G23 .40 Suppressor design progress

Yes, it can be legal to make a silencer. For everything Form-1, from silencer designs that are easily made, to filing forms with the BATF, to 3D modeling. Remember, you must have an approved BATF Form-1 to make a silencer. All NFA laws apply.

Moderators: mpallett, bakerjw

User avatar
mx201er
Silent But Deadly
Posts: 382
Joined: Wed Jun 11, 2008 4:24 pm
Location: NM, MT

G23 .40 Suppressor design progress

Post by mx201er » Mon Jul 16, 2012 12:38 pm

I sent my papers in on June 4th, and hopefully will have the approved form in hand in the next few months for my .40 suppressor, which will go on my G23.
So far I have collected 1.35” x .065”wall Ti tubing, along with 1.25” Ti solid round stock, and I have a couple feet of 1.5” Stainless stock.
I am planning on doing all of the baffles out of titanium; the booster, booster sleeve, and rear cap out of stainless; and front cap out of aluminum probably. Here is the design I have come up with so far, I had a few questions before I can do much more..
Image
Image
So first off, the booster. I wanted to do a sleeve around the booster for several reasons: I didn’t know if a stainless booster would wear funny against a titanium tube, when it starts to wear the sleeve and booster will be easily replaceable, and when the booster is cycling the sleeve will hold the baffle stack tight so they don’t slide in the tube. The measurements that work best for me will allow an inch long spring, and allow the booster to cycle half an inch. Think that will work? The other question I had was that a booster with an OD of .690, after threading 9/16x24 for the barrel, would leave me with just over 1/16” of material from the widest part of the threads, is that enough or should I make my booster wall thicker?
Image
Here is what I have for a baffle so far.
Image
I will be adding a port to it but was having some trouble figuring that out, since I have never seen a k-baffle first hand. I looked at the “K-Baffles step-by-step” thread ( viewtopic.php?f=10&t=17116 ), which says that he did the port at 45 degrees and size ~.31 which was a little smaller than his bore. I tried drawing a 45 degree port with .3 diameter, but it seemed to drill through the whole cone on my baffle, as seen below. Its hard to see in pictures how the port is supposed to go, I was wondering if anybody would draw it in for me or link to a cross section picture somewhere? Should I use a smaller port and change the angle to around 60 degrees?
Image

Also, would I benefit by making the flange on the front of the baffle thicker, so that I could do a deeper groove in the front? I’ll have to draw that when I get home from work to show you what I mean..

I would love to hear any ideas or thoughts, thanks guys!
I will document and share the whole build process once I am able to start the build :)
"If you are mad as hell and aren't going to take it anymore, grab your rifle and head outside.
If you're the only dumbass with a rifle screaming like a maniac, go back inside. It isn't time yet."

User avatar
Bendersquint
Industry Professional
Posts: 11359
Joined: Sat Jan 07, 2006 9:19 pm
Location: North Carolina
Contact:

Re: G23 .40 Suppressor design progress

Post by Bendersquint » Tue Jul 17, 2012 1:43 pm

Ball end mill plunged straight down, not at an angle. All the successful K Baffles use this method.

The 45 degree +/- angle will not give you the performance you are wanting.

User avatar
mx201er
Silent But Deadly
Posts: 382
Joined: Wed Jun 11, 2008 4:24 pm
Location: NM, MT

Re: G23 .40 Suppressor design progress

Post by mx201er » Tue Jul 17, 2012 6:08 pm

Bendersquint wrote:Ball end mill plunged straight down, not at an angle. All the successful K Baffles use this method.

The 45 degree +/- angle will not give you the performance you are wanting.
Thanks, its great to get info from somebody with experience. So straight down right up against the front of the k and all the way through one side (shown below), correct? Would ~.30 give good performance?

Image

I was thinking about dropping a baffle in order to be able to increase the thickness on the front of each K by 1/16", which would allow a deeper groove in the front.. I was wondering if that would be better or worse?
Image

Are there any other changes I can make to these baffles that you would recommend?

Thanks again Bendersquint
"If you are mad as hell and aren't going to take it anymore, grab your rifle and head outside.
If you're the only dumbass with a rifle screaming like a maniac, go back inside. It isn't time yet."

User avatar
Bendersquint
Industry Professional
Posts: 11359
Joined: Sat Jan 07, 2006 9:19 pm
Location: North Carolina
Contact:

Re: G23 .40 Suppressor design progress

Post by Bendersquint » Tue Jul 17, 2012 6:46 pm

mx201er wrote:
Bendersquint wrote:Ball end mill plunged straight down, not at an angle. All the successful K Baffles use this method.

