1311mikeg
Proactive member
Each spring has a unique motion characteristic related to it's diameter(s), weight and number of coils - ie: it's rate. This is complicated by the way in which the stored rotational energy is released and passes through the many coils - the stiffer, large wire diameter springs with a low number of coils are easiest to use, having less non-uniform expansion and producing less internal resonant energy with pulsing waves of energy. Stiffer springs are generally less fussy with guide fit, needing less damping.
Testing now includes rifle recoil and position, using the ever-expanding spreadsheet and the patterns that it is revealing, it is becoming possible to predict & demonstrate what tuning can do and what changes occur (with changes to spring & piston weight, etc for a given transfer port size).
This, of course, is only what several skilled tuners can already do from experience, notes or memory ref: @JimPlas.
The spring will do what the spring will do - it's motion envelope is essentially constant with different pellets, probably changing amplitude but not timing - the spring reverb energy remains constant. Cycle timing will only change for that particular spring with alteration to preload, piston weight or transfer port size. Changing pellet brand will alter the pellet exit time, this may be seen to shift relative to the firing cycle and the rifle motion null point - this is due mainly to shifts in pellet break time and consequent pressure/rate changes.
The problem with stiff, heavy springs comes at a point where a reduction in preload will cause high bounceback & slam just as their efficiency is improving, forcing the use of a lighter piston to maintain power ceiling. The optimum length of spring may indeed be closely set within a millimetre or so and is very much affected by spring coil end finishing styles - this can make finishing to length frustrating, often cutting a spring to find the opposite effect due to coil end finish issues.
V-Mach is a very good example of a stiff 3.3mm wire diameter spring with amiable characteristics of about 7.9N/mm, 27 coils at 226mm whereas the Titan08 rises to 10.3N/mm, 24.5 coils when shortened to 220mm. Optimum thinner wire multicoil springs might be around 6-6.4N/mm using 2.96-3.0mm wire with upwards of 30 coils.
If the spring is too stiff and heavy for the HW77K design limitations then it will remain harsh with poor timings, examples may be AirSporter and Titan 3 or 8 (shown below). Titan 10+ has more turns of smaller diameter wire and behaves much better . . if you can find one! WW Dagda looks similar.
Most tunes work easiest with heavier pistons as the spring energy may be reduced, the firing cycle is longer, giving good pellet break & exit timing to the null in rifle motion and the springs are easier to control or damp with good fitting guides.
The pictures below are representative of many spring & pellet combinations that were tested. They were made using a light 220gm piston because that would be most likely to move quicker and show any motion changes rather than a heavy piston that may mask smaller changes. And of course the spring will be required to give more propelling energy.
The spring coils can be seen to resonate at a characteristic frequency, usually around 1kHz, less & shorter time for higher mass or stiffer springs - this is regardless of pellet, changing only with external preload or spring length as piston weight changes are made. Spring reverb amplitude may change with pellet brand and may cause interference effects that will alter the pellet air coupling and speed in an unexpected way. Reverb energy is small but wasted energy.
Reverb amplitude can be muted by guide choices but guides will not alter not the frequency - there is a risk of having guides too tight that over-damp the piston motion - tight is good but not hammer-fit, remember they're loose when the spring is compressed and only tighten when expanded at TDC and may easily grab the guide at bounceback. In the extreme, a reverb pulse may cause the end few coils to collapse/open then grab and drag the guide off the trigger block - again why heavy one-piece steel or brass guides may be preferred.
Reverb amplitude, timing and frequency are easily changed. These affect not only pre-pellet exit and power transfer that may be related to pellet power, but also the rifle behaviour during and also immediately after pellet exit, altering the 'feel' of the shot cycle - although the pellet has actually gone by then.
Energy is left in the air - when testing pellet travel along the barrel, the pellet hits the measuring probe and instantly deforms, effectively blocking the barrel - so if the probe is towards the end of the barrel, the piston that would normally slam & stop with a little bounce, is forced backwards by the trapped air pressure into a repeated bounce & slam cycle. This allows the air excess energy to be estimated and compared to calculated efficiency - there seems to be a reasonable relationship but it's not constant.
