Fixer-upper Part 2: TM AK-47

So, it didn't exactly take long for me to buy another malfunctioning second hand gun with dreams of nursing it back to health.

Patient: Tokyo Marui AK-47
Symptoms: Without trigger pull, begins firing as soon as selector is moved off from safe

This time I got myself a classic, an authentic TM AK-47. This was now the second gun I've been sold as having a metal body when that was, in fact, not the case. This one was sold as a TM, and I purposely asked the seller (who was in another town) to make sure if the body was metal. I knew Maruis typically aren't, but I was prepared to entertain the idea that it would have had its body swapped. The owner seemed a bit less knowledgeable, but as the price was quite low, I took the chance. When it arrived, though, plastic it was.

The non-metallic body issues were quite nicely alleviated, however, by the fact that out of the four metal "hi-cap magazines" that accompanied the gun, only one was actually a rattling BB bottle, and the other three were proper Marui regular size mags (holding around 70 BBs). I had prepared to resell the useless mags and get proper ones at a later point, but now there was no need.

All the original wood imitation externals and the bolt carrier had been painted black (poorly) and the paint on most of the magazines had worn down quite nicely. This immediately sent my mind on a trip to a post apocalyptic setting or a STALKER themed game. That is actually the look I'll be going for with this thing, more on that in another post later.

Day 1

The gun came with the large Tamiya connector, but with no batteries, as I purposely did not want to have the previous owner's NiMHs lying around, since I've moved on to using LiPos rather exclusively. All the batteries that I had, though, naturally had the mini Tamiya connectors, so I could not test fire the gun at all.

Since I had been toying with the idea of changing all my guns and batteries to Deans connectors anyway, this seemed like the perfect time to take my first steps in that direction.

The old connector came loose pretty nicely by just pulling on it with pliers, which saved me the trouble of having to strip the wires. After that, it was just a case of soldering the new connector on.

The helping hands are a life saver when you need to solder stuff onto other stuff, and I quickly had the connector on. Next was the battery. I decided to use an old, low power NiMH battery as a test bed, so as to not mess up a more expensive and much more useful LiPo. All went well, though (which, considering this blog, is rather unusual). Pretty much the only two things I could have screwed up, though, apart from forgetting the shrink wrap, were to either solder the connector pins on the wrong way, or worse, cut both battery wires at the same time, shorting the battery with the cutters in the process. I hear the latter especially is possibly a Bad Thing™.

Since I luckily managed to avoid both of the above, I now finally had some juice to fire up the AK for the first time. I turned the selector down from safe, and sure enough the thing started shooting right away. Something was obviously pretty wrong on the inside.

I unplugged the battery and began educating myself on how to open up the gun so I could get to the internals and figure out what was going on. Turns out it's not too hard to do, but I managed to break something in the process. I didn't unscrew the hop up adjustment tab, which when I tried to get the cover off, bent finally and broke the plastic piece of the cover that has the graduated sticker on it. No real problem, though, it's not really a functional part.

As I continued taking the gun apart, it became apparent that the/a previous owner had painted the gun with all the parts still on, as is evident from the image above showing the part of the body that was under the fire selector. The gas tube inside the front handguard also had some interesting splatter pattern going on.

Once I got the gearbox out of the gun, it began looking as though the gun had never been taken apart even once.

The dirt and grime in the picture above continued on the inside as well. What isn't shown here is the fact that the GB shell was held together with Torx head screws, not the usual Phillips ones. Guess that might have deterred a less enthusiastic mechanic, but not this one.

As I separated the gearbox shell halves, it was pretty clear something was in the wrong place. How the spring (between the spur gear and the end of the piston above) had managed to stay out of the gears so far was definitely beyond me, but I was certainly happy that it had.

A pretty clear picture of what was (most) wrong with the gun had now formed. The trigger switch return spring had broken just from the end where it's hooked on the post of the GB shell. Before figuring out what to do with it, though, I went through the other parts in case they needed cleaning or replacing.

The head of the piston was covered in a grimy, oily residue which had bits of dirt stuck on it all around. It was almost as if I was looking at an AK that had been subjected to a real steel torture test.

