The Gunperson's Authoritative Internet Information Resource.
The Gun Zone banner

Glock e-toolA contributory factor

Glock Six-Part "Upgrade" FAQ

Behind the April 1992 Technical Bulletin issued by Glock, Inc.

In the matter of Glock's infamous "Upgrade" (let none dare call it a recall!) of 1992, the following Glock pistols became eligible for replacement of six parts to provide increased safety and reliability.

The mysterious Glock "spring loaded bearing." Those six parts are: The revised alpha designations, current through 2001, of models requiring Armorer attention:

Model Number / Chambering Serial Number Range
Model 17 (9 X 19mm)  AA through XG, inclusive
Model 19 (9 X 19mm)  …through XK, inclusive*
Model 20 (10mm)  …through WX, inclusive
Model 21 (.45 ACP)  …through XM, inclusive
Model 22 (.40 S&W)  …through YB, inclusive
Model 23 (.40 S&W)  …through SL, inclusive
* - Alpha-prefix EH require the "Plus" trigger bar.
Essentially, this covers all Glock pistols manufactured before mid-October 1991, estimated at various times by Glock officials at between 385,000 and 500,000 handguns.

Additionally, prior to alpha prefix ALD Models 21 Glock pistols are also eligible for further upgrade, a slide modification to regulate slide velocities to the .45 ACP cartridge. The large frame pistols were originally engineered for the higher pressure 10mm cartridge, and Model 21 owners unwittingly acting as "Beta testers" discovered that the .45 ACP guns suffered from reliability problems due to the mass of the slide2.

Pre- and post-upgrade parts Transition to the "upgraded system" was scheduled to begin on 1 August 1992, and Glock owners of record via Warranty Registration Card were supposed to have been contacted by the Smyrna, GA importer. Field reports suggest that this contact has been, at best, somewhat problematical.

If you are an owner of a Glock Model 17 (and 17L), 19, 20, 21, 22 or 23 pistol in the above Serial Number range and have not been contacted or already upgraded, or if there are further questions pertaining strictly to this matter, contact Glock, Inc. at 770-432-1202.

If you are the owner of a Model 24 or higher none of the foregoing need concern you. Be advised, however, that there have been other, subsequent "product upgrades" (involving, among other components, trigger springs, ejectors, recoil spring guide rods and rear frame slide rails) across the entire spectrum of Glock pistols, now up to the Models 39.

Contemporaneous Report

From the April 1992 issue of Gun Tests:

Glock Changes Firing Pins

Glock Inc. executives claim there is no basis to the reports that all of the company's pistols are being formally recalled. Nevertheless, they say that, as of last Sep­tem­ber, they have instituted a running change in the design of the firing pin safety mechanism. All of the pistols that were assembled after the middle of October include the redesigned parts.

The reports are apparently based on a pair of unintentional discharges experienced by a Suffolk County, New York, law-enforcement agency. Glock says one of the two pistols may have been altered outside of the factory. We understand, however, the department is steadfastly denying that claim. The second pistol was not examined by Glock.

A New York newspaper3 reported on the incidents, but Glock says the information was inaccurate, and was evidently passed on by a disgruntled member of the department. Glock also believes a competitor has faxed the New York newspaper article all over the country.

The improved parts will be refitted into all previously made Glock pistols "as they come to us for normal servicing and repairs," said national sales manager, Bob Gates. "This retrofit is important, not for safety concerns, but to assure product continuity 25 years down the road, because we don't want to have to stock parts for two separate designs4."

One Possible Scenario Behind the
1992 "Product Upgrade" Announcement

There has been some question over the years about what actually led to the six-part "Product Upgrade" announced by Glock, and many, buttressed by information developed by both the writer during his tenure as Industry Editor of Gun Week, and the "Industry Insider" column by the Editor of American Handgunner, concluded that it was directly related to "the AD heard 'round the world5" which occurred on 2 January 1992 to a Suffolk County (NY) Police Officer.