The cluttered picture below (sorry!) tries to show the relationship between Piston motion and Acceleration (lhs) to Position and Velocity (rhs). It tries to explain the timings and significance of the waveforms - the waveforms were all captured at the same time from the same shot - the marker pulses and X-axis times (in milliSeconds) help to relate timings. Comparisons may be made as tuning changes are made to similar other pictures, but that may be too much to post here (there's hundreds of them!).
I can't claim to relate all of the waveform shapes to performance or grouping yet - just some of the more obvious ones.
This recording was made using a heavy, tight Crosman Premier 10.5g pellet, included because it favourably shows the best timing of Pex to the rifle recoil null (r/h top red trace) for this spring and piston, albeit with poor, high recoil. Using a light pellet would cause the Pex pulse to shift earlier to the left by several millimetres. This light piston setup can produce average grouping but has a dislikeable feel with a lively recoil that quietens a bit with some lighter pellets.
Poor timing of Pex to the (l/h lower blue) acceleration trace causes the pellet to exit after surge acceleration crosses through zero - good tunes almost invariably have Pex before or at that zero crossover point. This is proportional to the late break time and consequent breech exit time of the pellet shifting along the piston motion waveform (top l/h red trace) with increased initial peak pressure but further, faster piston travel afterwards. The break time of lighter, softer pellets will occur earlier in time & piston position (to the left of the trace) usually with a proportional reduction in breech exit and Pex.
Tuning success is thought to be seen as a reduced size of the waveforms or changes in waveform shape - for example, the l/h lower blue trace shows a particularly bad large 60mm+ recoil shape - it looks like a chainsaw drive link with fast mid & final rates of change, rather than a nice gentle 1/2 sinewave - together with it's large size, this affects the 'felt' initial recoil. This is due to spring resonance and a large pulse(s) of energy arriving at a bad time(s).
Certainly, all of the tested successful tunes have reduced recoil, velocity change and good timings whilst demonstrating increased efficiency - meaning that a reduced spring may be used, and vice versa.
The spring reverb energy may be seen later in the r/h lower blue waveform as influencing the normal high speed rise after TDC but before pellet exit Pex - this is also felt as surge reversal recoil. The parasitic reverb energy particularly shows itself in the large burst of high speed energy following Pex continuing off the r/h edge of the page - this again may be felt - dunno how you can feel 1kHz though? This blue waveform is unusually large in size and would be typically be much 30-40% smaller, without the reverb. A light piston does not suit this spring, things do improve a bit with increased piston weight but it really is too stiff and not suited to the HW77 - it needs to be longer. Pellet exit may be forced to other points along this waveform area by tuning (or pellet type).
A marginal improvement is made with one spring open end, ok where it is supported in the piston bore but an open end on the breech end has problems, presumably where the open coil end tends to skid outwards or distort the spring coils (even on the guide) - presumably why Vortek uses a tubular breech outer guide. More work needed to either make a economy version of Vortek or even counterbore the piston or breechblock to accept a support sleeve - I can't shorten my piston since that's where my Hall Effect sensor magnet is mounted. A simple drilled location hole, shaped into the metal guide flange seemed to help but it's not nice! The effect of the open ends is to reduce the spring coefficient to about 8.5N/mm and reduce the non-linearity at low compression - it gets a softer feel on shooting.
Two graphs are shown below, one for Titan TS08 @ 227mm and one for V-Mach @ 221mm, both produce roughly the same power with a standard transfer port. They have similar barrel timings vs distance. The Titan8 has a light 220gm piston whereas the V-Mach has a 264gm piston - they 'feel' totally different despite the pellets performing the same. The Titan has less delay in pellet breech time, so the pellet accelerates too early, timings are poor, recoil is high & efficiency is lower.
A perfectly good spring like several others, but when shortened to fit, it does not suit the HW77 because it becomes too stiff, it also suffers from high resonant energy that can be badly timed. A better compromise can be reached if the piston weight is increased and the spring duly shortened. There is only about 6mm difference in length when using 220gm to 260gm pistons for the same power output. Tight guides, tophat and a sleeve help but cannot control it - the waveforms without guides are interesting - it's like a piano!
The graphs below (and others) show that the HW77 is currently dominated by its transfer port - hence the same axial pellet barrel speeds. Any attempt to delay the 'knee' of the curve (pushing it upwards) by tweaking the pressure reduces overall efficiency, seeming to alter the final slope only. Selecting the pellet is probably most effective way to alter the first slope, or to alter the transfer port size. Both springs produce a JSB452 pellet motion in the breech of about 60-65mm at TDC.
