Converting a CP2 from CO2 to High Pressure Air
I own an SMK CP2 Convertible CO2 gun. Also sold under other names (including Artemis CP2 and Diana Chaser) the gun's unique selling point is that it ships with two barrels allowing easy conversion between rifle and pistol forms.
As you might expect, it _can_ be quite a satisfying gun to shoot. Unfortunately, though, the use of CO2 as a propellant does come with drawbacks.
The pressure that drives the gun is derived from liquid CO2 in the cartridge transitioning to gas. The rate at which this conversion occurs is influenced by temperature: the lower the temperature the slower the output pressure recovers.
It's not just ambient temperature that's a factor: CO2 is a refrigerant, so as it expands, the temperature of the cartridge drops, with a knock on impact on pressure. Although the cartridge eventually warms back up, this dynamic limits the rate at which the gun can effectively be fired.
Temperature variances affect the gun's output power which, in turn, affects _accuracy_ across shots.
Putting all of this together meant that, sat outside in the British winter, I had some _quite frustrating_ sessions with the gun - to the extent that I quite strongly considered selling it on.
Instead, though, I decided to try something else.
This post talks about my experiences converting an SMK CP2 to use regulated high pressure air (HPA).
* * *
### Overview of Changes
The conversion process isn't _quite_ as simple as filling with air instead of CO2.
Regulated airguns require a plenum (an air space between the regulator and the exhaust valve). The air in the plenum sits at the regulator's output pressure and needs to contain _at least_ enough air to propel a shot.
As well as adding a plenum, I needed to be able to choose and attach a pressurised air bottle.
The most readily available vessel for high pressure air is a paintball cylinder, with the additional benefit that they normally have regulators fitted.
Conveniently, Froggys in the US make a drop block for the CP2 which not only provides a plenum, but includes a physical step down allowing the bottle to sit under the barrel:
The end of the plenum screws into the front of the valve unit (replacing the face that mates with the CO2 cannister).
One small downside of conversion is losing the ability to convert to a pistol: Not only does having a bottle on the front make it quite ungainly, but the power output would very likely come in over the UK legal limit for pistols (6ft/lb) making it a **very** illegal section 5 firearm.
* * *
### Aesthetic Changes
I ordered the drop-block from Froggys just before Christmas and they shipped on boxing day.
However, thanks to the wonders of the US Postal Service the package ended up repeatedly going in and out of the Chicago processing centre.
Knowing that I might have to wait some time, I started editing a photo of the gun to mock up how I wanted it to look once I was finished.
The gun started out looking like this
I found and ordered a 3D printed block extender on Ebay, which would theoretically allow me to move the scope forward onto the extender's picatinny rail.
I also decided that I wanted parts of the gun to be green, eventually arriving at the following design:
* * *
### Strip-down
The drop-block finally arrived on January 26th: a full month after posting.
I set about stripping the gun down to install it - thankfully, it turns out that the CP2 is _incredibly_ easy to take apart:
* Remove the CO2 stop-cap
* Remove the bolt in front of the trigger blade
* Pivot the action up slightly at the front and remove from the stock
* Put your thumb over the plug at the rear
* Use an allen key to remove the plug retaining screw
* Slide the plug out
* Pull the bolt backwards to cock the gun
* Remove the hammer spring
* Use a small allen key to remove the three barrel retaining grub screws (on the top of the front dovetail)
* Use an allen key to slacken the screw in the barrel band at the front
* Use a screw driver to loosen the front sight screw (it pins the barrel slightly)
* Slide the barrel forwards and out
* Use a small allen key to remove the screw in front of the breech
* Lift the block up and away, being careful not to lose the brass transfer port or it's 2 o-rings
* On the underside of the cylinder, there's an allen bolt in front of the trigger, remove it
You should now be able to tip the cylinder up and lightly tap it until the valve slips out of the front.
If it won't budge, use an allen key to remove the dog bolt from the top and then slide the hammer back and out (you'll need to pull the trigger to allow the front of it to pass the sear). You can then use a long implement to push the valve through.
