Lane

Here is my brass from extractor testing. I stripped the bolt carrier again and tested a new o-ring. It has the same issue, and cannot be assembled by hand with it inside. The o-ring that came with the bolt originally has crush marks on it. I think reducing the volume of the o-ring by about 25% would allow it to assemble and function as intended. I'll photograph how it actually goes together with a standard o-ring when I try a reduced o-ring. I didn't fell like cleaning up to take pictures of that yesterday. I did polish out the extractor edges quite a bit so the next rounds shouldn't have anything more than magazine and bolt marks; these already look pretty good in that regard. About to dive in to the shot timer code and I made a final to-do list last night. Still need to dig out some more AA battery holders for the new hardware build, and make something happen. My goal is to get two or three sensors in that design. New parts are still a few days out; but those don't really have much bearing on this build. I put some temporary BUIS on the 7.62x39 because of the broken dot sight. The rear takedown pin feels a lot better, but I'll pull it out later to measure and fill the dummy detent with JB weld when I get some new ones. I need to make one more special pin for another polymer lower that is on the way. Turns out the scales I ordered a while back aren't going to show up. eBay seller has about a 36% satisfaction rating all the sudden. I'm sure I'll get my money back; but I won't have scales to modify coming this week either. Don't really care about that data short term, as I still need to build a stupid long optical chrono to do any actual testing.

This I get. This is why I think it's a great idea... I live in a state where duct taping a knife to my gun makes it an illegal assault weapon. As if bringing a mounted knife to a distance gun fight is really going to be a problem. Go figure. I assume you want one that flips out based on your initial inquiry? That Geissele mount does look bulky, but shouldn't be all that heavy. Since I can't have a bayonet here; I haven't researched this at all. I don't know that you'll find a lot of people that want to weld a hand rail. You could modify some rail attachment parts and bolt them on for mounting something unusual on the rail as opposed to using the gas block. Your milage may vary; but I am certainly curious to see what you end up with.

Fired off two rounds to inspect brass, and the extractor was still gouging a bit. I stripped the bolt carrier group which was immaculately clean to find out that the extractor hadn't worn in at all. I couldn't move it with any amount of pressure, and it still looked wrong. I wasn't even able to clip a cartridge in to the bolt with my hand. I'm honestly not sure why it was working as well as it was given the situation. I searched around for other people having problems with this; and sure enough two reviews of the bolt only part mentioned exactly the same problem I was having. Apparently the o-ring inside occupies too much volume and keeps the extractor from moving appreciably (at all in my case). Simply removing the o-ring was enough to resolve that problem, and now the extractor pin can be removed/replaced with finger pressure. Two more test rounds showed nicer brass; but there is still burs on both ends of the extractor. I'll strip it again and polish those... One of the reviews mentioned grinding it like a shark tooth, I think I'll be fine with 1000 grit sandpaper. I'll try the other o-rings I have laying around and see if any of them fit better while I'm at it. I think you're saying I'll be ok here with those Speer .311" projectiles. I shoot regular off the shelf 7.62x39, but I plan to keep this gun as a brass queen. No cheap steel case ammo in this particular build. The Lee loading data says .311" projectiles. The specs I can find on 7.62x39 says 7.92mm projectile, which is .3118"; some call this .312". I'm just not sure why so many vendors sell .310" for the 7.62x39... I measured the projectile in the Winchester rounds I have, and they appear to be .311". I gather that shooting .308" projectiles would be a little like a hotdog in a hallway; launching bullets with less pressure and decreased accuracy.

Sounds like you got this covered sir. I toyed with those numbers a while back, and decided on a long barrel by accident. I don't regret it for a minute; and that last build balances perfectly in my left hand, (right handed shooter). You can mount rail attachments if needed; or just bolt on the component you need. I don't shy away from drilling or cutting a rail to make something work; that what makes it unique.

Runs great; and the brass looks wonderful. I just want to make sure everything is working perfectly. I saw someones out of battery picture the other day and I'm not interested in that kind of testing/experience... My split times were 0.5 seconds longer than with my .308 build; even when charging manually. The .308 moves so much more on the shoulder that I have to take time to fire again. The 7.62x39 only takes that much time to charge a new round. There were some audio file images posted a while back that show them side by side. You can see a slight tick between shots when I charged the next one on this straight pull build.

