Thoughts on Hammer Travel Adjustment

[rsterne]

OK, to continue.... Let's assume for a minute that my new style "travel adjuster" is now a "lift adjuster".... How is it different than the bstaley mod?.... Well, for one thing, it is recessed into the front of the hammer instead of the way the MRod (or Challenger) adjuster sticks out the front.... This INCREASES the available hammer travel instead of decreasing it.... Secondly, it has a spring guide attached to it.... How about the lloyd-ss "long stroke hammer"?.... Lloyd of AirGun Lab has been using hammers with a hollow front for quite a while to increase the stroke, a simple and effective method.... He takes that a step further by machining the back of the Disco valve to shorten it, allowing even more hammer travel.... I believe he leaves a central "stub" on which he mounts an O-ring to seal the stem, perhaps he will chime in with the details....

What I've done here is to combine several good ideas, none of which I can take credit for.... However, the result is a hammer for the Disco that first of all increases the stroke, secondly provides adjustability for the valve lift, and thirdly incorporates a spring guide / cocking indicator.... I believe I first saw that idea used by Ribbonstone although his simply sat in the hammer, held in place by the spring.... There is even a fourth possibility, and that is adding a cocking knob to the end of the spring guide (which Lloyd and I and others have already done).... That would additionally add mass to the hammer, and could be turned to adjust the lift.... Lloyd uses a captive O-ring on that shaft as a visual indicator of the valve lift, enabling direct measurement of that difficult to ascertain measurement....

Anyway, there you have it.... I'll be doing a bit more work on this over the next few days, trying to develop it and optimize it.... I am hopeful that the result will be a hammer that not only increases the potential FPE of the Disco, but provides easier tuneability and hopefully even an increase in efficiency and/or shot count....

Bob

[rsterne]

After writing the previous two posts, I went out into the shop, pulled the hammer out of my Millenium Pumper, and modified it slightly.... I removed 0.040" from the back end of the adjusting screw to allow one more turn of travel.... I can now recess the hammer 0.25" from the front face, reducing the maximum valve travel to 0.060" at the point the hammer face impacts the valve body.... I set the RVA to a whisker short of coil bind on the 2.00" x 0.045" hammer spring, maxed out the travel, and took a shot at 1800 psi, recording the velocity.... I repeated this, adjusting the position of the travel/lift adjuster.... I then decreased the hammer spring preload a turn at a time and did everything again.... Here are the results....



If you look at the curve for the hammer spring near coil bind (blue line, reading from right to left), you will see that as you increase the hammer travel (and hence the Energy and Momentum) the velocity increases and then plateaus.... This is the expected reaction to increasing hammer strike, as I've demonstrated many times on regulated guns (ie at constant pressure, in this case 1800 psi).... However, the last 2 turns of travel show something different happening.... At about 1 turn out, where the valve lift can no longer exceed about 0.10" without the hammer hitting the back of the valve, the velocity starts to decrease.... In that last turn of travel (1/24th of an inch = 0.042") it drops like a stone.... This is clear evidence that the hammer is hitting the valve, and that the loss of valve lift is killing the velocity.... At 2 turns out, we see the same pattern, with the velocity dropping off during the last turn of adjustment.... At 3 turns out, the dropoff doesn't occur until the lift is less than about 0.080" (1/2 turn out).... and at 4 turns out it doesn't occur at all, with the maximum velocity occurring at maximum hammer travel.... That tells me that at 850 fps, the valve is only opening 0.060" or less at 1800 psi....

Sooooooooooooooo.... the facts fit with the theory.... I can now adjust the valve lift providing that I am running lots of hammer spring preload.... There is one problem, however.... The adjustment, even using 24 TPI (0.042" per turn) is VERY finicky.... One turn (with the spring at coil bind) is over 250 fps.... Considering that, it remains to be seen if there is any practical application.... It will depend on whether or not I can achieve better efficiency running on the "left side" of the curves (where the velocity is determined by valve lift directly) compared to on the "right side" where the velocity is determined by spring preload (and it's effect on lift and dwell).... More experimenting is called for.... and the Millenium Pumper may not the the ideal candidate for that....

