I thought I would add more Socialism to this thread.
Hatcher has a number of omissions in his book "Hatcher's Notebook". It is unfortunately that it is the closest we have to a comprehensive account of the problems of the low number 03's, and yet, it is neither comprehensive nor unbiased. After working in a professional, hierarchical organization for decades, Hatcher had just retired, but before retiring he had climbed his way to the top of the Ordinance Department. He lead the Ordnance Department through WW2, a very difficult task . In all respects Hatcher is an exceptional man, probably an outstanding manager, and he is ambitious. He did make it to the top of an Organization where even now, it is very difficult to make General, never mind lead the thing. Though out his career he gave himself a career boost by making himself a celebrity. He wrote extensively in the Army Ordnance Magazine, in the Arms and the Man/American Rifleman. He was a well known individual with high brand recognition within the shooting community, and incidentally, within the Army. He wrote a number of books, becoming an authority in his field, and deriving a good income. This guy must have been a 16 hour a day sort of work a holic. A cynic would interpret his career in the light of career advancement and financial remuneration. There must be more, but then, I have met any number of people who were not happy unless their name was on top of the Organizational Chart. (Why does anyone want to be President of the United States? Pay is insignificant compared to industry, retirement laughable, and no one is happy, yet every four years, there are candidates.)
Hatcher retires and in 1947 writes Hatcher's Notebook, a book still in print. He has already embarked on his second career, and that is climbing to the top of the NRA leadership. Incidentally, something which he accomplishes. I have looked at current pensions for Major Generals, and while good compared to Social Security, it is chump change compared to a top rank NRA salary. Current top level NRA salaries are around $600,000 to $800,000 a year. And you probably get free parking. One very critical, perhaps the most critical job requirement at the NRA, is good relations with the Army. It has been almost 50 years since the Army punted the NRA out of the Pentagon, (1968), so no one today remembers the close relationship the NRA had with the Army. But it was it was more than close. The NRA was a hyper patriotic organization, staffed by super patriotic Army Ordnance retirees, who were more Green than Army Green, and you can clearly see this in the 1960's American Rifleman magazine. The Army financed the National Matches, which was equal to at least 25% of the total NRA budget. Each year you can go through the NRA annual meeting sections where one year the Freedom Flintlock is presented to the Army Chief of Staff, the next year the Marine Corp Commandant, and it is all kissy, kissy, huggy, huggy with the Armed Services. There are all sorts of stories about Service Men, whom the NRA trained, shooting their way to victory in Vietnam. This was how the NRA proved that its existence was of value to the Department of Defense. NRA writers could call up the Army and receive technical answers, latest research and you see that in the magazine.
So, after retiring, Hatcher is writing another book. More income, but he wants to maintain good relationships with the Army. If you notice in his writings, Hatcher is always supportive of the Army, never critical. Hatcher is a master of the misdirection, nothing is ever the Army's fault. He masterfully spins the low number 03 failure into an Army Triumph. The Army overcame those evil, rascally forge shop workers and built the ultimate 03: the double heat treat. In my opinion, the whole low number episode was a complete Army failure. They built 1,000,000 rifles that as a class, were so defective, that they should have been scrapped. This information was not released to the public, seems to held only within the Ordnance Department till 1927, and even then, a reasonable accounting of the dangers of these things does not come out until 1947. In 1927 an Army board recommended scrapping all of the low number rifles. The Army decided to keep them in service till the rifle blew up, or wore out. Once the weapon went to depot, then the receiver was scrapped. The Army knew that a certain percent would blow up and injury someone, but it was cheaper to injure a Soldier, Sailor, Marine, than replace the inventory of defective rifles. I consider this decision immoral, not Solomonic. They don’t do this today, in fact the Army is rather safety sensitive, because now, the costs for long term care come out of the Army budget. Back then it was different.
As I wrote earlier, Hatcher has been the only account about the M1903 low number problem, and he spins the story into a morality play: Good Army, rascally workers. This absolves Army management from all responsibility in the matter and scapegoats some rather insignificant people. Makes for a wonderful tale.