The extremely dirty piston had naturally also ground away at the cylinder wall, which was full of scratches and this one weird, large, deep scratch.

The light alloy sector gear (pictured from the flip side above) had its last tooth worn down cleanly, apart from a small flange that had been left behind. One more for the pages of the upcoming international bestseller Never Seen That One Before. The gun had basically short stroked itself somewhere along the way.

The dirty insides continued with the nozzle, the contents of which are seen above on the two ends of the Q-tip. I didn't dare even look inside the barrel at this point.

The trigger contacts had also certainly seen better days. The pitting is pretty extensive in my opinion. I took a small file to the metal it and filed it back down to a smooth surface on both sides.

I soaked all the metal parts in brake cleaner to get the old grease and dirt off. All of that residue is seen above. I was sure the gun would work a bit better without all that inside. Or was this in fact the mystical Marui magic dust without which the gun was ruined? Only time would tell.

As mentioned above, the trigger switch return spring had broken completely. It did look like it had also been between the gears at some point, as is evident from the bending above. Since I didn't have a replacement spring, my first thought was to see if I could re-use use this spring somehow.

I took two pairs of small pliers and bent the broken spring somewhat, so that the shape again resembled that of the original.

After a bit of fitting and cursing, my fancy makeshift repairs seemed to certainly do the trick! The switch was sliding back and forth just the way it should.

Since I hadn't located any other problems on the gun, I came to the conclusion that the always-on trigger was the only culprit, and began putting the gun back together, because I wanted to test fire it now with all the parts functioning.

Everything went pretty smoothly there, I re-greased everything and quickly tested the compression, which turned out to be surprisingly good, apart from the nozzle which had no o-ring.

Once I had the gun in one piece again, I set up the chronograph to see what kinds of results I would get.

With an 0.20g BB, the chrono read around 80 m/s (262 FPS), and with a 0.25g, 70 m/s (230 FPS), so around 0.6 J. It was obviously rather underpowered even by Marui standards, but this was to be expected with the self-short-stroking sector gear, the leaky nozzle, and what must be the original hop-up bucking. Since I had no replacement parts, I set the gun aside until I found some, particularly a new sector gear and spring.

Day 2

As luck would have it, I got some very slightly used second hand standard XYT steel gears and a new plastic piston and piston head for pretty cheap, and a completely unused second hand Vanaras green spring, which corresponds to an M110. I also happened upon a bundle of three more Marui metal regular capacity magazines, which brought my total to six. Those should certainly last a while for sure.

This time when I took the gun apart to install the parts, I also took care to unscrew the hop-up adjustment nub. It made removing the cover a whole lot easier.

Since the gun had originally also come with plastic bushings, I ventured to change those. As luck would have it, I had a set of 6mm ball bearings lying around from when I began working on the AUG and had mistakenly ordered the wrong size. It's good I have a hard time getting rid of stuff.

With the new(ish) parts in hand, I set to work. I decided to re-shim the gearbox as well while I was at it. 

After everything was set and the GB was back together, I noticed the motor had some trouble turning the stronger spring. At first I thought this might have something to do with the plainly sucky NiMH battery I was using so, encouraged by my success with earlier soldering, I decided to put a Deans connector on a stick LiPo as well.

Even with a fully charged LiPo the motor would occasionally not cycle all the way on semi-auto. Again my scavenging nature came in handy, however, as I still had the original motor from the AUG. 

Changing the motor (naturally) brought on a new, interesting problem. Since I had shimmed the gun using the Marui motor, and the AUG motor was just slightly different spec, the pinion gear now connected much too tightly with the bevel. Furthermore, the AUG motor sat ever so slightly higher than the Marui original that even turning the motor height adjustment screw all the way down was not enough.

Rather than reopen and re-shim the whole gearbox for a second time, I opted for another solution. I took the motor out of the cage, and screwed off the adjustment set screw. 

The screw head was a raised one, so I took out my trusty file set again and went to work.

After some ecstatic rubbing, I had managed to make the tip disappear. With the new flat headed adjustment screw, the motor was able to sit low enough to mesh with the gears very nicely, and I finally got the GB to cycle with the LiPo. And it did it ever so nicely. After a bit of height adjustment, the sound of the GB cycling also became very smooth, which I was definitely happy about.