The 45 degree +/- angle will not give you the performance you are wanting.
Thanks, its great to get info from somebody with experience. So straight down right up against the front of the k and all the way through one side (shown below), correct? Would ~.30 give good performance?

Image

I was thinking about dropping a baffle in order to be able to increase the thickness on the front of each K by 1/16", which would allow a deeper groove in the front.. I was wondering if that would be better or worse?
Image

Are there any other changes I can make to these baffles that you would recommend?

Thanks again Bendersquint
Give me a call and we can discuss it alot easier than trying to type it out.

User avatar
mx201er
Silent But Deadly
Posts: 382
Joined: Wed Jun 11, 2008 4:24 pm
Location: NM, MT

Re: G23 .40 Suppressor design progress

Post by mx201er » Wed Jul 18, 2012 5:08 pm

Pm sent
"If you are mad as hell and aren't going to take it anymore, grab your rifle and head outside.
If you're the only dumbass with a rifle screaming like a maniac, go back inside. It isn't time yet."

User avatar
delta9mda
Silent But Deadly
Posts: 2304
Joined: Wed Dec 07, 2005 4:32 pm
Location: miami, florida

Re: G23 .40 Suppressor design progress

Post by delta9mda » Wed Jul 18, 2012 10:46 pm

thanks bender for assisting. very lacking k baffles.
NP

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Wed Jul 18, 2012 11:00 pm

not sure I would use the ball end mill...not plunge cutting. you can use a "pot chuck" in your mill then plunge mill with a center cutting end mill with better results IMHO. Easier setup for op too.
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
Bendersquint
Industry Professional
Posts: 11359
Joined: Sat Jan 07, 2006 9:19 pm
Location: North Carolina
Contact:

Re: G23 .40 Suppressor design progress

Post by Bendersquint » Wed Jul 18, 2012 11:05 pm

jus2311 wrote:not sure I would use the ball end mill...not plunge cutting. you can use a "pot chuck" in your mill then plunge mill with a center cutting end mill with better results IMHO. Easier setup for op too.
There is a reason why almost all the manufacturers that make K baffles use a ball end mill. Just sayin.......

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Thu Jul 19, 2012 1:07 pm

plunge milling? really? any 1st year machinist student knows better than that...just saying
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Thu Jul 19, 2012 1:08 pm

maybe for the rear of the baffle for the "ramp" area for gas flow but def. not the side of the K or any other area similar in design to that
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
Bendersquint
Industry Professional
Posts: 11359
Joined: Sat Jan 07, 2006 9:19 pm
Location: North Carolina
Contact:

Re: G23 .40 Suppressor design progress

Post by Bendersquint » Thu Jul 19, 2012 1:58 pm

jus2311 wrote:maybe for the rear of the baffle for the "ramp" area for gas flow but def. not the side of the K or any other area similar in design to that
who said anything about plunge milling for the side of the k we were talking about the ports.

I was talking about the face divot and the underside divot/port opening.

Where the arrow is pointing on the face of my picture below, is a straight plunge with a ball end mill.
Image

And the plunge from the ball end mill as seen in the lower right baffle below.
Image

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Thu Jul 19, 2012 3:18 pm

I stand corrected...my apologies....on pain meds for neck surgery. Will try and pay closer attention next time I am trying to keep my mind of the pain...
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Thu Jul 19, 2012 3:20 pm

I would still not plunge mill from the face of the baffle with a ball nose end mill
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
Bendersquint
Industry Professional
Posts: 11359
Joined: Sat Jan 07, 2006 9:19 pm
Location: North Carolina
Contact:

Re: G23 .40 Suppressor design progress

Post by Bendersquint » Thu Jul 19, 2012 3:35 pm

jus2311 wrote:I would still not plunge mill from the face of the baffle with a ball nose end mill
Then how would you make the dimple on the face of the baffle?

Every single manufacturers K baffle dimple is done via a ball end mill.

User avatar
Enfield577
Silent But Deadly
Posts: 779
Joined: Mon Mar 28, 2011 7:29 pm
Location: New Zealand

Re: G23 .40 Suppressor design progress

Post by Enfield577 » Thu Jul 19, 2012 3:44 pm

jus2311 I can only say that I followed these guys advice (including plunge milling the vents) regards K baffle design and have had excellent results.

I would strongly urge you to listen to what they say, I gurantee you will be happy with the results.

If you do a search on here under my name you can find the post I did with photos of the complete process to manufacture Ks

Best of luck
Of all the things I've lost it's my mind I miss the most

rqlasl
Silent But Deadly
Posts: 182
Joined: Sat May 30, 2009 9:20 pm
Location: lititz, PA.