I am going to offer two paradigms. The Hatcher account, widely accepted by the population, follows the paradigm that bad products come out of bad factories run by bad workers. A different paradigm is that bad products come out of bad factories run by bad managers. It turns out, Darwinian Competition is eliminating institutions those corner stone philosophy is to blame workers for the poor products they produce. These institutions are being out competed by groups that that blame bad managers for bad product.
Again, who has the authority to organize the production flow, write work instructions, buy capital equipment? Blaming workers for poor product allows lazy managers to abrogate their responsibility in planning, managing and organizing. When defective M1903's are going out the door, who in the chain of command is asking "why are we producing defective rifles?". Who else has the authority and responsibility for finding error inherent processes or messed up production flows?
It is my memory that the first reported low number blowups from Hatcher's Notebook were at a cartridge manufacturer. The first receivers listed as blown up were a 1907 vintage receiver and a 1917 receiver. These receivers blew at National Brass & Copper Tube factory, a factory making ammunition for the US military during WW1. The 1907 receiver fragmented and blew a piece of shrapnel piercing the lung of the operator.
National Brass & Copper Tube had college educated and trained metallurgists, you can see that in the report. They were able to refute any self serving BS coming out of Springfield Armory about their receivers. Proving an Arsenal is wrong is not a task that a lowly Infantry Officer could do, or would even want to do. Something that is important to remember that hierarchical organizations are incapable of self reform. The first thing they do is shoot the messenger of bad news and deign there is any problem. Reform can only occur due to outside influences, and here you have it, an organization not under the War Department, outside the chain of command, pointing out that Springfield Receivers were burnt.
Old type factories had chaotic process control and unstable production lines. You just have to have gone though old line factories to see piles of in process product, defective product tossed in corners or on the side of a bin. In time no one knows why there is pile there and the defective product gets put back into production. With piles of in-process parts in bins, the time between when a process got out of tolerance and the discovery, a lot of defective product has made its way down the production floor. It was acceptable to make bad product. Inspection criteria assumed a certain number of defectives per lot. Mil Std 105 shows this inspection process. Lets say the Standard allowed 2 defective parts per 100, and the inspector found three, then he was supposed to pull more, and if lets say out of the next 50, only found 1 defective part, the whole lot was shipped, even though they had just found that 4 parts out of 150 parts were bad! The production control system assumed that some bad product was going out the door. Production was monetarily incentivized to ship everything. Quality Control was trying to separate out the good from bad. Product quality was a cat and mouse game between the production department and quality control. Also, it is well known, if Quality Control rejects too much product, management will fire the Quality Manager. This practice continues to day. Just look at the GlaxoSmithKline, lawsuit, the drug maker fired their Quality Control manager.
http://www.nytimes.com/2010/10/27/business/27drug.html?_r=0 Join the lawsuits here:
http://www.texaslawyers.com/coomer/pharmaceuticalqualityassurancelawyer.htm
I recall a quote from a General Motors Executive from the 1970's. A huge number of these cars were defective. The Executive was asked about this, and he said
"Marketing sells it, manufacturing makes it, and Customer Service makes it work". That was the attitude in manufacturing for a long time. I toured the GM Arlington plant in the early 1980's and at the end of the line, at the time I was there, seven guys were pushing a car away because it would not start. I have toured a couple of modern automotive factories now, and if you ask, "when was the last time a car did not start", no one knows. If it happened, it happened so long ago, they don't know. Anyway the GM culture towards product quality almost bankrupted GM. If a customer spends a year salary on a new car, they don't want to spend days or weeks at customer service, waiting for Dealer Ship grease monkey's to fix the thing. They expect a good product the day they drive the car from the lot.
GM was incapable of self reform, and it took the Japanese with their cheap, but quality cars, to force GM to make decent cars. I remember people who bought those cheap 1970's Japanese cars during the gas crisis, and the owners found out Honda's, Toyota's, Datsuns, were reliable. The little suckers ran and did not fall apart. When the Japanese moved into the mid scale market that almost killed GM, Ford, Dodge.