The gun had originally been wired to the rear, but I didn't like the hollow sound of the stock or the fact the battery was bouncing around in it, and thus made one final move, to rewire the gun to the front. This would also allow me to use the same stick LiPo I had bought for the MP5.

After some cutting and soldering, and some more soldering, and finding a new place for the fuse, I finally had everything back in working order. Then I naturally had to chrono the gun again to see the fruits of my labour.

I was slightly disappointed to see that I was "only" getting around 92m/s (302 FPS) on a 0.25g, which is about 1.06 J, but at least the figures were on an incredibly small spread. I had a few times when two consecutive shots both read the same exact speed down to the tenth of a second. 

Despite the test results, I declared this leg of the project done at this point. After all, the gun was now fixed and was performing a whole lot better than it had when it arrived. It was ready to be taken out to the field at the first sight of opportunity. Later on, I will definitely still check and most likely change the hop-up bucking and the nozzle, which should both increase compression and will most likely get the gun to the expected power range of around 1.2 J, or around 100m/s on a 0.25g. 

In a later post I will also get back to the visual modifications I have in store to properly post-apocalyze this gun and the magazines for end-of-the-world scenarios.

Fixer-upper Part 1: G&G M4

It has been a while since I last posted. Summer is here along with its typical time sinks, so I haven't had the time to write out the following until now.

I recently decided to dive in the deep end (of the kiddy pool, I guess) of teching, and bought myself a broken second hand gun. A local Facebook marketplace had a for sale post containing an M4 that had a mysterious malfunction of not working at all. The listing said it was a metal bodied G&G with a Marui gearbox (why, I don't know) and that the owner didn't know what was wrong; "It just stopped working one day". I thought I'd buy it off since the price was pretty cheap, and I figured I'd fix it easily enough.

What follows is the first in the (likely long) series of fixer-uppers.

Patient: G&G CM16 (mostly)
Symptoms: "It doesn't work"

Day 1

The previous owner was kind enough to drive the gun to my place. The moment he unwrapped it from the bag he brought it in, my suspicions rose. For a metal gun, the body colour looked really matte and dull, and sure enough, as soon as I got my hands on it, it was clearly plastic. I gave it some thought and decided to buy it off of him anyway, and after a bit of haggling we agreed on a new, lower price. After all, I wanted a project and a challenge, and the going price was well worth the gearbox alone in my opinion.

There were some immediate shortcomings on the externals, such as a missing charging handle (evident already in the picture above), and the lack of a dummy bolt catch.

No handle, no bolt catch

The gun also had a rather groovy paint job on the handguard both on the top and the bottom. That should be easy enough to get rid of if needed, though.

Camo for a nuclear reactor, maybe?

After the previous owner had left, I immediately went to work. My first go-to was naturally the fuse. I was secretly hoping that it would be just that easy. After prying the front end apart, I was sadly disappointed. Fuse was intact. So, onwards.

I had never before taken apart an M4, or any other V2 gearbox gun, but everything was still pretty straightforward. After screwing off the base plate on the pistol grip, I noticed that the positive wire was soldered onto the motor post, while the negative had a connector. That shouldn't really matter, though, and sure enough, a quick check with the multimeter proved that the solder was fine. So I proceeded to solder it off, naturally, to get to the gearbox. If you're not familiar with M4s, the only thing holding the motor in place is the pistol grip, there is no motor cage. And one of the few things holding the gearbox to the body are the screws on the inside of the grip. I'd get to know these screws a bit too well in the process.

Once the gearbox was finally free and out, I unscrewed it and checked the insides. The shell of the gearbox might have been a Marui, but the insides were definitely G&G, at least for the gears.

I'm somehow pretty sure that's a G&G gear right there

I was hoping for a poor solder or broken wires, inside the gearbox; anything that would be easily fixed. The two things that got my attention instead were the metallic dust on the inside walls and a shim that had experienced some sort of an impossible catastrophe.