Glock 19 That infamous event took place as the officer was preparing for his late duty shift. Alone in his home, he removed his unloaded service issue Model 19 Glock from its safe-keeping location, inserted a magazine, and attempted to chamber a round in the conventional method. As he released the slide, the handgun discharged inside his bedroom. Fortunately, as he had been observant of Firearms Safety Rule #2, there was no personal injury, but the officer was clearly shaken. He contacted the Suffolk County P.D. Range and requested that an armorer be available to inspect his service pistol. (As with most modern law enforcement agencies, all rounds discharged off the range must be accounted for, and if it could not be shown that the gun had experienced a mechanical malfunction, the officer might very well have been subjected to retraining, a departmental review, and an investigation with Internal Affairs personnel questioning his neighbors about whether they had observed any "problems" in the officer's marriage or had sensed a substance abuse situation.)

Although the Firearms Training Section personnel immediately concluded that the officer had disobeyed Firearms Safety Rule #3 (and Glock Commandment #1: "off-target, off-trigger!"), the man was adamant that all safety procedures had been observed during the loading sequence, and that he was not going to go on duty without his Model 19 having been thoroughly checked. The Section Sergeant authorized an hour's overtime for one of the armorers to stay on station to inspect the Model 19. When the MOS arrived the armorer was so certain that the errant discharge had been "operator error," that when he test-fired the pistol, he didn't even go onto the range, choosing instead to step outside the range HQ building and perform an administrative arming of the Model 19. Reminding the officer of the "keep a straight finger" dictum, the armorer inserted a magazine, and racked the slide.

The Glock discharged, sending a 124-grain +P JHP into the ground behind the building.

Quickly moving to the range proper, the validated officer watched as the armorer thrice more attempted to arm the Glock. Twice more it "slam-fired."

On the third and final attempt, it lapsed into a three-shot burst. (As has been oft-observed by the author, full-auto fire is always exciting, but an event best planned for… spontaneity is good in many things, not, however, in firearms!)

The officer was issued another Model 19 so that he could go on duty, and an emergency call was placed to Glock Regional Manager Steve Tretakis first thing on the morning of 3 January 1992. Within hours the malfunctioning pistol was FedEx'd back to Smyrna, Georgia.

Although the Suffolk County P.D. Range and Armorer's Section is less than five miles from the author's home, it wasn't until a week later, while attending SHOT Show in New Orleans that I became aware of an integral part of this story in a late night conversation in a hotel bar with then Glock V.P. Karl Walter and some of his staff… which conversation, by "sheer dumb luck," is recorded on a microtape cassette. In the early morning of 12 January 1992, Walter revealed that Glock Ges.m.b.H. had redesigned several parts, including the firing pin "safety plunger," a rough sketch of which he provided on a cocktail napkin which showed the new "flared" tooling at one end.

The full import of that discussion did not become apparent 'til the following week when the details of the S.C.P.D. accidental discharge became known to the writer, along with the information that Glock, Inc.'s armorers had inspected the malfunctioning Model 19 and declared that it had been "tampered with." When Glock Service Manager Bill Haberland flew to New York to return the pistol and inform the armorers of their finding, a shouting match ensued about who did what, if anything, and when. The critical point came when Sergeant Ray Flood threatened to issue an alert about the incident over the national law enforcement teletype.

At that point, Haberland backed off a bit and handed over those above referenced "six parts," indicating that those updated components should address any concerns that Suffolk County P.D. might have, yet continuing to disclaim any responsibility for the malfunctions which had occurred.

Understandably feeling vindicated at this point, S.C.P.D. armorers requested enough new parts to service the 1200 Models 19 then in use… but received only 250 two days later, all that Glock, Inc. had at that time. The armorers reported, and Karl Walter confirmed, that those sets of improved parts had to be stripped from 250 new (post-October 1991) pistols in Glock's inventory.

Approximately eleven weeks later, with Walter insisting that it was a "voluntary product upgrade" rather than a "recall," Glock issued its April 1992 Technical Bulletin announcing the six new parts. Originally projecting that all affected pistols would be upgraded in "a year or so," the undertaking was so massive that it was still on-going as recently as 1998, mostly because many of the owners of those early Glocks never got word of the upgrade availability.