Through Google Books there are other sources of information that have recently come available, and they provide enough information to confound the Hatcher morality tale about the low number rifle problem being due to a few bad apples in the forge shop.
Report of Tests of Metals and Other Materials made in Ordinance Laboratory at Watertown Arsenal Mass, Fiscal Year 1918, War Department Document 901, 338 pp.
https://books.google.com/books/read...printsec=frontcover&output=reader&pg=GBS.PA25
I have copied a couple of the Watertown summaries:
Examination of Receivers from United States Rifles Model of 1903, burst during Navy Target Practice
Conclusions:
One of the receivers was neither case hardened nor heat treated. It was in the perlitic condition as shown by the micrographs. The structure was very coarse.
Examination of United States Rifle, Caliber 30, Model of 1903, which failed at Camp Greene NC
Conclusions.
The receiver of this rifle had not been properly heat treated prior to being put in service. Microscopic examination showed the metal to be very coarsely crystalline and the structure was that obtained by cooling at a fairly rapid rate from a high temperature. The metal was very brittle under impact, as evidenced by its being readily broken when struck a light blow with a hammer.
Broken Bolts from United States Rifle, Caliber 0.30 Model of 1903
Object:
The object of this investigation was to make a complete examination of these two bolts and if possible determine the cause of failure.
Conclusions:
It is the conclusion of this laboratory that the failure of both of these bolts is due to the same factor. Both were very hard and brittle and their resistance to sudden impact was very low, as could be ascertained by securely fastening the metal in a vise and striking light blows with a hammer. the brinell hardess number on one of these was 430 and on the other it was 489. these structure of both was martensitic. This structure is characteristic of very brittle material
The chemical composition of these bolts is not within the specified limits with regard to carbon, manganese, and silicon
Investigation of cause of failure of United States rifle Model of 1903, No 108 448, which failed in target practice at Camp Shelby
Conclusions:
It is the conclusion of the laboratory that the failure of this rifle was due, at least partially, if not entirely to defective material composing the barrel. Chemical analysis shows the phosphorus to be 0.083, which is entirely too high and should not be allowed. The metal is very severely banded or streaked longitudinally, as shown by etching polished sections with Stead's reagent. Nonmetallic inclusions were present in considerable quantity and were greatly elongated in the longitudinal direction of the barrel.
The weakening effect of these streaks, rich in phosphorus and containing a large amount of nonmetallic inclusions is particularly apparent under shock. The brittleness under sudden impact caused by high phosphorus and nonmetallic inclusions, especially when in the banded condition, is without doubt the predominating factor contributing directly to the failure of this rifle. The metal of the receiver was very hard and brittle, as can be at once understood by observing micrograph 4215 at 500 diameters , which shows the structure to be martensitic.
A couple of issues fall out, and they are not as simple as a morality play that vilifies forge shop workers. Firstly, the materials they are using are inconsistent and don't meet specifications. . They evidentially don't have an incoming material inspection nor a material certification of incoming materials. I know from historical sources that the materials of the age were widely inconsistent, and the chemical analyses by Watertown shows that. These are plain carbon steels with slag, inclusions, high amounts of Phosphors or Sulfur, and all of these defects weaken the end product. I remember Colt had a metallurgical department that performed chemical analysis of the steels Colt received. These Arsenals don’t order steel in 10 pound, 20 pound lots, they order steel in lots of 100,000 pounds or more. We are talking about costs in the $250,000 dollars per order. Why did not Springfield Armory require a chemical test as part of the purchase order or do it in house? Heat treatment temperatures are based on the percentage of carbon in the steel and the heat treatment is specific to the carbon content. Improper carbon content results in a improper heat treatment, and the end result is a part with improper material properties for the application. Like too hard, too soft.
This is an excellent post by Firstflabn
http://www.jouster.com/forums/showt...on-(NOT-for-the-reason-you-re-thinking)/page2
A sure sign this discussion is in big trouble is when I have to save it with my metallurgical expertise. But, since in the land of the blind, the one-eyed man is king, I'll do like OJ and take a stab at it.