Nothing seemed to be seriously off at the first glance, apart from the metal dust

I can't begin to understand what had happened to the shim and how the gearbox had ever managed to run with it inside. Still, it wasn't stuck between the gears or anything, plus the fuse would have blown if the box had been completely jammed by it.

How exactly does one get a shim to do that?

I proceeded to clean all parts with brake cleaner, and while they were soaking, I checked out the rest of the gearbox.

A nice bath in brake cleaner

The trigger contacts didn't seem to touch all the way, which I suspected might be the problem. Stupidly I didn't test them beforehand with the multimeter, so there was no way of knowing if that had caused the malfunction, and to see if my meddling helped, I'd have to put it all back together again and test the gun.

As mentioned, the positive wire had been soldered onto the motor, and suspecting I'd have to put the motor in and get it out again more than once during the project, I hoped to find a connector somewhere. The hardware store across the street didn't have ones small enough, and it was getting late, so I ditched the idea of getting one somewhere else. The positive wire, however, had been extended at one point, and had a very interesting looking lump on it under a piece of shrink wrap, and I suspected it was the positive connector which had just had the extension soldered on.

Now what is this?

Turns out it was just a very... interesting joining of two wires. I decided to scrap the whole wire and put in a bit of brand new 16 awg silicone wire. Since I'd fixed the trigger contacts, I assumed the gun would work now.

Somewhere along the way, as I was putting the gearbox back together, I managed to cut my right index finger on a sharp edge.

So ok I wasn't quite bleeding out, but it was still a pretty crappy wound and slowed down my work

After finally managing to get the box back together, threading the wires through the lower receiver, soldering the positive wire onto the motor, screwing on the base plate and hooking it to a battery, the gun... naturally didn't work.

No other choice but to go back in. So I repeat the steps in reverse order: unsolder the wire, unscrew the grip etc. Somewhere along here I also noticed that the 16 awg wire was quite a bit larger than the original one, and decided against rewiring the negative side with it, in order not to create more problems at this point. Right now I just wanted to figure out why the damn thing didn't work. I was seriously perplexed, since everything seemed to be ok when testing all positive and negative contact points from the battery all the way to the motor with the multimeter.

Thus I began eliminating possible shimming problems. Once the box was again open, I removed everything but the gears, screwed it back together and tested the shimming by hand. Everything turned freely, so it obviously wasn't too tight. There was an excess of space, if anything. Then I even re-soldered the motor onto the wires without putting it in the gun, just to see if it turned outside the gun.

Plz work?

The motor did turn. All looked good. So, solder off, grip on, solder on, base plate on, battery in, pull trigger, it no turn. Frustration? Yeah, ask me about frustration.

Once again the soldering iron got some mileage in, as I once again removed the grip and once again opened the gearbox. This time I removed all gears except for the bevel. Then I (once again, damnit) screwed the grip back on, soldered the contact, screwed on the base plate, hooked up the battery, and pulled the trigger. Lo and behold, it turned, for once it finally turned inside the gun! The motor did make some god-awful grinding noises, but at least the bloody thing finally did what it was supposed to. I could finally go to bed tired and somewhat happy.

Day 2

Since last day's endeavours had led me to a place where the motor would actually turn when the trigger was pulled, I thought I was finally onto something. I just needed to find out why it didn't work with all the gears in and I'd be set. The now familiar unscrewing, unsoldering, testing, and cursing begin anew. As I opened up the gearbox this time, I notice there was again an accumulation of fine metal dust in the gearbox.

That can't be right

At this point I realized the grinding noises came from exactly that, grinding. The pinion was sitting so low that its edge was carving out a ditch in the gearbox, so in short the motor height was all sorts of wrong.

How do you spell gouge?

How the previous owner had managed to run the gun with the motor height this off, I'll never know. After cleaning the dust out of the gearbox shell and all the parts, I set out to get the motor height right. With only half of the gearbox in use, I screwed the grip on, and put in the motor and base plate.

It's a lot easier to see things this way

After turning the adjustment screw for quite a bit, I finally got the pinion high enough to come out of the hole in the gearbox and properly mesh with the bevel gear.

