Simplified Balanced Valve

[PikeP]

Bob I can understand why Domingo uses 1/3rd mass of air...another way to look at the air column is like a 3 car train...once the first train collides with the projectil it gets "derailed"...itll still exit at great speeds but not as if there were no projectile to derail it...that's my analogy ...in fact dry fires are incredibly louder...likely due to the vastly increased air speed and muzzle pressure combined

Alternatively if I simply calculate pressure drop and KE loss drop the more likely air speed at muzzle is 2200-2400 in the best conditions...that's using very raw napkin math but 2800~ peak at 4500 psi doesn't seem too unreasonable, only if you were able to shoot and measure just air...maybe that 2900 fps Texan shot? Idk..not giving credit but between your losses measuring 2200+ fps projectiles and a Texan measuring 2500+ unknowns I cant really discount the possibility of a 2850~ peak fps for air at 4500 psi

[mann]

I was thinking it would be easier to put in the transfer port adjuster in the valve that way you can adjust to what ever you want to do it. .050 reduction moved me down almost 200 fps which isn't severely choking the Valve and I can still get to my desired power level

[PikeP]

It's a very viable approach mann. That's why I mentioned it a few posts earlier. You're on the right track with your thinking. Either greatly choke the flow to create a range to tune in or solve it at the valve itself as discussed at length here. There should be no question as to how to do it provided you're measuring volumes and venting correctly with a reasonable balance ratio.

[mann]

Yeah everything is working properly other wise and it would be a set screw in the valve which the air tube all ready has a hole for it so it can be adjusted from the outside of the tube and wouldn't have to pull the valve if I wanted to change it . That's what I'm going to do .put in my .200 transfer port and restrict the tp at the valve . Thanks guys for all the help

[PikeP]

Ok I chose the 3ms dwell to model. Here are my results. These are approximate models not actual interpreted data..

Both charts make sense in terms of flow..

3ms dwell is 1.07 fpe/ci  with roughly 5370 mg ejected at an average rate of 1790 mg per second..

2.66ms dwell is 1.156 fpe/ci with roughly 4670 mg ejected at a rate of 1759 mg per second.

Bob would have to confirm which efficiency above represents his real world results as I cannot..they are both so close...

Notice I clip the flow once lift is beyond the point where it equals the ports ahead flow value.

I am sure Bob would rather the 2.66ms prediction be correct but it very well may be the 3ms...I thought a flow model would throw it off but its seemingly feasible! I find the 3 ms much more reasonable now although I was only figuring a few tenths less..and I was under the impression of higher efficiency but I believe that was before Bob determined it to be off by a small factor which threw me off..

[rsterne]

Domingo's 1/3 factor was (I think) based on the mass you include for a spring, fixed at one end and expanding at the other.... like a hammer spring.... You include 1/3 of the spring mass along with the hammer when calculating the acceleration due to the force of the spring.... An air spring would work in exactly the same way, and that would apply after the valve closes.... but I think not valid while the valve is open.... Think of the mass of a "packet" of air released each uS as becoming part of the mass that now must be accelerated by the next "packet" of air released in the next uS…. The mass to be accelerated keeps increasing every uS as long as the valve is open....

Mike, like you say, the models are very close, I have no way to choose between them.... One thing bothers me about your lift curves, though.... They have two distinct slopes to each side, four slopes in total.... How is that even possible?.... Should not the lift curve be something of a parabola?.... I get it that the flow rate may not be, particularly if the flow is restricted by the curtain effect, or port size.... but I can't wrap my brain around the poppet having four different, and symmetrical, parts to the acceleration/deceleration.... which is the only way you can end up with a curve that isn't a smooth one....

Bob

[rkr]

A question, what material and wall thicknesses could be used for the thimble? I mean we have full reservoir pressure against atmospheric pressure but much smaller diameter "tube" in this case. Would delrin be practical material and what wall thickness is needed? How about aluminium and brass?