* * *
#### Splitting The Valve
The valve consists of two halves: the (black) CO2 piercing unit and the exit valve body.
The two halves are (quite tightly) screwed together. Gripping the black half slightly in a vice was enough for me to be able to unscrew them, revealing the exhaust valve spring:
The valve itself is mechanically quite simple, it consists of a spring and a bung with a metal protrusion.
The protrusion pokes out through the front of the valve body, so when the gun is fired the hammer hits and pushes it in, allowing air to pass through before going up into the barrel.
A combination of spring and propellant pressure then recloses the valve:
The conversion doesn't require the black half of the valve body, so I bagged it up and put it into my spares box.
I then popped spring and bung back in before screwing the silver portion onto the end of the drop-block:
Before reinstalling into the cylinder, I spray painted the (other) end of the drop-block green.
This, however, turned out to be a bit of a mistake, as I ran into my first issue.
The dropblock is _very_ well machined but the gun itself is less so (something that Chinese made airguns _do_ have a bit of a reputation for). I'd assumed that the dropblock would just slide into the cylinder, but because the cylinder had been quite roughly machined, it kept fouling.
Ultimately, I had to add a little bit of gun oil to the outside of the dropblock **and** use a hammer to perform some percussive maintenace.
Although I could quite easily tap the dropblock _in_ , there was enough friction that I couldn't rotate it to correctly align the valve body with the holes in the cylinder, so I had to tap the block a little way back out, rotate and then tap back in to check alignment.
This process, of course, _ruined_ my carefully applied paint job, but eventually the dropblock was installed with the valve perfectly aligned and I was ready to reassemble.
* * *
### Measuring Output Power
With the gun re-assembled, I was ready to fire and measure output power.
Although I have a radar chronograph, I decided that it was better to measure at the muzzle: although the readings should be pretty close, if the gun were ever to be tested by the Police, it'd look pretty bad if they could say that I'd built it and failed to test properly to ensure that it was under the legal limit.
All chronograph readings in this post were collected using the same chronograph:
When the gun fires, the pellet passes over the (infra-red) LEDs, casting a shadow on the sensors in the top of the unit. The chronograph divides the distance between sensors by the time between shadows to calculate and display the pellet's velocity.
This provides a reading in Feet Per Second (FPS).
But, to assess _legality_ we need _energy_ rather than velocity. A feet per second reading can be converted to Foot Pound Energy (FPE) by taking the weight of the pellet and doing the following:
pellet_weight
FPS^2 X ---------------
450240
Rather than doing this manually every time, I created a conversion utility.
To ensure safety and consistency, all measurements were taken with the gun angled downwards, firing into ballast (specifically, a big bucket of soil).
* * *
### First Shot
I loaded 5 pellets into the mag and ran a test.
With a 15.89 grain pellet, the chronograph reported velocities of around 370 FPS, which might _sound_ quite fast.
However... it's not:
The gun's power output had dropped, by _almost half_.
* * *
### Under Pressure
Although disappointing, this _did_ make sense.
Depending on temperature, a CO2 cartridge will output somewhere between 850 - 1100psi.
A standard paintball bottle regulator is configured to output 800 - 850psi, so the bottle was actually providing _less_ pressure than the peak provided by a CO2 cartridge.
* * *
### Increasing Power
Although I didn't _necessarily_ need the gun to be sat right on the legal limit, I **did** need it to be pushing enough power to push pellets over longer ranges.
To be useable, I needed to pull the power up quite significantly.
If we look back at my firing cycle gif we can see that the gun is mechanically very simple: the hammer spring propels the hammer forward, pushes the exhaust valve open, allowing air/CO2 to vent into the barrel, propelling a pellet, before that same pressure closes the valve:
Ultimately, power output is driven by how much of the propelling gas makes into the barrel before the pellet has passed out of the muzzle.
This means that, whether we're firing air or CO2, there are a finite number of ways to increase power.
Those ways broadly fall into into one of two categories.
