Grizzley makes a gunsmith lathe for like $6500? I think that's probably going to be a decent machine.  

If not that, some old engine lathe for like $3500, but those are damn hard to find now that a bunch of people drove them to Mexico by the truckload for a decade or so when Mexico was buying them up.

Most of the old engine lathes that filled shop space and needed to be liquidated as CNC equipment replaced them have been sold and finding one now is hard.

By the way, it's not just shit for power, it's shit for rigidity also.  All that "we use castings so our 300lb machine is comparable to a ton" is a load of bullshit.  The light machines will vibrate like hell, break and wear out tools, and make crappy surface finishes.  They will require light cuts and like River rat said, they'll have you for $100 worth of time doing $25 worth of work.  It's not worth the trouble messing with crappy little import machines.

Little stuff like that has no spindle bore- like.75" - 7/8" spindle bore capacity.  That's horrible.  You can't even chuck a piece of 1" stock without it being saw cut to a small length.

I had a CNC hobby lathe once and I found it very marginally adequate for turning aluminum, but parting and grooving operations were ridiculous- it would vibrate and make tons of noise and sometimes even snap brand new carbide inserts from Kennemetal even with aluminum, unless the feed was perfect and flood coolant kept aimed at the tool 100% of the time.  The flood coolant of course is not enclosed so you're going to be wearing it, and it's going to be on the floor and the walls and ceiling.  Basically without running carbide it was worthless to me, and it would turn stainless but a parting operation would literally stop the spindle, so unless you want to cut off parts with the hack saw and face them on the back side, you're screwed if you plan to cut anything harder than 6061.  In stainless it would make a .025" cut on the diameter, and in aluminum .050.   So you'd make a part on it in about 7 minutes that a HAAS (A relatively entry level industry CNC turning center) would spit out in 45 seconds.  And the HASS part would be held to possibly even .0005" and the part it would make would be held to about .010" even possibly on the diameter where it's most critical.  

The backlash on the machines is bad.  Like .010" being fairly normal, and that means precision is going to drop off, as well as chatter and broken bits become more common, basically people who are going to use a crappy hobby lathe to make a silencer would be better off buying one from a company in business making quality silencers.  

If you don't have $3500 and time to wait for a great buy on some old iron in good running order, or $6500 for a grizzley gunsmith lathe, or $4500-$6500 for a running but aged out CNC turning center, my recommendation is to just not get involved.  Unfortunately making good parts is going to be damn near impossible on a cheap lathe.  

You'll have to at best compromise on materials and use something like brass instead of 17-4 PH, and at worst you're going to become frustrated and compromise on part dimensions and end up with a unit that looks like a 75% grade on a technical school shop project.  

The worst part is after the $750-$1250 hobby lathe and $750-$3500 in fixturing equipment, measuring tools, and cutting tools, you'll have spent enough money to buy good silencers and you'll have crap silencers to show for it.  

By comparison a buddy of mine has an old crappy American lathe that's huge and it will take .300" on the diameter in steel without chattering.  Cutting stainless steel like carving soap and moving the handles like an etch-a-sketch not really even trying to get proper feeds.   The HAAS will do .300 in aluminum and .150 in stainless probably.  That guy's family owns a company that makes pumping stations and that lathe was used when they bought it 50 years ago, and has turned tens of thousands of impellers to size, and it's ugly as sin, but it's a workhorse, despite abuses of dozens of people not formally trained to operate lathes, it soldiers on making better work than a brand new hobby lathe and cutting like a hobby lathe couldn't dream of cutting.  

The hobby lathe was really built to "optimistically cut" aluminum.  It's more aimed at plastic, delrin, and lighter materials.  The larger machines mentioned were designed to cut steels.  There is a huge difference there.

A piece of PVC pipe filled with aluminum screen sounds good.  That's not to say it's quality and utility compares to a Gemtech Outback II.  If you have 50 hours to make a can, you might be able to do something with similar tolerances (might).  You'd have to pay a minimum anodizing fee of like $150 to get the one off anodized though... so all I'm saying is being a tinkerer isn't always promoting more value.  

I once competed three 80% Ar receivers on a mini mill.  But that took like a week of my work, and in the end the result wasn't industry quality, though the receivers did work.  Factor even $12 an hour which for machining parts is like minimum wage, and I invested $480 worth of time, $350 worth of tooling, and 3 $405 worth of 80% receivers in the project so they were $411 not industry quality receivers when I finished it.  

Unless you have the lathe, you need  various chucks (maybe collets) some measuring tools (mics, depth mic, caliper, thread mics) a wide assortment of cutting tools etc, by the time you have what you need to make a good go at it, you're into it in thousands of dollars, not just the $1275 Jet mini lathe which is already expensive.  

Can someone make his own silencer?  Of course he can.  In the same way I could make my own $411 AR receiver.

Now if somebody died and left you a bridgeport and a south bend lathe and $7500-$10K in tooling and you know how
to use it- absolutely you can make your own silencer AND save money doing it.  

Without training, you probably wouldn't know how to grind a cutting tool so that would be a discouraging deal... but if you had indexable cutters, an understanding of the basic theory of operation, and a high tolerance for poor quality designs and garage gunsmithing, you might be perfectly happy with results.  In the 60's-70's silencer companies were using cap screws to attach end caps on silencers.  That's a less than optimal way to do that, and it's as ugly as it is relatively weak compared to threads or circumferential welding, but it nevertheless has a proven track record for having worked and it is within the skills of someone relatively inexperienced as a machinist.  

I guess you could build one with less tooling and equipment by designing it around that lack of equipment.  

With a lathe, two pieces of tubing, a bar of round stock, one outside turning tool, one parting tool, three drill bits, two taps, 8-15 cap screws, an allen wrench, and a hack saw, and some judicious application of elbow grease hacking it all up, you could make a silencer.  So I guess I was just thinking in terms of making something market competitive, and not in terms of replicating some 14" long design from the 70's.