Testing now includes rifle recoil and position, using the ever-expanding spreadsheet and the patterns that it is revealing, it is becoming possible to predict & demonstrate what tuning can do and what changes occur (with changes to spring & piston weight, etc for a given transfer port size).
This, of course, is only what several skilled tuners can already do from experience, notes or memory ref: @JimPlas.
The spring will do what the spring will do - it's motion envelope is essentially constant with different pellets, probably changing amplitude but not timing - the spring reverb energy remains constant. Cycle timing will only change for that particular spring with alteration to preload, piston weight or transfer port size. Changing pellet brand will alter the pellet exit time, this may be seen to shift relative to the firing cycle and the rifle motion null point - this is due mainly to shifts in pellet break time and consequent pressure/rate changes.
The problem with stiff, heavy springs comes at a point where a reduction in preload will cause high bounceback & slam just as their efficiency is improving, forcing the use of a lighter piston to maintain power ceiling. The optimum length of spring may indeed be closely set within a millimetre or so and is very much affected by spring coil end finishing styles - this can make finishing to length frustrating, often cutting a spring to find the opposite effect due to coil end finish issues.
V-Mach is a very good example of a stiff 3.3mm wire diameter spring with amiable characteristics of about 7.9N/mm, 27 coils at 226mm whereas the Titan08 rises to 10.3N/mm, 24.5 coils when shortened to 220mm. Optimum thinner wire multicoil springs might be around 6-6.4N/mm using 2.96-3.0mm wire with upwards of 30 coils.
If the spring is too stiff and heavy for the HW77K design limitations then it will remain harsh with poor timings, examples may be AirSporter and Titan 3 or 8 (shown below). Titan 10+ has more turns of smaller diameter wire and behaves much better . . if you can find one! WW Dagda looks similar.
Most tunes work easiest with heavier pistons as the spring energy may be reduced, the firing cycle is longer, giving good pellet break & exit timing to the null in rifle motion and the springs are easier to control or damp with good fitting guides.
The pictures below are representative of many spring & pellet combinations that were tested. They were made using a light 220gm piston because that would be most likely to move quicker and show any motion changes rather than a heavy piston that may mask smaller changes. And of course the spring will be required to give more propelling energy.
The spring coils can be seen to resonate at a characteristic frequency, usually around 1kHz, less & shorter time for higher mass or stiffer springs - this is regardless of pellet, changing only with external preload or spring length as piston weight changes are made. Spring reverb amplitude may change with pellet brand and may cause interference effects that will alter the pellet air coupling and speed in an unexpected way. Reverb energy is small but wasted energy.
Reverb amplitude can be muted by guide choices but guides will not alter not the frequency - there is a risk of having guides too tight that over-damp the piston motion - tight is good but not hammer-fit, remember they're loose when the spring is compressed and only tighten when expanded at TDC and may easily grab the guide at bounceback. In the extreme, a reverb pulse may cause the end few coils to collapse/open then grab and drag the guide off the trigger block - again why heavy one-piece steel or brass guides may be preferred.
Reverb amplitude, timing and frequency are easily changed. These affect not only pre-pellet exit and power transfer that may be related to pellet power, but also the rifle behaviour during and also immediately after pellet exit, altering the 'feel' of the shot cycle - although the pellet has actually gone by then.
Energy is left in the air - when testing pellet travel along the barrel, the pellet hits the measuring probe and instantly deforms, effectively blocking the barrel - so if the probe is towards the end of the barrel, the piston that would normally slam & stop with a little bounce, is forced backwards by the trapped air pressure into a repeated bounce & slam cycle. This allows the air excess energy to be estimated and compared to calculated efficiency - there seems to be a reasonable relationship but it's not constant.
The cluttered picture below (sorry!) tries to show the relationship between Piston motion and Acceleration (lhs) to Position and Velocity (rhs). It tries to explain the timings and significance of the waveforms - the waveforms were all captured at the same time from the same shot - the marker pulses and X-axis times (in milliSeconds) help to relate timings. Comparisons may be made as tuning changes are made to similar other pictures, but that may be too much to post here (there's hundreds of them!).