Bob

[john]

sure wish i had a drawing of the parts to better follow your description.

[rsterne]

There is a photo and description of the hammer at the top of the thread....

Bob

[bstaley]

...
Lots of precedence here, I'll mention two that I am aware of.... One is the bstaley mod for the MRod, where he uses O-rings against the back of the valve to limit the hammer travel and turn the stroke adjuster into a valve lift adjuster.... I never really grasped this until he wrote up an excellent thread on how it works, which you can read here.... http://www.gatewaytoairguns.org/GTA/index.php/topic,37695.0.htm .... My compliments to bstaley on figuring this out and implementing it.... The idea (to paraphrase) of not affecting the first part of the shot string, while limiting the lift (and possibly also the hammer bounce) in the second half of the shot string are a perfect example of critical thinking followed by a simple solution, and I congratulate him for it....
.....

I think you give me more credit than I deserve.  I came up with the idea while attempting to tune my p-rod, which was only my 2nd pcp.  My only prior tuning experience was installing an RVA on a disco.  Beginners luck more than anything.

It's only after learning more about how pcp's work, by reading posts by yourself, the others you mentioned in that post, and many others, did I really start to understand how and why it works as well as it does....and I'm still learning.

Looking forward to your experimentation with this, and would buy one of those hammers from you in a second if you decide to sell them!

[lloyd-ss]

A lot of folks have contributed a lot of good ideas on this forum, and many of those ideas are coming together in this thread.  Bob has already mentioned a lot of the people.  I think Bob is planning to combine those ideas into an ultimate... airgun device, what ever it may turn out to be.
To digress a little, I used the recessed hammer face on my first version of the Disco Double, in Sept 2011,
http://www.gatewaytoairguns.org/GTA/index.php/topic,17637.60.html

There is a series of pics in about the middle of the thread that show the hammer and cocking rod. Bob has now added adjustability to the actual striker, and that is a good thing.
Bstaley has added the o-rings to limit, and tame, the hammer stroke, and that is also a good thing.
But, there may be another material choice that has better characteristics than the usual o-rings to limit the hammer travel. Back when I was first working on mustang Mike's carbine, I again used the recessed hammer face, but I also added an impact absorbing pad between the hammer and the back of the valve.
The energy absorbing material takes a little thinking to understand.  It is an elastomer, it has some hardness to it, and can absorb
quite a bit of energy.   I have been using energy absorbing elastomers for years to protect delicate equipment during shock events.  They work well when chosen correctly for the application. Elastomers all have elasticity, but just HOW elastic they are that makes the big difference. Imagine hitting a golf ball that is laying on the floor with a hammer. Major bounce-back. Now take that hammer and hit a memory foam mattress.  Almost No bounce-back.  They are both elastomers, but the speed at which they release the energy that has been  imparted into them is very different.  The impact absorbing pad is like a very dense  version of that memory foam.  The hammer bounces back very little, and at a much reduced velocity. The yellow impact absorbing rings fit around the gland nut on the valve, and snub the hammer when it is an over-travel condition. Hammer bounce is virtually killed.
Here are the pics of valve and hammer and parts and pieces of that make up that high powered set-up.
Notice also in that pic where I borrowed Bob's Disco valve idea of the one-piece front end. 

Lloyd-ss

[Motorhead]

Got to shoot my M-rod with the hammer design shared earlier in this thread over the weekend ... Was not charting shot strings but did fool around throttling up and throttle down the power by ONLY changing travel ( kept same spring preload ) and found gun to shoot Very very well keeping shot report extremely quiet telling me dwell is staying very low.  Air usage indeed seemed to be just sips at all power settings. 

* Do need to remove a tad of my hammers inner sleeve length on back end so contact surface can be drawn further back increasing travel further yet.

[rsterne]

OK, guys.... here is a question for you.... Have a look at the velocities at maximum hammer travel, where the valve can only be opening about 0.060".... ie the very left hand side of the graph above.... Look at the velocities for the four hammer spring preloads (and I'll add one not on the graph)....