* * *
#### Approach 1: Increase Valve Open Time
* **Increased hammer spring strength** : this causes the hammer to hit the valve harder, knocking it further back so that it takes longer to close (this can be achieved with spring replacement or by using spacers to add preload to the spring).
* **Heavier hammer** : the hammer imparts more energy into the valve, knocking it further back
* **Weaker valve spring** : weakening the valve spring allows the valve to be pushed further back without needing to increase the energy imparted by the hammer
There are drawbacks associated with these:
* Increasing the hammer strike can add kick, which'll impact accuracy.
* Increasing spring strength makes the gun harder to cock
* Overdoing any of the above can increase hammer dwell time (how long it's sat on the valve), wasting propellant by allowing it to continue to pass after the pellet has left the gun
* * *
#### Approach 2: Increase Airflow
Once the valve has opened, the air flows out through the top of the valve, into a transfer port and then into the barrel.
The transfer port is a small brass fitting with a hole drilled through it. The factory size of this hole differs by market, but in the UK seems to be 1.8mm, while the holes in the valve and the barrel are around 3.5mm.
Widening the transfer port bore allows more gas to pass through into the barrel, with no impact on cocking difficulty or recoil.
Various aftermarket suppliers sell widened transfer ports for this gun, so getting a 4mm transfer port cost less than £10
Note, however, that because the valve and barrel holes are 3.5mm, the additional `.5mm` is essentially meaningless.
However, it's also not a perfect solution: there's very little benefit in increasing transfer port size if the valve isn't staying open for long enough to matter.
* * *
### Incremental Changes
I decided to work through incrementally, testing and measuring the impact of each change.
For convenience, I prioritised simpler changes over more complex (or harder to reverse) ones, measuring power output after each change and choosing the next step based on results.
No | Config
---|---
1 | As now - standard transfer port, uprated hammer spring, adjustable hammer (with adjuster fully wound forwards)
2 | As above, but 4mm transfer port (adding some grease to o-rings when changing)
3 | Standard hammer spring
4 | Hammer adjuster backed off by ~5 turns
5 | Hammer adjuster backed off by another ~3 turns
6 | Hammer adjuster move forward by 1 turn
7 | Factory Hammer
8 | Hammer with medium adjuster
I fired 5 shots for each, recording the following velocities:
Config 5 didn't result in any successful shots: when wound back that far, the adjuster screw protruded into the inside of the hammer and fouled on the spring guide. This meant that, although the hammer could be pulled back, it wouldn't go far enough for the sear to engage.
The biggest incremental jump came between the first config and switching to the 4mm transfer port, with velocity jumping by about 115 FPS.
However, these increases still didn't feel like _quite_ enough: although output had improved significantly, the gun was peaking at 10.3 ft/lb.
* * *
#### Weakening the Valve Spring
At this point, I decided that the next step was to either drill the valve and barrel holes out to 4mm or weaken the valve spring.
Of the two, the valve spring was more easily reversible: a replacement spring can be purchased for about £8. A replacement barrel is not only more expensive, but because of firearm controls, would need to be collected in person.
So, much to my dismay, it was time to knock the drop block back out of the cylinder.
I unscrewed the valve body and removed the spring, before using a pair of pliers to chop the end off it:
This might not seem like very much at all, but that's sort of the point: we only want to weaken the spring very slightly.
Before reassembling, it occurred to me that I should also drill the valve exit hole out to 4mm: it'd have no immediate power impact (because the barrel was still at 3.5mm) but would save me from having to remove and re-fit the drop-block again later.
The refitted valve delivered a modest power increase:
* * *
#### Drilling the Barrel
I decided that it was time to, metaphorically, bite the bullet and drill the barrel inlet out to 4mm1.
This delivered about the same level of increase as weakening the valve spring had:
I then fitted the smallest adjuster screw to the hammer and wound it right back to prioritise travel over pre-tension, delivering quite a mixed result:
The shots averaged out at about 10.3 ft/lb, with peak output being 11.1.
