NATURE OF THE BEAST
Part 1 - The Problem
By Dave Scott

Several years ago, Precision Shooting columnist Buddy Salyer took exception to Col. Whelen's contention that only accurate rifles are interesting. The Kentucky counselor and high power advocate asserted that inaccurate rifles are the really interesting ones.

Well, if Bud currently holds that view, he'd find my one-and-only factory rifle plumb fascinating. This rifle, which I call the "Beast" in polite company and whose moniker is typically preceded by a string of blistering adjectives in the privacy of the range, was so named because over the last 28 years it's shot like hell. Apparently, it's the nature of the Beast.

It wouldn't be so bad if the rifle was a rusty relic that had rolled off an American or German production line in the frantic midst of WW II. Or I could show compassion if I had rescued it from a dysfunctional family of obsessive barrel scrubbers. I could even be supportive and nurturing if the rifle was a two-fisted .30/06 trapped in a scrawny, 6-pound body.

None of these is the case. I bought the rifle new in 1967, shot it now and then for 12 years and retired it without so much as a speck of rust in its bore, a dent at its crown or an oval lapping job in its throat. The Beast is a Winchester Model 70 in .225 Win., and if abuse can't be blamed for its maddening inconsistency, neither can a lack of heavy metal. This beefy number features a straight-taper, 24" barrel measuring a broad .78" at the muzzle. Likewise, it would be folly to imply that the Beast's problems might center about its .225 Win. chambering. The semi-rim cartridge is, after all, a replica of the legendary .219 Improved Zipper and should perform quite well, thank you.

I recently retrieved the rifle from the closet because I figured it would be an ideal subject to begin testing the theory that narrow clearances between case neck and factory chamber tend to eliminate flyers and thus improve accuracy. This occurs because "thick necks" restrict the amount of bullet tilt in the neck, throat and barrel, and a bullet undamaged by tilting is a more accurate bullet. Or so the theory goes.

If ever there was a candidate for a thick-neck test, the Beast is it. The rifle has a long history of flyers. I know this because I've saved every group ever fired with it. From 1967 until 1979, the rifle, on average, threw one or two bullets in about every other group. For 5 shots at 100 yards, the best groups were typically in the low .5" range. Groups with flyers exceeded 1" now and then and topped out at about 1.8". With 52-grain Speer hollow points and 33 to 34 grains of old Reloder 21 powder, the 12-year average for all 100-yard groups was .85".

Predictably, when I again began working with the Beast a few months ago, it repeated the same pattern, except in somewhat wider dimensions. Now, the tighter groups averaged in the mid-.6"s. Groups with bad flyers mostly exceeded 1", up to 1.7". Even though fewer than 500 rounds have been cycled through this rifle, and the Speer bullets I've always used still engrave on the lands, I'd suppose we can attribute the deteriorated accuracy to advanced age. However, there's the possibility that the Beast just isn't as fond of my current powder, H414, as the Reloder. After going through several powders in the early load-development stage, the slow Hercules propellent showed the greatest promise in this rifle.

In recent testing of the Beast with three sets of cases having case neck-to-chamber clearances of .008" (.004" all around), .002" (.001" all around) and.001" (.0005" all around), no accuracy advantage was found between those cases with plated thick necks and regular factory necks. Two 5-shot groups with each set of cases produced averages of .82" for factory cases, .85" for the plated .001" cases and .93" for the plated .002" cases. All shots were chronographed to assure similar velocity levels.

Any accuracy comparisons between the sets of cases are meaningless because accuracy is hopelessly sporadic, and a total of only six "official" test groups were fired before I threw up my hands. This is not to say there are no revelations coming from the temporarily discontinued experiment. The important finding is that thick necks did not succeed in eliminating flyers and improving accuracy in this rifle. A secondary conclusion is that although, up to this point, thick necks haven't improved accuracy, neither have they harmed accuracy or presented any complications whatsoever.

So, based on the results thus far, have I thrown in the towel on this Model 70 thick-neck project? Not by a long shot. I firmly believe that fat necks eventually will benefit this tortured rifle, even if in its present condition they haven't. What we have learned is that in some rifles, such as this one, there are flaws much more destructive to accuracy than bullet damage due to tilting, and the truly dire ills must be cured before the more subtle improvements can be gauged.

Lest anyone level charges that I discontinued the project prematurely, be advised that the half-dozen test groups shown here by no means constitute the total effort devoted to tuning this gun to acceptable accuracy levels. The same distressing inconsistencies occurred in extended load development and experimentation leading up to the six test groups, which were produced merely to visually demonstrate the rifle's grouping tendencies.

In most shooting sessions, the first group of the day (after a fouling shot) displayed the worst flyers, usually measur¬ing in excess of 1". Typically (but not always), the second group was quite acceptably in the .6"s. For the most part, only two groups were fired at each ses

Temperatures ranged from 60 to degrees, and the barrel warmed only slightly in firing.