Coil Bound ~690 fps
1 turn out (not on the graph) ~720 fps
2 turns out ~740 fps
3 turns out ~780 fps
4 turns out ~ 850 fps

There is clearly a trend opposite to what you would expect.... More spring preload (and a higher force driving the hammer) is producing less velocity.... However, the distance the hammer travels from being released by the sear until it hits the valve stem is the same for all.... In addition, the valve stem can only be pushed 0.060" before the hammer hits the back of the valve.... so the lift is limited the same for all....

Anybody want to have a stab at explaining why?....

Bob

[Motorhead]

OK, guys.... here is a question for you.... Have a look at the velocities at maximum hammer travel, where the valve can only be opening about 0.060".... ie the very left hand side of the graph above.... Look at the velocities for the four hammer spring preloads (and I'll add one not on the graph)....

Coil Bound ~690 fps
1 turn out (not on the graph) ~720 fps
2 turns out ~740 fps
3 turns out ~780 fps
4 turns out ~ 850 fps

There is clearly a trend opposite to what you would expect.... More spring preload (and a higher force driving the hammer) is producing less velocity.... However, the distance the hammer travels from being released by the sear until it hits the valve stem is the same for all.... In addition, the valve stem can only be pushed 0.060" before the hammer hits the back of the valve.... so the lift is limited the same for all....

Anybody want to have a stab at explaining why?....

Bob

Sounds like the carnival sledge and bell ....
A fast glancing blow sends the weight up striking bell.
A solid hard hit only jumps the weight a bit.

Kinetics at work here and honestly don't have a clue ... tho likely one of our resident brainy aches will enlighten us

[bstaley]

perhaps the harder the hammer hits the valve face, the faster it rebounds, reducing dwell.

[lloyd-ss]

perhaps the harder the hammer hits the valve face, the faster it rebounds, reducing dwell.

Bstaley, that sounds like a reasonable explanation to me. 

Bob,
Now that we know more or less know what is happening with the hammer and valve travel in the graph at the top of this page, and you have indeed described it in your text, I have to say that I think the graph is misleading.  The sharp downturn at the left introduces another variable: limitation of maximum valve lift.  Also, I would think that the downturn is actually a sharp downturn, not gradual as the curve smoothing implies.  Once the valve travel limitation eliminated, I think all the curves might look somewhat similar to the 4-spring-turns-out curve.

Nice work and a good discussion.
Lloyd-ss

[PakProtector]

I think Mr. Staley has it in a nutshell. The reduced preload stroke will leave the valve opening and closing more slowly; the high preload gets the valve open quite quickly, and then the valve starts closing. I'm ready to bet heavily that the high preload, low velocity shot is going to be more efficient than a comparable velocity shot done with reduced preload.
cheers,
Douglas

[lloyd-ss]

I think Mr. Staley has it in a nutshell. The reduced preload stroke will leave the valve opening and closing more slowly; the high preload gets the valve open quite quickly, and then the valve starts closing. I'm ready to bet heavily that the high preload, low velocity shot is going to be more efficient than a comparable velocity shot done with reduced preload.
cheers,
Douglas

Douglas,
X2
I like what you said in your last sentence.
Lloyd-ss

[bstaley]

Yes, while metal on metal is very different than using o-rings or some other material as a buffer, using this method to limit power  should be more efficient than transfer port restriction also.  Restriction results in an attenuated, longer duration pressure pulse  because it reduces the closing forces on the valve.  Limiting valve lift does not attenuate the pulse until the hammer hits the buffer (quicker rise time than restriction.....the most important part of the impulse).  It also does not reduce the closing forces(for any given lift), so dwell is shortened rather than lengthened as restriction would do.

[rsterne]

OK, enough answers that you all agree with my (unstated) premise that DWELL is the only possible answer.... The higher preload is driving the hammer faster, it's hitting the back of the valve harder (and in a shorter period of time) and rebounding quicker.... thus the total dwell and hence (since the lift and pressure are fixed) the total amount of air released is less.... Makes perfect sense, and I alluded to that in my Millenium Pumper thread when I first mentioned this phenomenon....

bstaley, you mentioned metal on metal.... Well, guys, I left out (on purpose) one important bit of information until Lloyd posted about the energy absorbing elastomer.... The initial testing in the Millenium Pumper was metal on metal.... However, when I pulled the gun apart to modify the hammer adjuster to allow another turn of throw, I machined 0.100" off the back of the valve (leaving a 1/4" diameter stub around the stem) and fitted a washer of Lloyd's yellow elastomer to the back of the valve.... Therefore, the graph at the top of this page was done with the elastomer in place.... I agree with Lloyd's comments that I probably shouldn't have used "curve smoothing" on the graphs, there should be a (more or less) sharp discontinuity at the point where the hammer begins to contact the elastomer / back of the valve....

