About 1 in 10 shots will leave the empty case trapped in the bottom groove of the charging handle between the charging handle front tab and the upper side of the bolt's locking lugs.  The ejection pin had a rough surface, which I have since polished.  I have not tried the rifle since.  Does it sound reasonable that the brass was being caught on the rough surface of the ejection pin and being ejected up rather than to the side?

Could be I suppose but...

My guess is that the extractor is dropping the empty case before it gets ejected and the next round is pushing it up.

The bolt tends to move faster with short barrels and CAR stocks which tends to aggravate this problem.  Rough or sticky chambers can also cause this problem.

I like the heavy duty extractor springs but some will put a small rubber o-ring around the outside of the extractor spring.  I believe they provide some advantage in these cases. It's best to address the other issues also but a little extra doesn't seem to hurt.

(also... make sure there is no debri like tiny brass flakes down around the ejector making it sticky)

Oldguy,  Thanks for your input.

The rifle is in standard configuration.  I have been runing experiments on the recoil motion and find this rifle to be more or less the same as the other AR15/M16 models.  The extraction time being independent upon the barrel length.  On average this comes to 1.6-1.8msec.  I do think your observation of the case being dropped and pushed up by the next round to be entirely valid as the only time the case could end up above the bolt would be when the bolt was nearly fully retracted.  I will check the chamber to see if anything is amiss there.  

I made a bolt tool, like the Sinclair, and took the bolt apart.  I could find no indication that the ejection pin was sticky in its bore.  I have since cleaned it out and will try the rifle out at the range later this week.  I did notice that the extractor spring is weaker on this bolt than on other bolts, so it may be the root of the problem.

Fred

There has been more than a few extractor springs sold that are sub-par.  

You may just want to order/install a colt M-4 spring (black insert) or an extra power wolf spring.  Either spring will solve the problem if the weak spring is at fault.

I'm not exactly sure what you mean by "extraction time"?  

I know the cyclic rate for a full auto M16 is about 600 rounds per minute or 10 rounds per second.  Because the bolt takes the same amount of time to open as it does to close and because the bolt is nearly fully retracted before the empty is extracted...

I would estimate extraction time to be more like 50msec.

oldguy,  Your estimate would be correct if the action was driven entirely by recoil.  As it is gas driven the times are not symmetrical.  Extraction is set by the gas pressure, while the chamber time is by the recoil spring.  If the rifle is operating correctly, the recoil damper will never hit the bottom of the recoil tube.  Some 'hot' rounds will cause the damper to hit the bottom and you notice it in your shoulder.  As to the overall time of a cycle, from the time that the gas drives the bolt back until the next round is chambered is about 40-50msec.  However, the cyclic rate is not determined by that alone.  One must add in the time for the trigger circuit to drop the hammer, ignition to be initiated and the bullet to pass as far down the barrel to pass high pressure gas back to the bolt.  So, the overall cycle time is much slower and ranges between 300-400msec per cycle.  My spec data on the rifle states that the full auto rate is 150/200rpm, semi-auto at 90rpm, and maximum cyclic rate of 800rpm.  I don't know how the 800rpm rate is arrived at, but the gun never gets anywhere close to that.  My test data seems to show pretty much the full auto rate at 150/200rpm.  It is surprising how much oscillation the single coil spring imposes to the cycle time of the action.  More or less, until the spring stops bouncing the next round cannot be set off.  Other rifles that use dual springs don't have this problem and are ready to fire in a fast period of time.  The dual springs having frictional losses between them and dampen out much faster.  I have instrumented the rifle quiet extensively and have both calculation and test data to support my claims.  Been doing this as part of another project.

Fred

Fred
I believe you are mistaken on a few points.
TM 05538C-23&P/2
Full auto rate of fire 800rds/min
3 round burst 150/200rds/min
semi auto 45/65rds/min

If you notice the bolt clears the bolt latch with only about 1/4" of travel before the buffer bottoms out on both the carbine and full size stock.

I defy anybody to select springs and powder charges to get the bolt to reliably clear the bolt latch and not have the buffer rebound off the end of the receiver extension. Particularly with varying amounts of dirt, grit, and powder residue in the action or with bullit weights that can vary by a factor of 2.

Good point about the lock time.

Bolt bounce is the reason there are movable weights in the buffers. There are several weights of buffers available. I understand some are more appropriate for full auto than others.

i am having the same problem right now on a parts kit beater carbine that i keep just for others to play with, i thought it may be the rings, or the extractor, the ejector feels identical to a bushy unit that works, the extractor, and pin/spring were replaced with new bushy stuff, as were the gas rings, i polished the bottom of the carrier key, and when dissassembled i saw no indication of a leak, the bolt/carrier group from any other of my rifles functions flawlessly in the rifle, ammo, and magazines are not a factor, as this is common with any mag, and all the mags, and ammo function perfectly in the other rifles, the gun has 2500 rounds fired, the bolt feels good in the carrier, what next?

On the bolt I have that is giving me problems I noticed that the extractor itself was worn where it catches the case rim.  I also noticed that the new spring was noticably stronger than the original spring.  I have since replaced the extractor and it's spring.  Perhaps with the weak spring and the worn edge on the extractor the combination could not hold onto the rim sufficiently.  I will let you know how it works out at the range.

I'm new here so forgive me if I state something stupid.
From what I'm reading, in this post and others, if a spent case is stuck in the reciever, up by the charging handle, how could it be a failure to extract? The case is out of the chamber.
It sounds to me to be a case of failure to eject. Not enough bolt travel to get case to clear the ejection port.
Is this not correct?

From what I can see, the case cannot be forced up into the receiver unless the next round is being stripped off.  In my case there is a round partway in the process of being chambered.  As such, the bolt must have traveled far enough back to allow the next cartridge to be picked up by the bolt.  Basically the empty shell is pushed up be the next round as it had not been ejected.  The only place it can go is up.

Markmcjunkins,

When the extractor spring is worn, or the claw it's self is worn, the barrel extension lugs will semi hold the extractor claw on the case rim. As the bolt/extractor clears the barrel extension, the claw losses grip on the case rim, with the carrier/key pressure driving the carrier back (without retaining the case rim), and the remaining barrel pressure lightly pushing the spent case out of the chamber.  When the bolt goes to strip a new round out of the mag on the forward stroke, the spent case which is still sitting in the action is driven up by the fresh round and is driven into the only space that it can go, the charging handle.

Note: if you have a live round driven up (not fired), then the problem is the mag that is double feeding rounds (see cheap clone mags).

Had kinda the same problem with a Carbon-15 M.97. The non-ejected brass was forced up so hard by the next stripped round that the gas tube rear was bent, and so bad that the carrier hit it and wouldn't close the bolt into battery. I replaced the gas tube and did a complete tear down and cleaning/polishing of the carrier group. I also tried different mags and filled to only  one-forth capacity to see if mag shock early release (Tipped up bullet)was a problem. In my case with a super light gun like the carbon-15 the magazine takes a lot of shock apon firing and I was using plastic mags which probably made it worse. I must have done something right because the problem went away. Even when I went back to fully filled mags. Mikke