AR-15 Buffer Weight, Spring Rate and Gas Length That Cycle Together
A carbine buffer is 3.0 oz, H is 3.8, H2 is 4.6, H3 is 5.4 and a rifle buffer is about 5.0 oz, and one step of roughly 0.8 oz moves ejection about one hour on the clock. But buffer mass is the last thing to change, not the first — barrel length, gas system length and port diameter set the problem, and a buffer only trims the answer. Brass at 4 to 5 o'clock unsuppressed and 3 to 4 o'clock suppressed means the rifle is in tune.
Standard carbine buffer, 3.0 oz. H, 3.8 oz. H2, 4.6 oz. H3, 5.4 oz. Rifle buffer, about 5.0 oz. One step between those is roughly 0.8 oz and moves the ejection pattern approximately one hour on the clock.
That is the chart everybody wants, and on its own it will fix perhaps a third of the rifles that need fixing. The reason is that a buffer does not set the gas curve — barrel length, gas system length and port diameter do, and those three get specced by different vendors who never publish what happens when you combine them. A buffer trims. It cannot correct a mismatch.
So this page gives you all four together: the exact masses, the spring data nobody prints, the gas length and port diameter that belong with each barrel length, and an ejection-pattern table that converts an o'clock reading into an ounce figure. Start with the gas table, not the buffer table.
What in tune means here, measured
A correctly gassed AR-15 does four things, and all four have to be true at once. It locks the bolt back on an empty magazine every time, with any ammunition in its intended power range. It ejects to a consistent clock position. It strips and chambers the next round without the carrier arriving at the rear of the tube hard enough to be audible over the report. And it does all of that dirty, cold and dry, not just clean and freshly lubricated.
Reference condition for every recommendation on this page: 8.00 lb rifle as-fired, 5.56 NATO 55 gr at 3,100 fps from a 16 in barrel, 59°F at sea level, mil-spec bolt carrier group at 11.6 oz, standard buffer tube. Change any of those and the tune moves — a lightweight carrier at 8.5 oz behaves like about one buffer step lighter.
Ejection pattern is the measurement because it is the only one you can take without instrumentation. It is a proxy for carrier velocity and a decent one, but it is still a proxy — a rifle can eject at a beautiful 4 o'clock and be over-gassed if the extractor is doing too much of the work. Treat it as the first reading, not the only one.
Buffer masses, exactly
Nominal masses vary by a tenth of an ounce between makers, and a few makers use their own naming that does not map to the mil-spec ladder. Weigh yours on a reloading scale if the tune is close — a kitchen scale will not resolve 0.8 oz reliably at this range.
Buffer Nominal mass (oz) Internal weights Tube type Typical application Carbine 3.0 3 steel Carbine (7.6 in) 16 in carbine gas, unsuppressed, mil-spec BCG H (H1) 3.8 1 tungsten, 2 steel Carbine The best general-purpose default for most builds H2 4.6 – 4.7 2 tungsten, 1 steel Carbine Suppressed carbine gas, or short barrels on pistol gas H3 5.4 3 tungsten Carbine Heavily over-gassed builds with no adjustable block XH / H4 6.5 3 tungsten plus powder Carbine Last resort. If you need this, fix the gas instead 9mm blowback 5.4 – 5.8 Solid steel body Carbine Dedicated 9mm uppers only — no gas system involved Rifle (A2) 5.0 – 5.2 5 steel Rifle (9.7 in) 20 in rifle gas, fixed stock A5H0 3.8 1 tungsten, 4 steel A5 (carbine tube, rifle spring) A5 baseline, mid-length gas A5H1 4.6 2 tungsten, 3 steel A5 A5 with carbine gas, unsuppressed A5H2 5.4 3 tungsten, 2 steel A5 A5 with carbine gas, suppressed A5H3 6.5 4 tungsten, 1 steel A5 Short barrels, suppressed, on the A5 system Identify what you have by mass, not by the label stamped on it. If a build is one hour off on the ejection clock, move one row up or down this table; if it is three hours off, the problem is in the gas table below and a buffer will not reach it. The A5 system deserves a note because it is the most under-used fix on this list. It runs a rifle-length spring in a slightly longer carbine tube with a longer buffer, giving the carrier more travel and a flatter spring rate across that travel. The result is a noticeably softer impulse and a wider tuning window than the carbine system offers, at the cost of a proprietary tube and a receiver extension swap. For a rifle that has to run both suppressed and unsuppressed, it is the better architecture.
Springs: the part nobody changes
Buffer springs are consumable. A carbine spring cycled ten thousand times has taken a set, lost free length and is delivering measurably less force at the end of its stroke than it did new. That shows up as short-stroking which appears gradually and gets blamed on ammunition.
Measure free length against the specs below. A carbine spring under 10.0 in or a rifle spring under 12.5 in has taken enough set to replace, and springs cost less than a box of ammunition.