Why not a full (or near full) barrel length handrail; and a bayonet on the bottom of the rail? Not sure if you wanted a bi-pod too. Welcome indeed. What's your reasoning on the 20" barrel? I have a long barrel on my 7.62x39 build.

I laughed out loud at this too. One guy that hunts on the adjacent property sent me pictures of a bear that was in my back yard during hunting season (live update because he has a cellular connection). I didn't go looking for the bear; but he said it ripped his camera off the tree and pounded it. I would totally piss on the bear if I actually climbed a tree.

Cleaned up the screen and loaded a few raw values to finish out the timer display for now. I fixed some menu issues early last night too; I thought some things looked funny with the "-" before numeric values. Since they weren't negative numbers; I've begun looking at other options. In hindsight I should have picked any one of the other half dozen graphics libraries that works with this display. They all have similar commands, so replacing that in the future won't be much effort. The 7.62x39 build is pretty well sealed up; I ordered a sight for it the other day which should arrive soon (that dot thing on it is totally broken after using on an over-gassed .223). I'm a bit scared to take the bolt apart again after the first time; but I need to check on that when I clean it. It doesn't get very dirty without a gas system, which is good for the polymer lower in my opinion. I do also need to take out that rear takedown pin one more time to take off a hair more on the rear face. I'll be sure to measure it when I do and test it with a few more uppers. I ordered projectiles the other day (along with front takedown pins); and I'm still not 100% sure they are what I wanted. I got Speer .311" in 150 grain; some people use .310" for reloading though? My plan is to compare against some 150 grain .308 Winchester loads down the road. I've been pouring over this next timer hardware build and realized a few things. A simpler buck only DC/DC converter is more efficient. A single AA battery might be able to get decent runtime with that change (but two AA will break 12 hours non-stop easily). I need to start testing multiple audio sensors soon; and do some shot profiling at the same time. It's not usually a problem; but there are times I can trigger a shot when I don't mean to by tossing the unit around, or playing with it. That should be easy to filter out if I try to detect the profile (over width) of the shot audio peaks. Fumbling the unit should produce much wider spreads than a real gunshot (and have a vastly different drop-off rate). I realized that the buzzer delay is actually due to program file read and not the display writes. That is unnecessary, in that it was already loaded to display in the menu screen. If I undo that second file read things should be back to normal in terms of run time accuracy. I do need to animate the file loading when switching between files though, since it takes up to four full seconds to open/read/close the file before parsing the data back into the menu update. I'm thinking about scrolling "- - - - " vs " - - - -" across the filename should do alright to inform the user that it's working on it.

The shot timer screen is looking a little better; but those display memory writes sure do take some time. Looks like there is going to be some kind of unintentional pre-delay in the buzzer if I can't figure out how to weave it better. The first time display write takes forever; and I dare not try to run this is slow timing mode (animating the display in real time)... I bet it would be banging off of 5ms or so in terms of resolution. It's all good though; I'm dedicated to this one time display write, and only updating the screen when something changes. The big T in the upper left is the real feature of this device in my opinion. It lets you count shots or time. Of course the hybrid H mode allows you to get really granular if you wish. Looks like shot peak display works; but it's not calibrated to the system threshold yet. I still need to write markers as the shots go down so you can see them all on the graph. The other thing to do yet is calculate the position of the Shots/Time/Hybrid display graph. One hundred divided by the setting will return a multiplier for the graph; I think I have to split it lenghwise for the and/or hybrid mode. Still a bit of debug code on screen, and adjusting positions of elements; but I can see the end in my mind. I need to revisit the peak detection code and see if it actually works. A .308 should easily trigger the device and climb to peak in a few more time slices, though I doubt something quiet would. At the moment that requires a laptop outside hardwired to the device; and it was snowing yesterday. Intel assembly was a physics elective where I went to college. I had an absolute blast writing a 32 bit calculator in 16 bit assembly. I haven't even begun to optimize this at the bit shift level. Hell, I ignore all sorts of optimizations when writing a huge body of code like this. The 2.0 version of this will really blow your mind...