I can't claim to relate all of the waveform shapes to performance or grouping yet - just some of the more obvious ones.
This recording was made using a heavy, tight Crosman Premier 10.5g pellet, included because it favourably shows the best timing of Pex to the rifle recoil null (r/h top red trace) for this spring and piston, albeit with poor, high recoil. Using a light pellet would cause the Pex pulse to shift earlier to the left by several millimetres. This light piston setup can produce average grouping but has a dislikeable feel with a lively recoil that quietens a bit with some lighter pellets.
Poor timing of Pex to the (l/h lower blue) acceleration trace causes the pellet to exit after surge acceleration crosses through zero - good tunes almost invariably have Pex before or at that zero crossover point. This is proportional to the late break time and consequent breech exit time of the pellet shifting along the piston motion waveform (top l/h red trace) with increased initial peak pressure but further, faster piston travel afterwards. The break time of lighter, softer pellets will occur earlier in time & piston position (to the left of the trace) usually with a proportional reduction in breech exit and Pex.
Tuning success is thought to be seen as a reduced size of the waveforms or changes in waveform shape - for example, the l/h lower blue trace shows a particularly bad large 60mm+ recoil shape - it looks like a chainsaw drive link with fast mid & final rates of change, rather than a nice gentle 1/2 sinewave - together with it's large size, this affects the 'felt' initial recoil. This is due to spring resonance and a large pulse(s) of energy arriving at a bad time(s).
Certainly, all of the tested successful tunes have reduced recoil, velocity change and good timings whilst demonstrating increased efficiency - meaning that a reduced spring may be used, and vice versa.
The spring reverb energy may be seen later in the r/h lower blue waveform as influencing the normal high speed rise after TDC but before pellet exit Pex - this is also felt as surge reversal recoil. The parasitic reverb energy particularly shows itself in the large burst of high speed energy following Pex continuing off the r/h edge of the page - this again may be felt - dunno how you can feel 1kHz though? This blue waveform is unusually large in size and would be typically be much 30-40% smaller, without the reverb. A light piston does not suit this spring, things do improve a bit with increased piston weight but it really is too stiff and not suited to the HW77 - it needs to be longer. Pellet exit may be forced to other points along this waveform area by tuning (or pellet type).
A marginal improvement is made with one spring open end, ok where it is supported in the piston bore but an open end on the breech end has problems, presumably where the open coil end tends to skid outwards or distort the spring coils (even on the guide) - presumably why Vortek uses a tubular breech outer guide. More work needed to either make a economy version of Vortek or even counterbore the piston or breechblock to accept a support sleeve - I can't shorten my piston since that's where my Hall Effect sensor magnet is mounted. A simple drilled location hole, shaped into the metal guide flange seemed to help but it's not nice! The effect of the open ends is to reduce the spring coefficient to about 8.5N/mm and reduce the non-linearity at low compression - it gets a softer feel on shooting.
Two graphs are shown below, one for Titan TS08 @ 227mm and one for V-Mach @ 221mm, both produce roughly the same power with a standard transfer port. They have similar barrel timings vs distance. The Titan8 has a light 220gm piston whereas the V-Mach has a 264gm piston - they 'feel' totally different despite the pellets performing the same. The Titan has less delay in pellet breech time, so the pellet accelerates too early, timings are poor, recoil is high & efficiency is lower.
A perfectly good spring like several others, but when shortened to fit, it does not suit the HW77 because it becomes too stiff, it also suffers from high resonant energy that can be badly timed. A better compromise can be reached if the piston weight is increased and the spring duly shortened. There is only about 6mm difference in length when using 220gm to 260gm pistons for the same power output. Tight guides, tophat and a sleeve help but cannot control it - the waveforms without guides are interesting - it's like a piano!
The graphs below (and others) show that the HW77 is currently dominated by its transfer port - hence the same axial pellet barrel speeds. Any attempt to delay the 'knee' of the curve (pushing it upwards) by tweaking the pressure reduces overall efficiency, seeming to alter the final slope only. Selecting the pellet is probably most effective way to alter the first slope, or to alter the transfer port size. Both springs produce a JSB452 pellet motion in the breech of about 60-65mm at TDC.
cave man sir,cave man