Spring Wire dia (in) Free length (in) Coils Approx load, in-battery → fully compressed (lb) Notes Mil-spec carbine 0.0435 – 0.0450 10.5 – 11.25 37 – 39 4 – 5 → 10 – 12 The baseline. Replace at 10.0 in free length Mil-spec rifle 0.0435 – 0.0450 12.75 – 13.25 41 – 43 5 – 6 → 13 – 15 Rifle tube and A5 tube only Extra-power carbine 0.045 – 0.047 11.5 – 12.0 39 – 41 5 – 7 → 13 – 16 Roughly one buffer step on an over-gassed build Reduced-power carbine 0.042 – 0.0435 10.25 – 10.75 36 – 38 3.5 – 4.5 → 9 – 10 For genuinely under-gassed short barrels Flatwire carbine flat section 12.0 – 13.0 n/a 5 – 6 → 12 – 14 Quieter, longer service life, no spring twang 9mm blowback heavy 0.048 10.75 – 11.5 38 – 40 6 – 8 → 15 – 18 Pair with a 5.4 oz or heavier 9mm buffer Measure your spring's free length before changing anything else — a tired spring mimics under-gassing exactly. If you are chasing an over-gassed rifle, an extra-power spring is cheaper than a buffer and gets you about the same one-step correction without adding reciprocating mass. Spring rate and buffer mass are not interchangeable even though both slow the carrier. Mass changes the carrier's momentum through the whole cycle, including the forward stroke, so a heavy buffer can cause feeding problems on a weak magazine spring. A stiffer spring resists mostly at the end of the rearward stroke and pushes harder on the return. If a rifle short-strokes on the return — bolt stopping just shy of battery — you want less spring, not less buffer.
Barrel length, gas length and port diameter that belong together
Gas length is the distance from chamber to gas port. Dwell is what is left between the port and the muzzle. Longer gas length means lower port pressure and a gentler impulse, and it also means less dwell, so there is a floor below which the system cannot cycle. Port diameters below are the ranges commonly cut in 5.56 barrels.
Barrel Gas system Gas length (in) Port dia, common range Buffer, unsuppressed Buffer, suppressed 7.5 in Pistol 4.0 .073 – .081 H2 (4.6 oz) H3 (5.4 oz) 8.5 – 10.5 in Pistol 4.0 .070 – .078 H (3.8 oz) H2 (4.6 oz) 10.5 in Carbine 7.0 .070 – .076 Carbine (3.0 oz) H (3.8 oz) 11.5 in Carbine 7.0 .070 – .076 H (3.8 oz) H2 (4.6 oz) 12.5 in Carbine 7.0 .068 – .073 H (3.8 oz) H2 (4.6 oz) 14.5 in Carbine 7.0 .0625 – .070 H (3.8 oz) H2 (4.6 oz) 14.5 in Mid-length 9.0 .076 – .082 Carbine (3.0 oz) H (3.8 oz) 16 in Carbine 7.0 .0625 – .070 H (3.8 oz) H2 (4.6 oz) 16 in Mid-length 9.0 .076 – .082 Carbine (3.0 oz) H (3.8 oz) 18 in Mid-length 9.0 .076 – .082 Carbine (3.0 oz) H (3.8 oz) 18 in Rifle 12.0 .089 – .0995 Rifle (5.0 oz) Rifle (5.0 oz) 20 in Rifle 12.0 .0890 – .0935 Rifle (5.0 oz) Rifle plus extra-power spring Start here, not at the buffer table. Find your barrel length and gas system, check your port against the range — a pin gauge set costs about what two magazines do — and only then pick a buffer from the last two columns. If your port sits outside the range shown, no buffer will bring the rifle into tune. Mid-length gas on a 16 in barrel is the single best value decision in an AR-15 build. It lowers port pressure meaningfully against carbine gas at the same barrel length, which softens the impulse, reduces carrier velocity and widens the tuning window enough that one buffer will run the rifle suppressed and unsuppressed. Carbine gas on a 16 in barrel is a legacy of 14.5 in military barrels, and there is no engineering reason to prefer it.
Two hard constraints whatever the configuration. Leave at least 1.5 in of barrel past the gas port. And never pair rifle-length gas with a barrel under 18 in — the dwell is not there, and the rifle will short-stroke regardless of spring or buffer.
The selection solver
Six inputs: barrel length, gas system length, caliber, gas port diameter where you know it, suppressed or unsuppressed, and ammunition power factor. It returns a starting buffer mass in ounces and a spring specification.
Power factor is the input people skip. Commercial steel-case 55 gr runs meaningfully softer than 62 gr NATO-spec, and a rifle tuned on one will sit a step out on the other. Enter the ammunition you actually shoot most, and expect everything else to land an hour off on the clock.
