizontal stovepipe where the case mouth is wedged between the hood of the barrel and the back of the breech face. Tap roll rack and I am back to business as usual.

I press the trigger a few more times, and the gun cycles reliably, but brass hits my left arm, the rim of my hat, over my right shoulder. The last round hits the underside of my hat rim and decisively wedges between my eyebrows and safety glasses. After my well selected curse words and putting the gun down in a safe direction, I dislodge the still sizzling empty case from my glasses to the aroma of burning hair and skin... 

Does any of this sound familiar?

Possible solutions?
Our solution is to lower the ejection port, and change the ramp out angle to a 45 degree rather than the steeper factory angle. I designed our barstock extractor and it is in the testing phase.
I also have a prototype ejector.

Magsz donated his problematic Gen 4 19 which I believed he called Satan spawn. Most of our testing has been with Winchester ammunition. The gun was tested before modification and as each new component came online. The lowering of the ejection port was the last operation, as from prior work on competition Glocks I knew would have the greatest impact.

With the lowered ejection port, our extractor and the prototype ejector 400 rounds have been fired without incident, and the ejection is consistent at the 3:00 position regardless of who was shooting the gun.

Today I swapped out the proto ejector and reinstalled the #336 factory part. The result was the same- consistent extraction and ejection being at 3:00 for all three shooters (two left handed shooters and myself) We shot the gun one handed, right and left hands, limp wristed, no pinky support, thumb and middle finger only grip as well as freestyle for 150 rounds.
The brass ejected at 3:00 to the shooter's right side regardless of shooting style or grip strength. NONE, I repeat NONE of the shooters had brass come near their head!

Earlier in this post, I wrote about the accusation of limp wristing being the cause of my problems. I believe that I have confirmed (at least for myself) that it is a fallacy. Properly set up, as long as there is at least 20 pounds behind the gun, a properly designed and executed gun should function, extract and eject without fear of stovepipe or brass hitting the operator. 

The downfall of this is that in milling the slide, you void the warranty on your Glock. Ultimately, it is up to you the owner to decide whether it is an acceptable risk.

Until Glock takes a serious look at the ejection port geometry, I fear that problems will plague the Gen 4 guns no matter how many changes to the ejectors and RSAs they make.

Thank you for reading this, and as always, I could be wrong...

-Randy