The back story is long and what you see here is the culmination of a lot of other people's work together with what I've learned from the other cans I've built. There are some general truths about silencers that I've verified and can repeat with some degree of confidence, so first about the theory and principles...
Basic premise #1 is that to achieve the ultimate suppression, you have the tools of internal volume, baffle efficiency, baffle count/spacing, and surface area. The thermal coefficient of the materials plays into the surface area equation a little. Of these things, baffle efficiency is the most important factor. What does baffle efficiency mean? It means how well each baffle can disturb the laminar flow through the hole in the middle of the can, i.e. make the central jet of gas following (and passing) the bullet deflect away from the central axis and thus become trapped by the baffles. Also, efficiency means how effective the baffles are at retaining the gas and not letting it leak out the bullet aperture. "Clipping", or modifications to the baffle apertures which make these subtle yet all-important gas disturbances happen, is where the real secret to good suppression lies. It is a simple matter to part away expanding gas from the stream and compartmentalize it (as in simple cones or even flat washers), but another matter entirely to try and break up the central stream and create a self-propagating pressure restriction at each baffle orifice.
Silencers are for the most part a "mature technology", with most of the current evolution involving decreased weight and baffle efficiency. Creating more efficient baffles is where the Voodoo, research dollars, and legendary designers come in to play. Also, the "drama", which is legendary within the industry as well. The problem with engineering for the DIY maker is that we aren't permitted to "experiment" with our designs and try different things on one build. An ATF 5320.1 form allows us to make a silencer ONE time, in ONE configuration, and if it sucks we're stuck with it, no Mulligans. Any modifications, alterations, or even repairs of a completed Form 1 silencer are only allowed to be performed by licensed manufacturers who have paid the extra "Special Operations Tax" which entitles those manufacturers to gunsmith on the Title II stuff. This is because a completed DIY can becomes an NFA item upon completion and Those SOTs who are willing to recore and repair silencers are few and far between. The rest of us have little option but to keep buying tax stamps and keep refining each build until we understand how to do it well. Fortunately, the internet has brought together some silencer cranks who have a similar mindset to us here and freely share information, particularly good information, and a few rare breeds of SOTs (who, as manufacturers, can make prototypes and experiment with designs to their heart's content) that are willing to allow their ideas and information to be public domain. Knowledge in this industry costs a lot of money to gain, and most of the SOTs and designers guard their secrets like their lives depend on it, because in a way it does, at least income does. I say all this to give a better understanding of the limitations and challenges all of us "form 1-ers" face.
Back to the designs, spacing, and "funny notches". Like any form of engineering undertaking, one must first consider the purpose. Purpose will determine the material and strength requirements, size and weight limitations, host firearm requirements (such as semi-auto, rate of fire, rifle or pistol, type of action, and priorities of weight, cost, durability, suppression ability, and size. For this can, my requirements were maximum practical lifespan and maximum suppression from full-power .308 Winchester, on a bolt-action rifle, with a 20" barrel. The industry has established that a safety margin of about 2:1 is achievable in this instance with a Grade 9, seamless titanium tube having an ID of 1.375" and at least .060" wall thickness. End caps must have at least 20 tpi threads and the thread engagement must be at least .3" in length. 24 tpi is better. Maximum operating temperature will be about 600F before the safety margin begins to drop off significantly. To prolong the life of the silencer, I chose to use a muzzle brake as both a mount and as a set of sacrificial baffles, being that the first baffles take the brunt of the heat, pressure, and ablative effect of particulates. Legally, muzzle brakes are not silencer parts and are thus user-replaceable, so we use that to our advantage as much as possible by making silencers that mount on muzzle brakes instead of directly threading them onto our muzzles. "Brake mounts" also help mitigate muzzle crown wear, thread wear, and offer a quick/easy way to swap a single silencer between multiple hosts. The length of the muzzle brake determines the "blast chamber" length, or the amount of volume available for the first chamber and the placement of the first baffle in the stack, called a "blast baffle". Minimizing the blast chamber volume is a priority for minimizing the factor of "first round pop", particularly with subsonic loads, where the first shot encounters a silencer filled with cool, dense, oxygen-rich air that can create a secondary burn of remaining propellant coming into the can. Many sub-sonic loads necessarily use a powder which isn't fully consumed within the barrel (semi-autos are the main example needing such loads to function sometimes), and this combustion makes a further "pop" inside the suppressor on the initial shot. Creating a blast chamber that is both small and generates a lot of turbulence mitigates FRP. What seems to work out as a general guideline is a blast chamber about 1.5 to 2" long, and a blast baffle located about 3/8" to 5/16" from the end of the brake. Supersonic loads with much higher pressure, where FRP is less relevant anyway, are typically optimized for bigger chambers as this makes the can more durable, lighter, and less complex. This can is an experiment in optimizing both subsonic and supersonic ammunition, so I had to make a small blast chamber with a baffle close to the muzzle brake, but since this increases stress on the baffle and subsequent few baffles with supersonic loads, I had to make them out of a more durable material than titanium, with 17-4 precipitation-hardened stainless steel being a good choice for those of us not equipped to deal with nickel alloys like Inconel.