* * *
### First Shooting Session
I wasn't _particularly_ happy with the amount of fluctuation but decided that, before spending time tuning, I should probably take the gun to the range and check that there weren't any other issues.
It started off pretty well, shots were quite consistent and the scope was well zeroed by the time I'd gone through the first magazine.
However, part way through the second mag, something changed: shots dropped a **long** way down and right.
The possible causes that I was able to think of boiled down to two broad categories:
* Something changed in the gun, leading to a drop in power
* The gun was fine, the scope wasn't
There was no audible change, suggesting the gun's power was the same (unfortunately, I didn't have the radar chrono with me to be able to confirm).
I brought the gun home, re-gassed it and re-attached the chronograph, firing off a string of 10 shots:
The first shot was much higher: I'll come back to this in a bit.
The remaining shots, while a little inconsistent, measured at about the same power as before I'd gone to the range.
So, I decided that the scope (and/or it's fixings) had probably been to blame. The scope was, after all, mounted on the picatinny rail on top of the block extender.
Because it didn't grip the barrel as tightly as I'd expected, the block extender was partially held in place by a strip of double sided tape - I wondered whether the warmth of the sun hitting it had caused it to shift slightly.
* * *
### Tweaking the Hammer Adjuster
Whatever the cause of the aiming weirdness, I also needed to try and make the power output more consistent, so I began moving the adjuster and testing:
Although it didn't deliver the highest power, having the adjuster just 1/4 turn forward delivered the most consistent set of velocities.
* * *
### Polishing The Hammer
My testing so far had been while the hammer was coated with moly based CV grease.
Wet lubes like grease and oil allow the hammer to fly forward at a higher speed, but (particulary) for grease, that can come at a consistency cost.
Just as with CO2, the performance of grease varies depending on temperature. Over time, wet lubricants also trap grit and detritus, increasing friction.
So, I decided to remove and polish the hammer (and cylinder) before switching to a dry lubricant.
After cleaning away all traces of grease, I used autosol to polish the hammer:
I then sprayed it with resin bonded molybdenum disulphide (dry moly for short):
Once the moly had dried, I fitted the hammer back into the cylinder
Although I still think that it was the right choice, this came at a much greater power cost than I'd been expecting:
* * *
### Changing Regulator
Going back to the post-range testing graph then:
The first shot had a much higher velocity: so high in fact that it came in over the legal limit2.
Because I'd refilled the gun prior to taking these measurements, I theorised that the bottle's regulator was probably bleeding pressure during refilling. So, rather than being at the regulated level, plenum pressure had crept up.
It also occurred to me that this might offer an alternative explanation for the strange drop-off I'd experienced at the range: if the plenum pressure was able to sustain enough shots, I'd have zeroed the scope based on a (much) higher than normal power output, only to be way off once output normalised.
I considered removing and disassembling the bottle's regulator, but ultimately decided against it3.
Instead, I decided to find a new bottle with a higher pressure regulator.
Standard paintball regulators output between 800 and 850 PSI, however some higher end markers require what's known as a Super High Pressure (SHP) regulator which output 1100 PSI.
* * *
#### How Much Pressure Is Too Much
There are a couple of considerations which need to be made when increasing pressure
* Can the gun take it? Having something made of metal explode when your face is near it is, _at best_ , suboptimal and, on average, life changing.
* Will the valve work with it? The valve needs to open for the gun to fire, if the hammer can't hit the valve hard enough to overcome the increased pressure, the gun will get valve locked
In this case, I considered that the answer to both of these questions was probably "yes". Although CO2 averages about 900 psi, in particularly warm temperatures it can approach 1100 psi.
* * *
#### Test Firing
Once the new bottle arrived, I fitted it and (without making any other changes) did a chronograph run.
The power started high but progressively dropped, eventually failing to register a shot at all. I realised that the bottle had come loose in the drop block so wasn't efficiently feeding the plenum.
