Well, trigger mods always seem to need tweaking.... The lighter spring I used was too light, so I went back to the original one, at reduced preload.... This produced an "acceptable" trigger, but I really need a spring in between the two.... Using a longer, weaker spring with increased preload will give a more constant trigger pull over the 1st stage travel.... which is what I wanted.... but the light spring I had was just too light....
I noticed one other thing I'm not crazy about, and I have seen this before with QB triggers.... When you lighten the trigger spring, the sear has a tendency to not reset if you pull the trigger through the first stage and then release it without firing the gun.... The reason is that the load from the hammer is too great to overcome the friction between the sear and the trigger.... The term for this is "balk firing", and it is VERY common in Springers, where you usually have to re-cock the gun to reset the sear.... I can't tell you if the stock Chief trigger was doing this or not, I didn't check for it.... but the surfaces were so rough I would bet on it, even with the stiff spring.... It is something to keep in mind, if you start to take a shot and then change your mind, you can leave the sear "hanging" on the verge of firing.... The way to clear it is to recock the gun, by pulling the bolt back all the way to take the hammer load off the sear.... It is a good practice to get into, and especially necessary if your sear doesn't reset when you partially pull the trigger and then back off on the shot....
While I had the gun apart, I pulled the barrel and receiver and checked out the ports.... They are pretty small, and all the same size.... The barrel port, receiver port and valve exhaust port are all 3.5mm (0.138"), and the flat O-ring type seal between the valve and receiver is also the same ID.... In addition, the valve in my gun is slid back a bit, so the exhaust port is partially covered by the seal, restricting it even further.... That 25 fps that Ribbon had over my gun could be right there....

.... The valve has slid back because the holes in the tube for the valve screws are a bit larger than the screw heads.... With 2000 psi of pressure pushing back on the valve with 1178 lbs. of force, the valve slides back until the screw heads are tight against the back of the holes, and the exhaust port, that should have lined up with the seal and receiver port, is back from where it should be.... A seal with an oversized hole would probably help, or if, as I am, you plan to enlarge the ports, you can carve more out of the front of the exhaust port than the back, to re-center the port relative to the hole in the tube and the ports in the receiver and barrel....
Anyways, with the gun back together, I tethered it to a regulator set to 1800 psi and proceeded to go through the entire range of preload adjustment, shooting five JSB 18.1 gr. Heavies at each turn of preload and recording the average.... The results are in the chart below....

This is a classic curve, with a "plateau" where additional preload adds NO velocity and only wastes air.... a "knee" where the velocity starts to drop, and then a "downslope" where every time you reduce the preload the velocity drops.... Stock preload on my gun was 4 turns out, right at the beginning of the plateau.... Therefore adding more preload does nothing but make the gun louder and waste more air.... On the other hand, the knee of the curve is at 5-6 turns out, and that is a good place to experiment.... Less preload than that looses more velocity.... Now this was at 1800 psi, at the 2000 psi fill pressure, the entire curve would move up (faster) and to the left (more preload).... That 4 turns out stock setting is probably right at the top of the knee at 2000 psi.... Anyways, now that I knew where the knee was, I could try reducing the preload to see what happened to the shot strings.... Here are the results....

Only shots within 4% of the maximum velocity are shown on this chart.... The red string is from yesterday, but I omitted the shots below that 96% velocity point, leaving only 8 shots within a 4% ES, the velocity dropping with every shot.... pretty dismal the way it came from the factory.... This is what happens when you tune a gun on the plateau.... I reduced the preload 2 more turns, and shot the string in black at 6 turns out.... The starting velocity was only 742 fps, it wasn't until shot #5 that the velocity climbed within 4% of the 794 fps peak (which is why the first 4 shots have been omitted from the black line).... I got 22 shots, ending at 1400 psi, so the usable string was 18 shots within a 4% ES, starting at 1900 and ending at 1400.... I increased the preload a turn, to 5 turns out, and shot another string, recorded above in blue.... This is about as good as it gets for my 2000 psi fill.... The first shot was within my desired 4% ES, the string peaked at 809 fps, and I got 16 shots with the string ending at 1500 psi.... Here is the summary of these 3 strings.... all of which were shot with JSB 18.1 gr. Heavies, from a 2000 psi fill....
4 turns out - 8 shots, average 814 fps (26.6 FPE) @ 0.98 FPE/CI
5 turns out - 16 shots, average 794 fps (25.4 FPE) @ 1.31 FPE/CI
6 turns out - 18 shots, average 779 fps (24.4 FPE) @ 1.41 FPE/CI (1900 psi usable fill)
So, there is the key to tuning your stock Beeman Chief.... Try backing out the preload adjuster a turn or so, to get a proper bell-curve and double your usable shot count, with a big increase in efficiency....
Bob