Motor height screw sits pretty high up there now

Once I'd figured out the correct height, I took off the grip again. I could now put the motor in the grip, replace the base plate, and see the correct pinion height just by pushing the grip on the gearbox half. This way I re-shimmed the bevel until it sat very nicely on the pinion. After that I went on to re-shim the rest of the gearbox according to the bevel height, and with just the gears in, put the box back together. After one more episode of the Screw-and-Solder Show, I finally got the motor to turn all the gears, and with a very smooth noise to boot. Happy times!

But I was not out of the woods yet. I had a seriously loose main spring (less than M90) lying around, which I decided to use for now just to see if the whole package would run and cycle with all the parts inside. After lubing everything up, I began to put the box back together. At this point I felt it might be wise to also check the compression of the cylinder and nozzle, since I was going to chrono the gun now that it worked. The cylinder was super tight, but the nozzle had no o-ring, and thus leaked quite a bit. Since I had no replacement anyway, I didn't really bother with it, and just slapped all the parts back together.

Before putting the box in the gun, I wanted to make sure it worked. Once all the screwing, and the threading of the wires, and the fitting of the motor, and the soldering, and the baseplate tightening had been once again done, I proceeded to test fire the gearbox with the grip on. I manipulated the selector plate by hand, just as I'd done at least thirty times before this when testing the box. I first moved it to semi, and fired off a few shots. Worked like a charm. Then, flipping it all the way to full-auto, the selector plate comes loose, and THE CUTOFF LEVER SPRING FLIES OFF!

There is something missing here

Finding a spring with the dimensions of a third of a matchstick is easy in a large room full of stuff: Y/N? You get two guesses.

I had already had the gun working, it was ready to be put back together and used right away! And now, because of one single spring going AWOL, it was useless. I'm not sure how many letter A's to put in DAAAAAAAAAMMMMMNNNN to properly depict my emotions at the time. Oh, hey, it was again seriously late and I had to turn in for the day. With such warm and fuzzy feelings, who wouldn't sleep tight?

Day 3

New day, fresh eyes, optimistic attitude. After a bit of searching online, I found out that cutoff lever springs are not generally sold separately, but as a part of packages that contain all the springs used in a gearbox apart from the main spring. They're also priced somewhat steeply. As I wanted to get the gun back in working order as soon as possible, I wanted a different solution. A bit more searching revealed that several disposable lighters contain springs that are very nicely sized to be used in gearboxes. I just happened to have a Colt (how fitting) lighter nearby, which I promptly proceeded to destroy. It indeed had two very nice springs inside, the thinner of which looked just perfect for the cutoff.

Thank you for the gift you have bestowed upon me, Disposable Lighter God!

I took some measurements from the gearbox, and nipped off a fitting bit of spring. I had to use tweezers to get the spring in place, and you betcha they slipped in the final phase, firing YET ANOTHER cutoff lever spring into extinction.

Older and wiser, I cut off a second piece of the spring, set the gearbox inside a plastic bag, and worked in there with the tweezers, so as not to lose any more parts. Of course the spring went on like nothing this time around.

Once I was confident enough to remove the box from the bag, I could admire this perfect replacement spring

A couple of test shots proved my replacement to be functional, so I proceed to do the merry-screw-and-solder-around once more. With the gearbox in the gun, I reassembled everything and set up a chrono so I could finally see how the gun performed.

Finally I was in a position to take pride in my work with glee! I fired off some shots into the chronograph. Smile was suddenly replaced by disgust, as I looked at the wonderful reading of 58 m/s (190  FPS). There was obviously a huge airleak somewhere.

During the test shots, I also found the gun doing a weird thing where the motor would turn for a bit, but stop mid cycle even with the trigger held down. At other times it worked just perfectly. I assumed the negative wire on the motor was a bit loose, since it had felt a bit wobbly when installing. During this test I also noticed that my King Arms M4 midcaps didn't feed very well on the gun unless they were held forward by hand. The mag well was simply too wide from front to back, and the magazine follower would push the mag to a slanted position where it no longer fed properly.