After screwing it in tightly, the gun was coming in quite significantly over the legal limit, averaging about 14 ft/lbs,
I tried adjusting the hammer, but rather than reducing power it slightly increased it (annoyingly also delivering _exactly_ the type of consistency that I'd otherwise want).
Next, I switched back to the factory hammer: this got the gun back below the legal limit, but by too much: it was back down towards 10 ft/lbs.
Switching to an uprated spring pulled the power back up towards the legal limit:
I was worried, though, that this could be a little _too_ close.
If the Police were ever to test the gun, they'd likely try a range of pellet weights, so I needed to be comfortable that using different pellets wouldn't push me over the limit.
I fired three different weights through and, with the exception of a single shot, all came in under the limit:
The amount of fluctuation was concerning, but the configuration also meant that there wasn't much that I could do about it: running on the factory hammer meant there wasn't a way to tune hammer strike.
* * *
#### Transfer Port Bores
To enable me to use the adjustable hammer, I decided to switch back to the factory transfer port.
Although this would probably reduce power by too much, I could then progressively drill it out (with the worst case being that I'd need to order a replacement).
As expected, the standard size led to a significant decrease in velocity. Drilling the port out to 2mm led to a small increase and bumping up to 2.5mm increased it further:
Unfortunately, my drill bit set only had .5mm progressions, so the next step up was 3mm.
I did a quick bit of maths to try and predict whether this would be too large of an increase:
Average at 2mm : 489.6
Average at 2.5mm: 539.23
--------------------------
Increase : 49.63
Average at 2.5mm: 539.23
+ 49.63
=========================
588.86
This is where I fucked up.
For a 15.89gr pellet, the legal limit would be 583.1 FPS - _less_ than my predicted outcome.
However, I decided that as my sums relied on averages, the outcome had probably been skewed up by the fluctuations in testing.
As it _felt_ close, I decided to press ahead.
The first three shots fluctuated heavily, so I took some additional ones to be sure
I'd over cooked it: the highest reading weighed in at 12.327 lb/ft.
The gun was only _marginally_ over the limit, but UK firearms law is strict and "marginally" is still potentially a prison sentence.
* * *
#### Right-Sizing the Transfer Port
I ordered a replacement transfer port and a 2.8mm HSS drill bit.
This time, the velocity sat almost exactly where I wanted it
With the gun seemingly quite consistent, I pushed a range of pellets through to ensure that nothing took it over the legal limit.
Pellet | Weight (gr)
---|---
JSB Hades | 15.89
BSA Storm | 15.43
RWS Hobby | 11.9
BSA Blackstar | 18.21
H&N Barracuda | 18.52
Bisley Long Range Gold | 14.2
Excite Spike | 15.74
Happily, they all came in under. The gun also, quite clearly, achieved the most consistent output with BSA Blackstars.
* * *
### Paint Job
Having reached the point that the gun seemed to be mechanically ready, I touched the paintjob back up and fitted a new scope.
* * *
### Range Trip Two
Happy that I was almost there, I took the gun back to the range.
It was more challenging to zero in than it had been on my previous visit - having run out of Blackstars, I was using JSB Hades and experienced quite a few fliers. As a rule, though, they were generally going where I wanted.
However, after a re-gas, the point of impact changed dramatically, dropping down and to the side **just like it had on the previous visit**.
This time, though, I _did_ have the radar chronograph with me, so was able to confirm that power output hadn't dropped off:
This time I could also rule out the scope and mounts: not only was I using a new scope, but it was mounted on the gun's dovetails rather than the aftermarket picatinny rail.
The best explanation that I could think of was that the bipod was causing issues: it mounts onto a small picatinny rail on the bottom of the barrel band:
My theory was that, if I pushed the gun forward, or put a bit more weight on it, the band pivoted slightly pulling the floating barrel down a little and affecting POI. The scratches in the paint on the back of the drop-block certainly supported the idea that the bipod had been moving.
* * *
#### Trigger Failure
I re-zeroed the scope and continued shooting, noticing as I went that velocities seemed to be higher and fluctuating more than they had before I'd refilled:
Suddenly, the gun stopped cocking: the bolt drew the hammer back but the sear didn't seem to be holding it.