The output is a starting point, not a verdict. Fit it, shoot twenty rounds, read the brass, and adjust from the ejection table below. Any tuning guidance that does not end in a range trip is guessing.
- Know your port diameter if you can. Pin gauges in 0.001 in steps from .060 to .100 will tell you in a minute and take all the guesswork out of the process.
- Weigh your bolt carrier group. Mil-spec is about 11.6 oz; lightweight carriers run 8.5 to 10 oz and behave like a lighter buffer.
- Note whether you have an adjustable gas block. If you do, tune there first and leave the buffer at H.
- Test cold and dirty, not just clean. A rifle that only cycles when freshly lubricated is not tuned, it is lucky.
Ejection pattern as a measurement
Stand the rifle up, fire five rounds, and note where the brass lands relative to the muzzle as twelve o'clock. Do it on a flat surface where you can see the pile. Five rounds is a pattern; one round is noise.
Brass lands at Diagnosis Most likely cause First correction Approx mass change 12:00 – 1:00 Severely over-gassed, or bolt-over-base Oversized port, or a suppressor on an unsuppressed tune Adjustable gas block. Rule out bolt-over-base first +1.6 to +2.4 oz if no gas adjustment 1:00 – 2:00 Over-gassed Suppressor fitted, or hot ammunition One buffer step up plus an extra-power spring +0.8 to +1.6 oz 2:00 – 3:00 Slightly over-gassed Ammunition on the hot side of spec One buffer step up +0.8 oz 3:00 – 4:00 In tune, suppressed — Nothing 0 4:00 – 5:00 In tune, unsuppressed — Nothing 0 5:00 – 6:00 Slightly under-gassed Heavy buffer, tired spring, or soft ammunition One buffer step down, or a new spring −0.8 oz 6:00 – 7:00 Under-gassed Undersized or misaligned port, carbon-fouled block Check gas block alignment before touching the buffer −0.8 to −1.6 oz Fails to lock back, brass at your feet Badly under-gassed Gas block clocked wrong, or gas tube not seated Pull the handguard and inspect. Do not change the buffer n/a No consistent pattern Not a gas problem Weak extractor spring, worn ejector, mixed ammunition Replace extractor spring and insert, then retest n/a Read your pattern in the first column and apply only the first correction listed — one change at a time, five rounds, re-read. The mass column converts at roughly 0.8 oz per hour of clock, and it stops being reliable beyond two hours in either direction, which is the point at which the gas system is your problem. Two cautions on this method, because it gets over-trusted. Ejection angle is influenced by extractor tension, ejector spring strength and case material as well as by carrier velocity, so steel-case ammunition often lands an hour further forward than brass out of the same rifle. And a rifle ejecting at 3 o'clock with a suppressor and 5 o'clock without one is not two rifles out of tune — it is one correctly tuned rifle, working in both conditions.
The 0.8 oz per hour figure is a working rule rather than a measured constant. It holds reasonably well within one or two hours of a correct tune, and progressively less well outside that.
Fix order when it will not cycle
Work down this list. Most of it costs nothing, and the parts that cost money are at the bottom for a reason.
- Check the gas block alignment. A block rotated a few degrees off, or slid forward of the port, is the most common cause of a rifle that short-strokes from new. Pull the handguard and look.
- Check the gas tube is seated in the carrier key and that the key screws are staked. Loose key screws leak gas and produce intermittent short-stroking that comes and goes.
- Measure the spring's free length. Under 10.0 in on a carbine spring, replace it.
- Check for carbon lock in the gas block, particularly on a suppressed rifle. A partially blocked port behaves exactly like an undersized one.
- Replace the extractor spring and insert. Cheap, five minutes, and it eliminates the most common cause of erratic ejection that gets misdiagnosed as gas.
- Only now change the buffer, one step at a time, five rounds between changes.
- If none of that works, measure the port with pin gauges and compare against the table above. A port outside spec needs a gunsmith or a new barrel, not a heavier buffer.
Why the buffer is usually the wrong first fix
A buffer changes reciprocating mass. That is a blunt instrument applied to the end of a chain whose first link is a hole drilled in a barrel, and it can only correct a problem that was small to begin with.
Consider a 16 in barrel with a .0935 in rifle-gas port cut into it by mistake instead of a .0700 carbine port. Port area is nearly double. No buffer in production absorbs that — you would be running an H3 and still battering the receiver, and the rifle would beat itself apart over a few thousand rounds while technically functioning. The answer there is an adjustable gas block or a new barrel, and any amount of buffer swapping is money spent avoiding that conclusion.
The corollary is worth stating plainly. An adjustable gas block on a build you intend to suppress is not a luxury item. It is the component that makes the rest of the tune possible, and it costs about what two buffers cost when you buy them in sequence. Fit one at build time and the buffer question mostly answers itself at H.