First, Hatcher was no metallurgist. Even in the context of the science of his day, he was a beginner. I think he says he took ONE course in metallurgy. He says a couple of really stupid things in his Notebook, but that discussion can wait for another day.
A couple of comments:
1. According to a chart prepared by the Tempil stick folks (if you don't know what a Tempil stick is, stop reading and go find out, you won't benefit by going further now), there is only about a 100 deg F difference between the top of the safe forging temp range and the bottom of the burnt temp range. Therefore, it is ludicrous to believe SA didn't have pyrometers in their forge shop, Hatcher or no. Further, the FY18 report in Brophy's SA book says they installed improved pyrometers in the hardening shop. Why would they have the good stuff - and upgrade it - to use on a much less critical operation at a 1000 F lower temp and none at all in the forging shop?
If the pre-1918 forge temp range is the same as the 1942 spec presented in Brophy (p. 549) - 2300-2340 F - then SA is into the 100 deg F no mans land between safely forging and burning.
2. The plot thickens - as carbon content rises, the burnt temp drops. Brophy shows a carbon range of .30 to .38 for receivers. Moving from .30 to .38 LOWERS the burnt temp by about 30 deg F. Interestingly, the FY18 report celebrates their brand new chemical lab which allows them to (apparently for the first time) do a chemical analysis "for all the steel entering into components or tools." This smells like they previously had only checked the paperwork from the outside supplier providing the receiver blanks. 30 deg may matter if you're bumping up against the safe max temp already.
3. According to the Tempil chart, the forging range is a bit over 600 F wide. Thus, SA wrote specs to operate at the tippy top of the safe range. They might have been worried about forging laps, etc., but that kind of defect would probably have been revealed in proof firing - so, they were concerned about budget (in peacetime) and production (in wartime), not burnt steel.
4. Interpreting the Tempil chart on my ancient monitor, there is no discernable difference in color between the top of the safe range and the bottom of the burnt range (100 F, remember). I'd have to ask someone with foundry experience, but Hatcher's story sounds suspect to me - he may not have known enough to call BS on the "cloudy days" cover story.
3. Another reference I have warns that care must be taken when working forgings that have been heated to near the max safe forging temp as getting in too big a hurry (whomping with the whomper) will raise the forging's temp from friction. My bet is that when that happens, the chewy chocolate center is hotter than the outside, so even modern pyrometers would not help.
I suspect this knowledge existed in heavy industries doing really big pieces - locomotive, ship building, hydropower, etc., but that the combination of stodgy old ordnance officers and budget parsimony created an avoidable (but inevitable) f*ck up.
What I wish I could find out is whether the acid etch test (to identify burnt steel) was available in those other industries before WWI. I have no experience with it, but from my materials lab knowledge it doesn't look too tough to perform and interpret. It's a destructive test, but I wonder if the edge of the tang or one side of the recoil lug could be tested and still have a functioning receiver (with possibility of leaving no more than a blemish). I'm no longer in the testing business, so I don't know who to call for a freebie. If it could be done and only leave a minor boo boo, somebody could have a booming business. The test is 100% reliable, though the interpretaion is visual, so might need a practiced eye.
I'm tired of hearing myself type, so I'm out.
I am going to summarize what I consider the important points of this post. Human eyeballs could not maintain the heat treat or forge temperature tolerances of Class C materials. So why were they using Class C materials, and not something else? These temperatures are specific to the carbon content of the steel, and yet, the Army is not ensuring that its vendors are supplying the proper steel.
Another issue shown by the Watertown document has to do with heat and temperature. There is either too much heat, too little heat, indicating no temperature controls. Generally the parts are brittle which is an indication of too much heat, (see above) but to have but to have a receiver which was neither case hardened or nor heat treated shows a production line out of control. How can a receiver get out the door and not have been either heat treated or case hardened? Springfield Armory is shipping junk, either intentionally, but more likely, because they don't know they are making junk. The Army’s premier factory is out of control and Hatcher, instead of pointing out that Army management is at fault, is blaming the forge shop workers.