As I went to adjust the hop-up to zero for a couple more test shots, I noticed that the face of the hop-up chamber did not actually mate with the gearbox face at all, even when the gun was assembled. The gearbox opening was simply too small for the hop-up chamber. There was obviously more work to be done. I left the gun that day, because it was again rather late and I couldn't decide whether to enlarge the gearbox hole or file down the hop-up chamber.

Day4

Once again I started the teching day with screwing off a certain base plate. As I unsoldered the motor for the nth time, I began to wonder how much of the metal shavings the pinion had been taking off might have ended up in the motor itself. Once the motor was off, I decided to take the brake cleaner can to the motor, but sadly the nozzle on it was way too short to accurately spray it. Luckily, the soldering iron was still warm, and I had some plastic q-tips that I'd used earlier, and with the help of a toothpick I managed to make a functioning extension nozzle for the cleaner.

A perfect, DIY, extensomatic nozzle

What came out of the motor looked pretty harsh. I sprayed the motor several times, and each and every time I got more metal shavings for my troubles. No wonder it had been running a bit erratically.

This should maybe not be in an electric motor

Once cleaned and sparingly lubed, I set the motor aside, with plans to take apart the gearbox one more time, file the gearbox hole to fit the hop-up (that's what I ended up deciding), and install the stronger, original spring.

I first tried to use the rotary tool to enlarge the gearbox opening, but soon realized that the bit I had on was too fine to make any proper progress, and the other ones were too gritty to progress slowly enough. So I went with a round hand file. Luckily there was only a little to take off to fit the hop-up. Once the hop-up could properly seat in place, I no longer wondered about the huge air leak from before, since the nozzle had been left about 2 mm from the hop-up bucking lips when in the front position.

Installing the original spring went, surprisingly, without any major screw-ups, so I finally had the gearbox in working order, and properly shimmed and lubed up. Before final assembly, I still wanted to check to see if the motor was now running better after the clean-up. I hooked up the motor outside of the gun, and attached the battery. Pulling the trigger did nothing.

No to worry, I though, maybe the old NiMH battery had just drained enough overnight or something. So I tried another battery, a fairly new LiPo. Still nothing. That was now too weird, so I went to test the LiPo in another gun. It worked just fine. What the hell was going on? I retried the LiPo on the M4, and this time it works the motor flawlessly. The motor was also running very nice and smooth now, so the clean-up obviously worked. I tried a third battery, a higher voltage one, so I could hear if the motor ran smoothly on higher voltages as well. Somehow, an 11.1 V LiPo that had been charged two days ago and had just sat unused would not turn the motor. But it did light a bulb in my head.

Only now, four days into the project, did I take a closer look at the mini Tamiya connector on the gun. The negative lead was a bit wobbly, and it seems it had occasionally protruded ever so slightly out of the connector, which would obviously result in no current passing through. As I pushed the lead back into the connector and retried all the previous batteries, each worked without a problem.

Whether or not the problem had been a loose lead in the connector all this time, I will never know, but it does seem plausible. At least the motor was now working all the time as it should, so I could finally wrap the thing up. A final round of screwing and soldering later I finally had the gun in working order.

As the first of two final adjustments, I made a quick fix to the magwell out of self adhesive furniture felt pad, since I would need the gun as a loaner just a couple days later.

Now this must be tactical felt, right?

The bit of felt set on the rear wall of the mag well worked very nicely right away, and the magazines began to feed without issues. For once something on this project worked the first time around. To further make my day, I also happened upon the previously escaped, Colt-cannibalized cutoff spring at this point. I guess it didn't want to run away from home after all.

The second cobbled-together thing was to take some steel wire, wrap it around itself for strength, and make a hook on one end and tighten the between the carrying handle and the rail. This made for a makeshift clasp for the ejection port cover, which would otherwise get in the way of the mag release button.

Ready to operate

After days of frustration, head scratching and blood sacrifice, I finally had a well functioning G&G M4, which chronoed a nice, steady 103 m/s (337 fps), and moneywise cost next to nothing. Couple of days later it functioned without a problem at the weekend skirmish I took it to as a loaner (I did play couple of rounds with it myself, of course. How could I not after all this effort?).

The next time I get a gun that doesn't shoot, however, I'll be sure to check the damn battery connector first.