Re: G23 .40 Suppressor design progress

Post by rqlasl » Thu Jul 19, 2012 4:49 pm

I used a ball end mill and clamped my k's to the table of my drill press and plunge "drilled" them slowly. No problems or issues and they look pro!

User avatar
Bendersquint
Industry Professional
Posts: 11359
Joined: Sat Jan 07, 2006 9:19 pm
Location: North Carolina
Contact:

Re: G23 .40 Suppressor design progress

Post by Bendersquint » Thu Jul 19, 2012 5:01 pm

i am still interested in finding out how he would make those without plunging.

vz58
Silent Operator
Posts: 72
Joined: Fri Jul 29, 2011 11:57 am

Re: G23 .40 Suppressor design progress

Post by vz58 » Thu Jul 19, 2012 9:09 pm

Plunge milling is absolutely acceptable if its the ONLY way to get a specific shape. Coolant and feed speed are key.

User avatar
Bendersquint
Industry Professional
Posts: 11359
Joined: Sat Jan 07, 2006 9:19 pm
Location: North Carolina
Contact:

Re: G23 .40 Suppressor design progress

Post by Bendersquint » Thu Jul 19, 2012 9:29 pm

vz58 wrote:Plunge milling is absolutely acceptable if its the ONLY way to get a specific shape. Coolant and feed speed are key.
Yeah, I know, I have tried to will metal into dimpling but apparently the force is not strong enough in me so I plunge with a ball end mill.

User avatar
Enfield577
Silent But Deadly
Posts: 779
Joined: Mon Mar 28, 2011 7:29 pm
Location: New Zealand

Re: G23 .40 Suppressor design progress

Post by Enfield577 » Fri Jul 20, 2012 4:11 am

I guess if you really had an issue with plunge milling (though can't see why) you could drop into the bore at full depth then index across to form the vent.

Just means you are taking quite a deep conventional cut on the cone end vent.
Of all the things I've lost it's my mind I miss the most

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Fri Jul 20, 2012 5:55 am

other end mills can be used that would plunge mill with better performance...
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Fri Jul 20, 2012 6:00 am

will try it with different end mills and see...I listen but will throw questions/comments in here and there regardless when I have some doubts on a particular op. or tooling used.
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
Enfield577
Silent But Deadly
Posts: 779
Joined: Mon Mar 28, 2011 7:29 pm
Location: New Zealand

Re: G23 .40 Suppressor design progress

Post by Enfield577 » Fri Jul 20, 2012 6:16 am

jus2311 wrote:will try it with different end mills and see...I listen but will throw questions/comments in here and there regardless when I have some doubts on a particular op. or tooling used.
No problems with you asking and making suggestions, hey that is how we all got on but I am not sure you got the idea of what you are trying to acheive.

The ball mill creates a scollop with a rounded base so that the gas is directed across the bore and is collected (scooped up) by the other port and sent to the out side of the cone. If for example you used a standard end mill or slot drill you would not really form a good scoop.

Anyway experiment (if you are allowed) and let us know how you get on (photos are good)

Oh and I love your drawings, excellent, I play with a very old version of AutoCAD LT, not in the same league.

Cheers
Of all the things I've lost it's my mind I miss the most

User avatar
jus2311
Silent Operator
Posts: 77
Joined: Mon Jan 10, 2011 2:16 pm
Location: ALABAMA

Re: G23 .40 Suppressor design progress

Post by jus2311 » Fri Jul 20, 2012 3:29 pm

Understand the scoop being cut with the ball end...just a little confused about the area leading to the outside of the baffle to vent the gas....Delta cleared it up for me....just was looking at doing the op a different way and it didn't click in the brain housing group....ty
I.M.O.
Trooper E.A. Hill R.A.C.
10/10/41

User avatar
mx201er
Silent But Deadly
Posts: 382
Joined: Wed Jun 11, 2008 4:24 pm
Location: NM, MT

Re: G23 .40 Suppressor design progress

Post by mx201er » Fri Jul 20, 2012 8:46 pm

I know the baffle design is lacking, that is just my basic starting shape.. I have never seen a baffle in person, and was having a very hard time trying to understand the angles in the pictures I have seen. I think I finally understand how the port is supposed to go!

I have been looking at AAC Element baffles and have found some good pictures from various angles, and I am going to try to model my baffle to be similar to them.

Thanks for helping me understand, even as slow as I may be :oops:
"If you are mad as hell and aren't going to take it anymore, grab your rifle and head outside.
If you're the only dumbass with a rifle screaming like a maniac, go back inside. It isn't time yet."

Post Reply