Any more comments, now that you know there was an energy absorbing bumper in place for the hammer to hit?....

Bob

[bstaley]

How thick is the washer then...0.10?  The thickness and elastic properties of the buffer material will have a significant effect on hammer and valve response.  I believe that having a somewhat thicker, compressible buffer does help improve the length of a shot string, but perhaps with a bit more ES, as a result of allowing lift to vary a bit more over the string.

[PakProtector]

The 3 o-ring stack comes into play gradually across 1/2-3/4 quarter of the shot string and is deflecting much further in the last quarter. For a maximum rate open and close I find the metal-metal striker/valve contact hard to beat conceptually. For a max efficiency reg'd rifle I would want a monster valve, a strong hit and a travel limited opening. Hammer will bounce fairly effectively on metal and the poppet will only have its own mass do deal with on close. Can tune this system with a bit of damper material and a minimum mass hammer( I for one have no issue with a nutty cocking effort ) to incease dwell at max opening.
cheers,
Douglas

also, can get direct valve lift measurements to simulate an unreg'd system response and get a fine idea of exactly what's going on in order to develop a set of adjustments that will let you shoot through a 2x spread in tank pressure v. an open loop operation.

[Motorhead]

Why not just place a thin washer/spacer made of delrin, hard polyurethane or a rubber washer up against face of valve to give just a tad bit of shock absorbing and set up so hammer DOES BOTTOM OUT every shot and JUST ADJUST lift for power ya want ??
Dial in to much lift and have hammer no longer bottom out ... add some more spring tension.

This ONLY for a regulated gun .. But another way to fix valves flow so it does EXACTLY the same thing shot to shot 

[rsterne]

bstaley, yes the washer is about 0.110".... actually the piece Lloyd sent me was squashed? in the middle and varied from about 0.080" to 0.120".... The whole point of this material is that IF the hammer hits it, it should, at least in theory, decelerate it quickly and not cause it to bounce.... pretty much the opposite of a stack of 3 O-rings, I would think....

Doug, I would think that metal to metal would have a VERY elastic collision, and coupled with any push from the valve stem the hammer would have the biggest possibility of bounce you could give it for that combination of air pressure, hammer strike, and springs.... It does make sense that the dwell of the valve would be much less, as once the hammer impacts the valve, it is no longer being pushed by the hammer (so it should stop sooner and reverse quicker), and if the hammer has rebounded from the metal valve body, it can close quicker as well (no hammer to accelerate).... Sorry, I don't understand your last paragraph....

Motorhead, that's kind of where I was going with this idea.... I was thinking of tuning the gun on the "left side" of the curve (which appears to be pretty much a straight line of velocity vs. lift).... You would select the hammer spring preload that allowed maximum lift (and velocity) to occur with the hammer still in contact with the valve body.... However, the adjustment required to dial in the velocity you want is VERY sensitive, in this case about 250 fps per turn on 24 TPI.... Using the "coil bound" example (the blue line), pretty much the entire range of velocities is available, zero to maximum, within about 2.5-3 turns on the adjuster.... At the very minimum, a finer thread is in order, IMO, to make this practical....

Bob

[PakProtector]

hey Bob,
I was thinking of gettimg empiracal valve lift v. velocity but that would only be applicable to a certain kind of collision( and that ov course dependant on the type of any damper material used if any). Was recalling discussion from a while back egarding hammer behavour and valve opening and thought to use this one as a test bed. It could yeild non-impact data if you ran over the hill past the collision-induced knee for any given preload, you'd know the lift to within a few thou if the velocity discontinuity were sharp enough.
cheers,
Douglas