Getting close now. I've loaded the program data into memory and parsed it into the shot timer. It's still a bit messy but much of the information is actually written to the display chaotically for the time being. There are a number of font changes to do in the process, and some other drawing operations. I'm still trying to figure out how to write all those updates in a clean way (less memory writes to the display buffer). The pencil drawing is my first sketch of how the timing screen should be displaying the relevant information. The one last thing to do is tie those timing variables in, and enumerate through the stations automatically in "Continuous" mode. The menu now shows a bit more information as well. Beside Go! is the filename of the selected program. Beside data is the shot file size. Next to Program is the currently selected program number. I think I can tie that to a push and twist of the rotary encoder so you don't have to leave the main menu to change it; but I'm still thinking through how to show and set the selected station number as well. There can be breaks inside a course; and one might want to select an individual station out of the complete course without running the whole thing. Maybe moving the program number to the Go! selection, and setting the station through the other menu option with some visual cues added in could work...

I'm a bit curious how many rounds per second this is. You know; research...

That's the other way around actually. I'm taking off the BACK side of the pin, so the lower is actually too long. I thought it might be a shrinkage issue originally; but it seems like it's the other way around. It's cold here, so there is no reason to believe this problem gets better in the summer time (or in the desert). Is there some reverse shrinkage with polymer; or is this overcompensation during manufacturing? When casting metal there is a shrinkage calculation; I presume someone did that wrong in this case. I am curious to try other polymer lowers though; and perhaps casting my own one day.

I finally got through the bulk of this last bit of code. I can write, read, and parse the course programming files through the web interface. The last bit is to rewrite the shot timing display screen, and link all those variables to relevant code functions. That will honestly be easy in comparison to debugging the dynamically generated HTML I had been working on. I might be able to save some RAM and increase station count if I remove every last useless space character from the html, and make it a giant block of text. Seems to work in the limited situations I've tested so far. The new threshold code is now modeled in to the course programming. The default will be "0" for system shot threshold. One would only need to modify the variable if they wanted to shoot quiet firearms. Any other single digit value will divide the global threshold setting; so setting a 1 would be 10% of the system threshold, up through 9 for 90%. Beyond that the number will set the threshold explicitly for that station up through 1024 which is the maximum. I'm about to sketch out the new shot timing display screen based on these variables. There isn't much to add as most of the programming variables are either delays in the code, or endpoints for existing shot timing code. I ended up devising a very elegant means of parsing the data file which didn't require any kind of array so far (near zero memory footprint). I think the shot timer will use a single 1D array; but it won't impact memory in any significant way.

Here are a few new observations. If I'm going to do this right; I'll need another field in the programming data. Select a range of preset thresholds; or a specific value. I know some people shoot multiple gun courses; and that's the only way this is going to work for everything. Shotgun seems to be in the etc category so far; but I haven't tested everything explicitly. Those are my .22lr times from earlier. There were actually two missed shots the first magazine; and one fail to fire the second. I didn't even bother to record the 10th shot on the latter. I did clean the FTF off; manually load, and fire it afterward. Pretty much what I expect from this last bit of cheap ammo. Those CCI quiets just jam the gun and it wasn't worth the effort to record quite yet. I'll get into that after I can select a per station threshold.

Got back into the shot timer code after thinking more about today's testing. I was aware that cheap .22lr rounds don't always sound the same, but there was a pretty wide spread in variation between the loud and quiet shot peaks recorded. I'm not sure if the code is efficient enough yet; but I did revisit the shot peak detection. It currently runs for about 25ms after hearing the first peak over the volume threshold, and collects audio input as fast as possible. If any of those samples is louder than the initial, it is sent back along with a new time. The main function writes an update in that case; and proceeds with the rest of a short delay to make up for echo. My goal here is to pick up the initial shot peak for a tighter timing accuracy; while still allowing for multiple calibers without any threshold adjustment. The extra decimal points are displayed in the shot timing file, along with a dummy text identifier. It looks like the last serious thing to do is link the programming data into the shot timer itself. I've been putting it off because it's a huge number of variables to store; which will most certainly have an effect on my overall memory footprint. I doubt I can get away with a file read between each station; I expect I'll need fill a two dimensional array with the entire course programming up front. Of course, after posting this photo I see an error in the file display routine (one missing digit when scrolled to a certain position). I'll have to record some longer shot strings to debug that since I just nuked my file again. I think the last column (peak volume) will be dropped from the file outside of debug mode. There is a better way to show that to the user; and it's mostly useless data inside the shot file. I've been using it to keep an eye where my threshold needs to be set up until now.