[PikeP]

A quick side note. Added wasted volume resulting from a balance valve proportionally effects efficiency by your shot volume...so for a small bore shooting 1 to 4 ccs per shot, a .25 cc chamber would use 25% to 6.25% more air each shot for equal energy of a unbalanced...no way around it...where as with Bob's shot at 19~ ccs his chamber volume is adding only 2%~ more wasted volume.. its definitely worth it for big bores but small bores have a larger trade off the smaller in caliber you go.

[PikeP]

Bob, I tried to be clear those were approximate models not actual representations and more based on what's required to flow the discharged air at those dwells..its simply your dwell time divided by 10 with its peak lift mid way.. if you didnt notice..I dont know your poppet lift time to the curtain or its curtain to close time but if you do and want to share I can model based off those...I also want to note the behavior of the poppet is probably weirder than you think..it goes from having 200 lbs against it to transitioning to ~200 lbs assisting all while its opening in under .5~ ms..not sure it's a perfect parabola with all the transitions it faces. But.. I'm not trying to produce actual representations, I cant make that clearer. I dont mind making corrections but throw me a bone or something. Any lift data would be nice.

[PakProtector]

A question, what material and wall thicknesses could be used for the thimble? I mean we have full reservoir pressure against atmospheric pressure but much smaller diameter "tube" in this case. Would delrin be practical material and what wall thickness is needed? How about aluminium and brass?

For a first cut, use the hoop stress equation. I don't think it matters if the hoop is under tension( pressure inside), or compression( this case; pressure outside). For a cylinder it is Pressure*radius/wall thickness. Do mind the units, yes? If the stress is acceptable, consider the Youngs modulus, and see how much the tube will shrink under this load; that could indeed cause binding you'd never see unless you hired one of Maxwell's demons to go inside and look... Just keep in mind, the likely dimensions are out of the range of the20:1 ratio of diameter to wall thickness...

Bob, the forces acting on the poppet are not symmetrical, and even work to push it open until the balance chamber fills( IIRC you posted this graph earlier). I can see this making an asymmetric lift profile.
cheers,
Douglas

[rkr]

A question, what material and wall thicknesses could be used for the thimble? I mean we have full reservoir pressure against atmospheric pressure but much smaller diameter "tube" in this case. Would delrin be practical material and what wall thickness is needed? How about aluminium and brass?

For a first cut, use the hoop stress equation. I don't think it matters if the hoop is under tension( pressure inside), or compression( this case; pressure outside). For a cylinder it is Pressure*radius/wall thickness. Do mind the units, yes? If the stress is acceptable, consider the Youngs modulus, and see how much the tube will shrink under this load; that could indeed cause binding you'd never see unless you hired one of Maxwell's demons to go inside and look...
cheers,
Douglas

Well, I'm not a math guy so I was more looking for guidelines with serious overkill. Like acceptable materials and wall thickness for say 1/4" or smaller ID thimble. I know it needs less thickness than bigger tubes. I have some aluminium bar and delrin I could use but I'm uncertain of delrin and uncertain of the wall thickness needed for aluminium. I could get some brass rod if aluminium is a bad idea but again I have no idea of the wall thickness. I assume someone did the math before they made their thimble?

[MJP]

Rkr how about actually calculating something, it is not that hard. There are even good calculators for pipe strength at engineersedge.
Put some effort in to it and don't expect a ready answer.