The fact that the barrel was thoroughly cleaned between trips to the range would lead us to suspect it shot better fouled, being that the second group usually was the tighter. Following that line of reasoning. I left off cleaning, and the first group next time again exceeded 1". Another theory going nowhere.

That the second group generally was smaller also would suggest the barrel was "walking" as it warmed. This has to remain a strong possibility, and upon conclusion of a few remaining tests, cryogenic stress-relieving will be tried.

I don't want to give the impression that I consider this rifle's sad accuracy a rari¬ty. Nor am I picking on Winchester. (A collective sigh of relief emanating from New Haven just rustled leaves across Connecticut.) All the factory varmint rifles I've ever owned have performed much like this one, despite my exhaustive efforts to coax from them consistent groups below, say, the .75" level. They exhibit good accuracy for one or two groups, then spray bullets across the target. Or they shoot well one day and then break your heart the next. Hell's bells, if you've ever fiddled with factory varmint rifles, you know what I mean!

The reason I demand more from this Model 70 is because with its relatively heavy barrel chambered for the fine .225 Win. cartridge, it should do better. After all, there should be some payoff to lugging around an extra pound or two of chrome-moly. But most of all, I've had it with rotten-shooting factory rifles. And I'm sick and tired of never knowing what their problems are.

So, for the time being let's leave the issue of thick necks. Instead, let's focus on the more pressing flaw in this particular Model 70. Why does it on average throw one or two shots out of 10? Folks, we're loitering at the threshold of the 21st century. Surely we now have unprecedented resources for resolving this question, considering the innovative services and cutting-edge technology available in the pages of this magazine. Let's take advantage of them and see what happens.

Granted, we could shortcut the process by truing the action and attaching a custom barrel, maybe throwing in a synthetic stock for good measure. However, that's not our purpose. If possible, we want to correct this factory offering exactly as it now exists. Most importantly, mah fellow shooters, we want to know what's wrong with it. Is it possible that there's a single factor accountable for most of this rifle's sorrow? If so, is it then conceivable that when (and if) we pinpoint it, we may understand what afflicts other rifles that have been carefully tuned and yet perform as erratically as this one?

Certainly, we may find to our disappointment that there's no single remarkable breakthrough that magically transforms this Model 70. Instead, a host of minor corrections may nibble at its troubles until the deed is done. Or tragically, we may find that all our efforts are for naught; that this rifle in its current state will never shoot worth beans.

But enough speculation and whining. Let's get on with the project, starting with a close look at the various factors.

The shooter

Could the problem be me? Dare you ask? While the Beast, with its anemic, little forend, is not an easy rifle to hold, I can only submit that I shoot my custom hunting rifles with some degree of proficiency and consistency. In my humble opinion, I'm holding the .225 suitably, using a technique I developed a few years ago for inadequate stocks on factory rifles. Since I've already confessed that every factory varmint rifle I've ever owned has shot as badly as this one, perhaps the jury is still out on my abilities.

All I can ask is that you give me the benefit of the doubt for the time being. If at the end of this exercise the Model 70 is still sick, I'll either turn the rifle over to another shooter or else bed the barreled action in a hernia-class Herter's laminated stock already inletted for the Model 70. In addition to its considerable weight, this chunk of laminated lumber has a fairly wide and flattened forend and may ride the bags well enough to be shot free recoil. If so, it'll go a long way toward either clearing or collaring me as the major flaw.

In an ongoing effort to combat the unfortunate susceptibility of the toothpick-dimensioned factory Model 70 stock to canting, a ScopLevel was closely observed throughout testing. I also watched the wind flags like a hawk. For those of you who haven't struggled lately with a truly flawed rifle, try to imagine the utter futility of waiting for the same condition in which to release a shot when the accursed gun is punching holes all over the target. Friends, at dismal accuracy levels, it becomes painfully obvious that wind is the least of your worries. Still, when one is striving to be a dutiful little scientist, one must adhere to certain test standards. I drew the line, however, at attaching a scope tube to deflect barrel mirage. There are limits to the foolishness even I will tolerate.

The scope

Could optics be the culprit? Nah. In the early days, the rifle sported a 12X Weaver. In recent testing, I used a 16X Leupold, then substituted the 36X Leupold from my bench gun. The rifle shoots as poorly with one scope as another. By the way, I checked and rechecked the rings and bases for problems, and the 16X Leupold had been epoxy bedded in the rings before testing commenced.

Action bedding

Before undertaking thick-neck experimentation, I once again bedded the action in metal-filled epoxy. After curing, I cleared the epoxy away from the front, bottom and sides of the recoil lug and drilled out the stock screw holes so the action screws don't contact wood or epoxy.

Then with the Beast shooting so poorly, I placed the rifle upside down on sandbags to check the bedding. A needle taped to the tip of the forend was positioned so its point was a few thousandths from the barrel. Tightening and loosening the front and back action screws did not noticeably change the gap between needle point and barrel, thereby suggesting an even bedding platform.

