I've seen loadings for 32-20 that were listed as 'very quiet' and 'nearly silent, check barrel before next shot!'
I'll see if I can google them up
something similar in 38 special
My requirement was to develop one load which I could stock in quantity for use in any .38 Special revolver or "Cowboy Rifle." In my testing with current production Alliant Bullseye power, the lowest charge with the Remington, factory-swged, soft lead 158-gr. semi-wadcutter bullet which would exit the barrel every time (100 rounds, 50 each in rifle and revolver) was 1.2 grains of Bullseye, but only when used with a 3mm diameter enlarged flash hole.
This is about as large as you can go in a case which uses the small size (.175" / 4.45mm) primer. This charge didn't always exist when using a the unmodified 0.078"-0.082" (1.98-2.08mm) flash hole, 2 bullets out of 50 rounds fired in the 18" Marlin carbine lodged within 3" of the muzzle.
The load was "plinking accurate in the basement at 25 feet, but at 50 yards in the Marlin groups strung vertically over a foot! NOT acceptable! Velocity averaged 300 f .p .s . in the 6" Ruger Security Six and 480 fps in the 18" Marlin, but velocity standard Sd was over 100, which set off alarm bells! Had I taken the velocities first I would have quit sooner... Just plain lucky I guess!
All test loads used the Lee Factory Crimp die to hold bullets securely in the case against telescoping from compression of the tubular magazine spring in the Marlin 1894 carbine. I have found that this also improves velocity uniformity, as it seems to prevent the primer blast from dislodging the bullet before powder ignition in light loads. Of course, the powder must be suitable. I have limited myself to Alliant Bullseye so far, because I have it on hand and it would appear satisfactory.
A charge of 2 grains of Bullseye very satisfactory, but much louder (500 f.p.s. in the revolver ) . Point of impact was 6" low at 25 yards, producing a loose 4" group with noticeable projectile yaw from the 18-3/4" twist . Velocities were more uniform and entirely acceptable. Noise-wise in the revolver it was more quiet than a factory-loaded target 148-gr. wadcutter.
In the 18" Marlin it was fairly quiet, producing a satisfying "thunk" rather than a crack, rather like firing standard velocity .22 LR match ammunitionfrom a short-barreled sporting rifle. Not, however, like the "Cat's Sneeze" equivalent of Eley Tenex fired from a long barreled target rifle. The velocityaveraged about 700 f.p.s., point of impact was useful for plainking with iron sights at 25-50 yards plinking without changing the sights from my regular carbine zero for 158-gr. factory .357 Magnum softpoints at 100 yards. I got 2" round groups at 25 yards, larger than I expect with the best loads, but reasonable.
As the powder charge was increased above 2 grains velocities became more uniform and accuracy improved. Using Alliant Bullseye of current manufacture it takes 3.8 grains with the Remington 158-gr. lead SWC to approximate the velocity of factory standard velocity lead-bullet .38 Special loads averaging 800 +/- 20 f.p.s. in a 6" revolver and 950 +/- 20 in the Marlin.
Normal extreme spread of these loads with iron sights is "one inch per ten" (yards) in a handgun and "one inch per 25" in the Marlin, out to 100 yards. I used Norma 158-gr. lead RN factory ammunition as a benchmark. It is repeatable at these velocity ranges and gives 1.5" ten-shot groups at 25 yards from the 6" Ruger Security Six revolver and approximately the same at 50 yards from the Marlin carbine.
At longer ranges I can reliably keep 10 out of 12 shots on a 12" steel gong at 100 yards from the 6" Ruger revolver and expoect the same when plinking at 200 yards with my Marlin Cowboy Carbine. I do not change the sights, but use "Tennessee elevation" (centering the "ghost image " of the gong in my other eye between bottom of front sight bead and the sight dovetail base behind the slimmed portion of the front sight blade).
I didn't load any test increments between 1.2 grains and 2 grains of Bullseye. I need to do that, will do so and report. The 2 grain loads I tested here had unmodified, standard flash holes. My next step is to load samples at 1.5, 1.7 and 2.0 grains, with enlarged flash holes and see how they do.