The DEA "Frisbee Test"

While Suffolk County P.D.'s "AD heard 'round the world" certainly lent impetus to Glock's "going public" with its upgrade, clearly this is something that had been in the works for a considerable period of time, and well before October 1991.

Glock Perfection Documents first provided by a source professing to be knowledgeable about the situation, and buttressed by other documents released under a Freedom of Information Act (FOIA) request, indicate that failure during the "Frisbee" portion of the 1991 test protocols (DEA-91-R-0023), led to the Austrian pistols being "rated as unacceptable" for broader adoption by the Federal law enforcement agency. (Whereas Glocks had been one of several DEA-approved handguns, subsequent to the tests, the agency selected the SIG P228 as their "official" handgun and removed Glock as an option. No existing Glocks were removed from service, however, and Glock's Walter stated at the time that "Glock is not going to protest the test results, since we feel they were fair and equitable.")

§L.16.1.b.(2)(c) of the agency's solicitation, Abuse Testing, required that submitted "weapons, with magazine inserted, will be thrown six times, three times in such a way as to land on the right side and three times in such a way as to land on the left side. The throw will be for a distance of approximately fifteen feet, not to exceed a height of approximately four feet, to land on a floor of quarry tile or concrete." From the FOIA request documents it was learned that beneath the Consensus Report heading of "Weaknesses" the evaluation committee stated:
Throw test: Frame 479 (with) slide 318, 1st throw left side the slide came off both rear rails. Frame 474 (with) slide 479, 1st throw right side slide came off right rear rails, rear pistol grip under landyard{sic} loop hole cracked and broke the grip after throw test, pulled the trigger would not fire. Tap, Rack, Bang would result in function of the weapon. Frame 477 (with) slide 305, slide came off right side rear rails on the first throw. Frame 318 (with) slide 474, 1st drop rear of pistol grip broken by the landyard{sic} loop hole. Based on the failure of the slides coming off the rear rails it was concluded that the weapons would not be further tested. Therefore no firing of the weapons took place after the throw test….
The original source explained:
Of all the tests of throwing Glock pistols, none before had ever been done with a loaded magazine in the pistol. Magazine out of the pistol, no problem. Magazine in the pistol, BIG PROBLEM. In the DEA tests the slide came off the frame and the pistol discharged. Gaston Glock and Reinhold Hirschheiter6 concluded that the slide rails were too short and that the firing pin safety needed improvement. Gaston determined that it would be too expensive to recall all pistols and replace the frames and firing pin safety systems.
At this point the narrative takes an especially interesting twist.
Karl Walter asked Gaston Glock to recall all pistols and fix the total problem. He refused and directed Karl to call the eventual project a product "upgrade" and NOT a recall.
The source expressed the opinion that this disagreement signaled the beginning of the end for Karl Walter at Glock, but, while stating that the upgrade of the "…firing pin system was the cheapest way out," described the chronology of a subsequent redesign.
Even if the slide would still come off, the pistol would not discharge due to the new system. However, Mr. Glock did know that the frame rails were a big problem, in particular with future Federal bids, so he quietly lengthened the rails.
This is why there are at least three7 different propagations of Glock receivers around, for the first attempt proved too long, and the subsequent "medium" length of frame rails seemed to do the trick.

In the final analysis, the April 1992 announcement of the "Six Part Product Upgrade" appears in retrospect to have been a temporizing measure while the frame was being re-engineered to withstand the demanding protocols of the "DEA Frisbee Test." Officials at Glock have refused to confirm this for the record, of course, but what is known is that the the newer pistols have not only been once again authorized by the DEA, but have been adopted for issue by the FBI.
by , formerly famous gunwriter.
Post questions or comments in The Gun Zone Forum.
© 2000-2008 by
The Gun Zone
All Rights Reserved.
Elite Operator TacticalElite Operator TacticalTM Gear is exclusive with EOT.

This page, as with all pages in The Gun Zone, was designed with CSS, and displays at its best in a CSS1-compliant browser... which, sad to relate, yours is not. However, while much of the formatting may be "lost," due to the wonderful properties of CSS, this document should still be readable.