Marko

[mann]

I used aluminum for thimble 1/4 in hole inside outside diameter I left at .400 seems to be working fine

[PakProtector]

I used aluminum for thimble 1/4 in hole inside outside diameter I left at .400 seems to be working fine

With Pr/t the stress at 3000 psi of a .250 ID and .02 wall is just under 20,000 psi. Given the repetitive loading, and 6061T6 yeild strength of about 35,000 psi, this is pushing it. If you choose 1144 steel, a .02" wall for a .25" balance piston should be sufficient. Even 1045 cold drawn at 78ksi looks good. The precision of keeping the hole on center with the OD comes into play of course...
cheers,
Douglas

[rsterne]

Mike, I don't have any actual lift or dwell data for this valve, or I would gladly share it.... I realize that the lift curve will not be a perfect parabola, after all, the forces go from 200 lbs. holding it close to over 100 lbs. negative when it cracks, to 40 lbs. closing it again once the balance chamber fills (sometime before full open)…. After that, however, the force is pretty constant (stem only), particularly with the poppet being shrouded by the thimble, and not subjected to much airflow, except a bit of surface friction as the airflow velocity increases.... At the very end of the valve cycle, when the flow chokes, then the closing forces increase again.... However, this should leave the center and most of the closing cycle somewhat of a parabola, at least that is how I envision it.... If your lift curves are only a representation it doesn't matter anyways.... I'll just ignore the shape and look at the maximum lift and dwell....

I used 1144 Stressproof steel (115K tensile) for my thimble, and on the SS valve on my .457 the big end (1/2" ID) is only 0.030" wall.... That gave me a 3.6:1 safety margin at the working pressure of 3600 psi in that gun.... For the much smaller ID chamber in this valve, I didn't even run the math, but I will now.... Using 6061-T6 aluminum (45K tensile), with a 1/4" ID and a 5/16" OD (0.031" wall) you would have a 3.3:1 safety factor at 3000 psi.... which considering it is contained inside the valve, and that inside a tube, would not concern me.... On my valve, with a 0.315" ID on the balance chamber, I would have to switch up to 2024-T3 (70K tensile) with  a 0.030"wall (3/8" OD), but the SF would be 4:1....

I have tensile strengths of 8.5K for Delrin and 16K for PEEK.... Using that with a 1/4" ID at 3000 psi would require much thicker walls.... With Delrin, the OD would have to exceed 3/4", so that is not practical.... With PEEK, you could get away with a 9/16" OD, resulting in a 0.156" wall with a 1/4" ID and have a 3.6:1 safety margin at 3000 psi.... Plastic, as you can see, is NOT very good as a pressure vessel, I would not recommend it....

Bob

[mann]

It came out to .413 I think is where I left it but I'm also running it In a regulated gun at 2100 psi after I did few tests from my spare tube from 3000 psi down. It will now stay in the regulated tube

[PikeP]

Bob not only does what you state effect the shape of lift curve over (dwell) time...but so does the flow through the valve..low lift low flow is going to act way differently than high lift high flow on both the open and close side...I wouldnt theorize anything as ideal or perfect..certainly in those conditions with wide variance of forces in play in short order. Jmo!

The thimble should ideally be so small that the material strength can be very forgiving..at least imo. I'd still use high quality metal rated for HPA as itll allow thinner walls that reduce volume taken away in the valve. I think a failure event would be fairly harmless unless a pellet were loaded..plastic versions more so as they could be the next projectile if they break..

[rsterne]

I think I may have a way to measure the actual valve lift.... If it works, I will give you a table of lifts at various SSG Gaps and two tethered pressures....

Bob

[PakProtector]

I'd still use high quality metal rated for HPA as itll allow thinner walls that reduce volume taken away in the valve.

What steel's are 'rated for HPA'? Or for that matter, any metal? Are you serious?
cheers,
Douglas

[PikeP]

I'd still use high quality metal rated for HPA as itll allow thinner walls that reduce volume taken away in the valve.

What steel's are 'rated for HPA'? Or for that matter, any metal? Are you serious?
cheers,
Douglas

Are you serious? Do your own research...rating for hpa would be based on Tensile strength and wall thickness and can be nearly any metal, provided you size it right...I was vague  because I'm not going to go through a list of every element in earth comparing tensile strengths and thicnkess... thats not what I signed up for...if you're building this valve or anything to do with HPA then that is YOUR responsibility..