I'm going to break that off for a minute and discuss baffle shape. Generally, supersonic, high-pressure rifle silencers are most effective with baffles that are steep and match the shape of the expanding gas plume, or about 60° included angle. While this also works well for splitting subsonic gas plumes, other angles and shapes are better. Sub-sonic gas plumes can be diverted, confused, contained, and have their heat absorbed better if they are made to deflect multiple times within a chamber before the pressure equalizes and peaks, thus a 45 or 50° cone, and in many instances a frusto-shaped cone, work extremely well to part and deflect/stir the gas in each chamber. An SOT acquaintance of mine developed this 50°, necked baffle style as an effective compromise with both subsonic and supersonic loads, very handy for a .30-caliber rifle silencer. I took it upon myself to include two, 60° true cone baffles immediately after the blast baffle to be most efficient with supersonic gas, and the following baffles to work with the lower-pressure loads. With the close spacing of all the baffles, this isn't a compromise for subsonic loads, but makes the can heavier than it needs to be for supersonics, particularly because the baffles at 50° need to be extra-thick to withstand the forces of supersonic gas impinging on them more directly than on a 60° cone whose surfaces are more parallel to a supersonic expanding gas plume.
On to spacing. Determining spacing is pure Voodoo and experimentation. If we can't experiment by changing parts, we can engineer our Form 1 cans to be internally modular so we can switch up the order of the pieces and see what works. A lot of this has been done and an arrangement of .75 to .8" tip-to-tip spacing of the first three baffles followed by a .5-.625" spacing of the rest is a very good "rule of thumb" for making a quiet can. Supersonic or subsonic, it all works although the supers can get by with fewer baffles. The last chamber needs a lot of space if one wishes to avoid point of impact shift, which I'll get to next.
Silencers have an annoying tendency to shift the trajectory of the bullet slightly. Some are worse about this than others, and often sights must be adjusted back and forth when shooting with and without a silencer. POI shift is due to several things which have finally been isolated to a great degree and I'll enumerate here: Asymmetrical baffle clipping techniques, parallel-sided bore apertures, and excessive exit chamber pressure. Many silencer manufacturers produce cans with clip styles which divert the gas stream off-center only one way, believing that symmetrical clips will cancel each other out. Best suppression with these single-clip baffles typically is achieved with all the clips lined up instead of alternating, so the bullet is pushed progressively off course more and more the same direction by each baffle. POI shift is generally toward the clips, and away from the high-pressure point on the side opposite the clip. Baffles and end caps having apertures with parallel holes through their thickness instead of tapered holes also focus high pressure gas around the bullet as the bullet passes the aperture, and any slight defect in concentricity or natural bullet yaw from the muzzle exit will affect the equality of pressure on the bullet, thus inducing flight path deflections.
The business that developed the clip style I'm using here came up with it while trying to maximize suppression, yet minimize POI shift. The clip shape is a combination of u-notch and step clip, and are clipped on two sides opposite each other, and when combined with a relatively large exit chamber and short bore apertures, yields zero POI shift. Very clever. Three clips don't work well for some reason, but the combination u and step work to split the gas plume and create a high pressure blockage during the initial shock wave (gas hits the four right-angle flats spaced equally around the aperture which "cut" the gas stream across the bore hole) and the u-notches create cross-bore jets as the chamber pressure deflates and gas attempts to find its way out the hole to the next chamber. Exit chamber from tip of baffle aperture to exit end cap hole should be about 3/4" to 1" in distance. There are also some tricks with varying bore size to help minimize POI shift and maximize suppression, such as making the exit hole a little bigger than the holes in the rest of the baffles to reduce the un-corking effect of gas flow suddenly being vented to the atmosphere, and increasing aperture size for the first few baffles to let more volume into the middle part of the can where overall pressure can drop somewhat and make use of the volume aspect of suppression while making the clips more effective farther down the stack. A certain amount of gas pressure is required to make the clips function, and if the can soaks up too much pressure in the first chambers, the following baffles won't work as efficiently as they could. Sort of counter-intuitive, but that's the way it works. Basically we want to charge the can up with pressure, make use of the volume, and feed the last baffles enough gas to maximize the clip efficiency (make a stopper). Double-clipping all the baffles helps "charge" the can in stages rather than all at once, and we can control a lot of the little details of clips and apertures to tune a silencer for a certain load or barrel length.
The particular shape of the 50° necked baffles is ideal for the double clip method, providing lots of meat where needed for the step clips and coned all the way to the orifice on the back side so that there is minimal pressure influence on the passing bullet. Titanium automotive valve spring retainers are excellent for re-shaping for this clip style.
There's a lot more to all this than I've mentioned here, in spite of rambling on about it, but should give a little insight into why I built this one the way I did. Just a note, I only u-clipped the two 60° cones because I wanted more flow for the subsonic gas through the first part of the can, otherwise the following baffle clips wouldn't be seeing enough pressure to operate correctly. With high-pressure, they will still part and store gas without deflecting too much and again feed gas to the remaining baffles in a controlled fashion.