The stock

In some rifles, excessively tall magazine walls or overly long action screws may deceive us into believing the barreled action is tight in the stock when in reality it's rattling around. Such is not the case in this Model 70. The screws are firmly and positively pulling metal against wood.

Accuracy problems can also result from the bolt handle bottoming out on wood when the stock cutout is too shallow. A quick inspection indicated the downward travel of the bolt was correctly stopped by receiver metal—and not wood.

After any tampering with the barreled action, the front and rear action screws were tightened with careful consideration of the bedding surface around them, thus preventing crushing both epoxy and wood. The center action screw was only firmed-up to prevent stressing the inadequately supported middle section of the action. Incidentally, in the Beast's 1967¬79 era, I experimented extensively with various settings of all three action screws in an attempt to settle this rifle down. Didn't work.

Years ago the barrel was free-floated with wide clearance. There's no chance of barrel-stock contact.

The load

It's tough working up a load when the rifle shoots equally rotten with any powder charge. I finally settled on a load in the 3,300-3,400 fps range for no other reason than I figured this might generally be an accurate velocity level. The Speer bullet has traditionally shot a bit better than others in my factory rifles, and in a recent limited test it grouped as well as benchrest bullets in this troubled Model 70. Years ago, I determined that the Beast preferred Speer 52 grainers over 52- and 53-grain Sierra Match bullets, 55 Hornadys, 56 Winchesters, 55 Herters and 55 Noslers.

While I have forever advocated seating any bullet into the lands, at the urging of astute PS columnist Jeff Aberegg I agreed to try "jumping" the Speers. No encouragement was offered by this arrangement. In looking over ancient targets, I note that I had years ago experimented with seating depths. Moreover, in current testing, three brands of primers and two powders—H414 and IMR4064— offered no accuracy breakthroughs.

Cases were not weighed or otherwise selected, but flash holes were deburred and the primer pockets trued. Plated necks were tediously outside turned, and the unplated cases were only partially cleaned-up with the neck turner. Loaded rounds were miked to confirm the quality of the turning operation and, just for the heck of it, spun to monitor bullet runout. All this for nothing. So what's a hand-loader to do? Of a hat, it was a simple matter to inspect the assembly for old, gummed oil that might be caked about the firing pin and spring, thus slowing things down. Everything looked fine, but just to make sure, I cleaned the disassembled bolt in solvent and oiled all moving surfaces before testing began. The firing pin seems to operate freely and strike with authority.

An annoying feature on various factory bolt actions is a spring-loaded ejector so strong it bends or otherwise mutilates case necks. Years ago, I replaced the Model 70's extremely stiff ejector spring with a much weaker version that merely assists in the removal of cases by hand. If aggravation isn't enough to prompt that alteration, some theorists advise the deed be done to prevent the ejector from shoving loaded rounds to one side of the chamber, creating concentricity problems during fireforming or when using full-lengthed cases.

Judging from wear, the unlapped bolt locking lugs give the appearance of bearing rather evenly against the receiver lugs, although their contact is only partial. It's a miracle, I tell you. Similarly, the bolt face shows fairly uniform wear and slight brass plating all around that circular surface. While I'm not certain it makes a lot of difference, the fit of the bolt in the boltway is as snug as I've ever seen in any action, factory or custom. I hasten to add that my observations obviously are superficial. I have no idea if the receiver threads, the front receiver face or any critical action dimensions are acceptably true. At any rate, no glaring glitches jumped out at me.

Although there's an inclination to sneer at Winchester's crude, basic trigger arrangement, in the two Model 70's I've owned, the triggers adjusted beautifully, releasing with light pull and no perceptible creep. The pull of this trigger is a crisp, consistent 1-1/2 pounds and presents little problem in bench shooting.

In total, the Winchester action looks like a nice piece of work, but if problems persist, at some point it will be trued.

The crown

Before embarking on the thick-neck test, I prevailed upon accuracy gunsmith Garland Gilbreath to touch up the muzzle. We can rule out a flawed crown.

The bore

A prime suspect has to be the barrel interior. Because all I'm outfitted to do is squint down the bore with all the comprehension of a dog watching television, I'll tote the barrel to Shilen Rifles in Ennis, Texas, to see if I can persuade Ed or Doug Shilen to borescope the Winchester tube and measure its innards with an air-gauge. Hopefully, they'll also oblige me with a chamber cast so we can get an idea what kind of alignment the chamber has to the bore. Based on their barrel examination, I'll try my best to elicit an explanation from the Shilens as to this rifle's disorders. Although of course they won't be able to confirm the existence of metal stresses, at least these veteran barrelmakers should be able to pinpoint bore flaws that could conceivably cause our grief.

A full report on the Shilen evaluation and a titillating experiment with barrel vibrations will follow in Part 2. Also forthcoming will be results from cryogenic barrel treatment, action truing (if needed) and any other measures deemed promising, including exorcism. Kindly join us next time as we begin unraveling the Beast's mysterious malady.

The action

The Beast's excessively long action is the post-'64 variety. Because Winchester's bolt design allows the firing pm assembly to be removed at the drop

---