Fixed. I modified the detent channel on a front takedown pin and added a new detent. I didn't see a good way to fill the existing detent; though epoxy might work. At the moment it has a dummy detent that won't release the upper. I took a bit of meat off the one face to account for the apparent shortness of the lower between takedown pins. I had to grind out the slot a bit further than what is shown to clear the upper when pulled all the way out. Now the pin sticks out the left side enough to operate with human fingers (as intended). The buffer retainer is now functioning as it should. It took quite a bit of bit spinning to open it up enough. The bit went straight in the hole though; as others had noticed before me, it needed a good de-burring for some reason.

I warmed myself up today and went out for timer testing. Glad I waited because it was actually relatively nice out for winter. Picked up all but one of the .22lr rounds; sadly not a single one of the CCI quiet rounds triggered my current threshold. That would need to be adjusted a bit to pickup anything less than a .22lr supersonic round, but it's a simple setting to change. I nailed about 200ms between rounds very consistently since it barely moves off target when firing. I roughed out a CAD drawing for a new rear takedown pin only to realize that the front takedown pin looks almost exactly like what I want. I plan to swap one of those into my polymer lower and hopefully address the buffer retainer hole at the same time.

That's not a bad call; if I'm going to do it, might as well do it right. I have a tilting vise on my drill press already (simply can't load the lower in that way). I can drill 1.125" wooden dowel at 6º though. Then put it in the buffer tube threads; line it up, and go. Still have to find a piece of aluminum rod to turn a new rear takedown pin. I think I can make a full diameter pin, and then take a crescent shaped side off with sandpaper. I want at least 1/4" in additional length, and I might as well add a right side grip if I'm going to make my own.

Yes; you are correct; numbers are hard. Looks like a drill bit could ride up against the top of the area where the buffer tube threads in. Probably be able to twist it by hand until I get the buffer retainer to fit in like it should. Seems like some others have had issues with the rear takedown pin location too. Most people egg out those holes which I'm not ready to do yet. I would be happier to find a long replacement pin and sand down the diameter slightly. I looked around and only found "long pins" that had a protrusions on the right side to grab. Nothing I found has a longer shaft to make it stick out through the wider polymer lower. I guess if I really had to I could make one on the lathe out of 6061.

My excitement about this polymer80 has waned considerably. The milling all went off without a hitch, but drilling the selector holes was problematic. In hindsight I should have drilled a number of pilot holes first; but there was nothing in the instructions about that. The drill bit punched through the right side making a bow tie shaped hole that I had to mill out before I could finish drilling. The red plastic also tended to clog up the bit on those larger holes. The fire control group all went in smoothly and functions perfectly. The bolt catch is stiff in the slot; but the real issue was the buffer retainer got stuck and I can't get it back out. I put some oil down the hole, and took it out to the workshop to chill down. The mag well is very tight with my ACS 7.62x39 magazine. I was surprised to find out that the rear takedown pin can't be operated without a tool. Furthermore; the lower itself almost seems warped, and I can see light between the upper and lower. The rear takedown also seems to be just a little bit off front to back; and it came from the factory that way. I have a pistol grip on this build for the time being; but I find it more comfortable to operate with a slanted grip. That feels more like a rifle stock; and having to charge for each shot, I naturally want to grab for a swept back grip. I skipped testing today because it was unpleasant outside. No idea what the wind chills were earlier, but even a few minutes was too much; checking this evening it said -14 degrees F. I do still need to pick out optics for this build. My next lower is aluminum for a 300 blackout build; I simply need a barrel and gas tube to get that done. I assume someone sells longer rear takedown pins that work in polymer lowers... Here are a couple shots pre and post stock adjustment along side my .22lr. I guess I could have a collapsable stock on this build; but I don't know that it matters much. I always shoot fully extended for comfort; and these both have fixed installs. Update: I got the buffer retainer pin to pull back up with needle nose pliers finally. I don't think it sees enough abuse during regular operation to fix itself though. I suspect I will need to remove it later and bore out the hole a bit more. As I recall it is drilled at something like a 4 degree angle, so I'll need to be careful how I do that.