Given that the power had started to fluctuate, and based on my earlier experience, I theorised that the adjuster had come loose and was now fouling on the spring guide again.
It was, sadly, time to go home. First though, I sat in the range's cafe and cheered myself up with a sausage roll and a (very good) cake.
When I got home, I stripped the gun down and, sure enough, the hammer wasn't moving far enough back for the sear to engage:
default />
I'd been right about the source of the issue.
The adjuster is held in place by a grub screw and a nylon ball - the grub screw pushes the nylon ball into the adjuster's threads providing friction whilst still allowing adjustment without needing to remove the hammer from the gun.
The nylon ball had gone awol, allowing the adjuster to turn freely: it had, presumably, unwound a little each time that I fired.
Accepting that it'd make adjustments harder, after resetting the adjuster, I wound the grub screw all the way in, locking the adjuster solid (if I have issues again, I'll likely either loctite or spot weld it in place).
I put the gun back together and, as the weather was nice, sat out using our garden range. Even after firing 100 pellets, there was no sign of the issue repeating.
I **also** found that the scope needed significant re-zeroing - essentially undoing the massive adjustment that I'd had to make at the range. As it seemed likely that it was the cause, I removed the bipod.
* * *
### Back To The Range
The gun _seemed_ to be ready so, this weekend, I took it back to the range to see whether it'd make it all the way through a shooting session.
It did! This time, there were no trigger issues and no unexpected changes in point of impact.
Over the course of about 2 hours, my radar chronograph captured 265 shots. FX Radar reported the velocity of each into my earpiece, saying reassuringly similar numbers each time.
Although the barrel is _very_ pellet fussy, the power output translated well into consistent accuracy, with the rifle comfortably able to hit a stick of chalk, end on, at 40yds.
The paintball bottle provided enough air for around 80 shots before the pressure dropped below regulator pressure and started to decline:
The heavier adjustable hammer does give the gun has a _little_ more kick than it had before, but it's not too severe of an increase.
* * *
### Final Configuration
Tying off all the loose strings, then, the result of this is a SMK CP2 converted to PCP with the following configuration
* High Pressure Air regulated to 1100 psi
* Aftermarket adjustable hammer, adjusted to 1/4 turn out
* Hammer polished and lubricated with dry moly
* Transfer port bored out to 2.8mm
* Slightly weakened exit valve spring
* Snazzy paint job
The barrel seems to particularly like BSA Blackstar hunting pellets.
Although they're a little heavier than I'd normally throw at paper targets, they do display a pretty significant level of consistency:
A max variation of 0.4 ft/sec has proven hard to beat! Admittedly, there is a greater initial variation just after filling the gun (or after it's been sat in a warm cabinet), but it settles back to about the level above (which works out to 11.3 ft/lb).
* * *
### Conclusion
Going through this process, there have been more than a few times where I've felt a bit like Father Ted knocking out dents:
Although there were multiple points that I could have declared "good enough" and stopped, I pressed on: sometimes completely invalidating the work that I'd done to get to that point.
A friend of mine remarked that it'd probably have been easier to buy a new gun, but I felt that this rather missed the point. I started on this journey not just because I wanted a more consistent gun, but because I wanted to explore and learn the impact that each tweak would have.
Obviously, I didn't want to end up with a _worse_ gun than I started with, but if the only purpose had been to get a better gun, I could just have got my HW100 out of the rack instead.
Although it's very pellet fussy, the result of this work is a remarkably consistent experience, with none of the frustration that I experienced when the gun was powered by CO2.
Time will tell, but I expect it's performance to be consistent across a range of temperatures and weather conditions.
* * *
1. A small side note for anyone doing similar - you should not drill the barrel bore beyond 4.2mm (for .22) otherwise the pellet skirt will fall in and end up getting damaged ↩
2. eeek ↩
3. Quite frankly, the idea of fiddling around with something that's protecting me from 3000PSI gave me the willies ↩