Do you have any more in the works? I've been reading some of those threads. I wish I could do one tomorrow; but I live by the rule of law. Sunrise here. I'm going to birth a new one.

I'm really pumped up by this one day lower build. I've had this sitting around for months; I don't know why I waited so long. It's quiet hours right now, but I'm planning to go out shooting with my morning coffee. It's going to be very windy; but still worth testing the shot timer with quiet .22lr, and whatever else I feel like dragging out there. Certainly going to put a few more 7.62x39 rounds through as well. Looks like the Pelican 1750 long is the way to go for rail travel; it has an interior size of 50.5" × 13.5" × 5.3". That should fit about 50 pounds of rifles without too much trouble. I don't see why I can't finish a .260 before fall. I don't know what I should bring really; having never been to a fall shoot before... Got back in to the shot timer code and just about to add decimal points to the recorded times, and increase the resolution recorded to include the tenth of a ms. My to do list is getting quite short. I need to add a few routines to help me profile my code, and look at memory in real time. I was just reading about a cycle count variable I wasn't aware of; and that can help with timing accuracy by counting instructions instead of microseconds. I had my timer in the truck with me the other day while running errands and noticed it rolled over the microsecond counter. Apparently that happens after 71 minutes, and will write negative time values to the file if it does. I'm not sure the best way to get around that besides a timed sleep function. My only concern is; what if it's running a course at that time? I could do filtering to look for that; but I would also have to count how many times the counter rolled over for very long sessions (like running all day). Seems like a pointless exercise; but there are a few cases where I can see this happening at something like a fall shoot. I did a search earlier and found a few threads with polymer80 .308s. Nobody complained; but I didn't see any long term reporting either. I don't plan to use polymer lowers for anything violent; and I would worry about my current .308 on a polymer lower. On the AR-15 side; the .22lr CMMG doesn't even use the buffer tube, so that seems like a good choice. My 7.62x39 is also a very tame gun; so why not? I haven't weighed it yet; but it feels great even with an aluminum lower. The balance is spot on right now, so I'm curious to see how it feels with something lighter. I've dug out the LPK, buffer tube, and buttstock to finish this lower. Only a few more hours and I can assemble it at sunrise.

Alright; I will do that. I'm well aware there are things I don't know, and haven't tried. While I don't like the giant lips on the jig; I'm really am impressed with this polymer lower. It mills so fast, smooth, and quietly; I'm well over 75% done. I think I could finish one in half an hour if I tried to do it in one sitting. Honestly going to need at least two more of these; though I'm still not sure how I feel about using one on a 308. I don't recall seeing anyone else around here with a polymer80 Warrhogg build.

I've been eyeballing those pistols lowers too. From a legal perspective I would need to get a pistol permit, which takes forever here and isn't exactly guaranteed. I did read somewhere that you can't be denied for a premises only permit though; which is totally fine if I'm making my own. I would really like to do a 1911 from scratch too. I have a friend in NM, and really enjoy being the southwest. I'm started to get more comfortable with the twisted gun laws here; though I would prefer they were repealed for upstate residents at least. Besides having no chance to play with a full-auto, I don't see that I'm missing all that much. There are a couple of oddballs though; like I can't build an Uzi with a 16" barrel, because of where the magazine lives.

Decided to take a short break on the timer software. I will probably revisit it later tonight; but in the mean time I want to get started on another lower. The jig that comes with a polymer80 is much taller than what I normally use (but the lower's profile doesn't fit in my regular side plates). The mill bit it comes with it also useless in my setup because of it's length. I'm impressed at how easy this material is to mill (it's my first polymer lower), but it does smell a bit strange. I think I'm better than half done and have only been at it for about 10 minutes. I upgraded the button location markers with some temperature sensitive "blueberry" paint; it turns bright blue when warm. Also ordered three doppler radar modules to test; not because I'm sure it will work to measure bullet speed, but it's something to try. One of them in particular looks like it might be made to function with an antenna modification.