Limited pressure drop design

[Ben]

For those who didn't read some of the discussion in APH customizations thread, in that thread I had the idea to use air-to-water ratios and high starting pressures to keep performance rather constant in simple air pressure. While it's not a new idea, no one really thought about taking advantage of this before. This is an interesting near CPS (or even simply CPS if you do it right) design I think has a lot of potential. Given that it's cheaper than LRT and CAP, easier to get high power than LRT, has no flow limits unlike LRT, and works like a standard PC like LRT I think it's a very viable option we should starting using.

To make a LPD design, you essentially build an air pressure piston water gun like Supercannon II but with a much larger air chamber. You can use bends, bushings, whatever you want, to make the air-to-water ratio higher. The idea is that the more air there is, the less effect a certain amount of expansion will have on the total pressure.

Yesterday at work I did some math figuring out how constant LPD design could be. The result surprised me. It's rather easy to get fairly constant pressure without any regulator or special parts aside from pistons.

Below is a simple chart showing how much pressure will drop for a given ratio.

ratio	%drop
0.5:1	67%
1:1	50%
1.5:1	40%
2:1	33%
2.5:1	29%
3:1	25%
3.5:1	22%
4:1	20%
9:1	10%
19:1	5%
99:1	1%

%drops of 33% or less seem perfectly reasonable to me. I should figure out the approximate %drop for LRT because LRT isn't perfectly constant. I bet it's something like 20% actually, which isn't much lower than some of these.

Note that you get diminishing returns fast though. I included 9:1, 19:1, and 99:1 ratios because they show how large you need to make things to get more constant pressure.

Because at higher pressures you get diminishing returns in range you can take advantage of these %drops to keep your range within a small range. For example, if a water gun with a 2:1 ratio starts at 100 PSI, your lowest pressure will be 67 PSI, which probably would only have a few feet less range.

Your water output will be affected much more than range however. But that will be minimized if you use a larger ratio like 3:1 or 4:1. 3:1 seems to be the best ratio to compromise between %drop and size.

I'm rather excited to build this type of water gun. I intend to build a test water gun of this variety for the review contest as the top prize, so if you want one for free or nearly free look into sending in some reviews.

[Silence]

Here's a link to the water gun review contest.

Sounds good. For a long time I thought CPS was better than precharging - but now that we've seen the limits of rubber bladders, namely pressure and flow (both of which are tremendously important) and how they can be outclassed by air pressure, prechargers are looking enticing indeed.

Ben, I'm surprised nobody (which implies you) has done correlations between pressure/output and range for a single nozzle. If I ever get a pressure gauge, that's definitely something I'll try.

[Ben]

Ben, I'm surprised nobody (which implies you) has done correlations between pressure/output and range for a single nozzle. If I ever get a pressure gauge, that's definitely something I'll try.

Nope, I haven't tried to make any correlations between them. It's too complicated and anything I make would be right on one blaster and wrong on another. I have tried some plots for pressure vs. output but from what I remember I "prettied up" those results because they weren't looking as good as they should have. I don't remember how much I prettied up the range graph too but I doubt I changed more than one range if I did.

We should do more testing. I'll see what I can do. I'd like to do something with regulated pressure and multiple tests to keep results most consistent and minimize the effect of outliers.

[Silence]

Maybe timing gets hard with small volume as shot time decreases. I can see why you were disappointed, as that should have been a square root correlation.

[Ben]

Something's making the flow more linear than the simple Bernoulli equation would suggest. I'm betting the most major reason has something to do with the fact that the pressure wasn't truly constant over the duration of the shot. Other things like minor loss, friction, viscous forces, etc. need to be taken into account too, and I'm sure there's other things I don't even know about that cause weird stuff to happen. Once I learn some sort of serious fluid mechanics I should be able to explain stuff like this better.

The timing was adequate, or I had hoped it was. We used a video camera for those tests with precision of 1/30 second. Maybe that's not adequate or our equipment was inadequate for the job. There's no other method to measure the time without affecting the water flow directly except for possibly sound analysis, which shouldn't be more accurate than video analysis.

[Ben]

I just completed some tests with LRT to determine exactly how constant LRT really is. While my pressure gauge wasn't too precise and ambiguous for pressures less than 20 PSI, I've found that the pressure starts at 20 to 23 PSI and ends at 16 to 18 PSI. LRT seems to have approximately a %drop in the range of 20% to ~27%, which is perfectly achievable by reasonable LPD systems. A LPD system with a ratio between 3:1 and 4:1 should perform very similarly to LRT. We can consider some forms of precharging to be constant pressure systems.

For the sake of classification, I'd like to propose the following based upon %drop:
- a %drop of greater than 50% is classified as non-constant pressure
- a %drop of 50% to 30% is classified as LPD
- a %drop of 30% or lower is classified as CP

Yes, it's arbitrary but it's reasonable.

Something interesting to note is that the "operating pressure" of the LRT is not the pressure reported on McMaster-Carr. The pressure spikes when filling and emptying right when the tubing is about to bulge. When filling it spikes to 35 to 40 PSI and when emptying it spikes to 25 to 30 PSI. I think the reported pressures refer to the first spike rather than the operating pressure. I don't know if there is a relatively constant ratio between the first spike pressure and the operating pressure, but if there is, it's about 0.65.

I'm going to update my latex rubber tubing article later today with these findings. I think we've found an excellent alternative to rubber CPS. To recap, the reasons to use precharging are abundant:

- easier to get higher performance because pressure and pipe diameter are not limited
- very often cheaper than LRT and nearly universally cheaper than CAP
- can be more constant than LRT
- does not require maintenance or reduce performance after use like LRT
- is less fragile than LRT
- works like a standard PC unlike CAP

I've attached a picture of the simple device I used to measure the pressure.

[Silence]

Wow, nice find! I guess you just don't notice even the 25% drop in CPS water guns since you'd need to see the streams side-by-side or take measurements. Plus, since there are diminishing returns in everything (pressure to flow, flow to range), that probably ends up being closer to a 5% drop in range.

Maybe we could use 3PD or 30PD to signify exactly how much pressure we're losing. Anyway, thanks for taking the stats - now we have real justification to use chargers.

[Ben]

Some shorthand like 30PD would be fine. I think 30%D would make more sense though because P often stands for pressure.

There's one advantage to LRT that's worth mentioning. LRT is small and can be made into very compact blasters, especially if you use the PC case as a reservoir. I don't suppose prechargers have that possibility, but you can still make them relatively compact depending on the design. LRT still makes sense for some designs obviously and it's not completely obsolete, but I honestly think precharging is the way of the future now after figuring out the numbers behind LRT for comparison.

[aEx155]

I have a reason to add to the precharger positive list: it allows for a more linear design. Using a piston, you get the ability to have the PC in any orientation, meaning that you can have it inline with the nozzle like in CPS guns.

[less important stuff below]

I might try this now, instead of CPS. I really like this idea. In fact, I think I thought about it (haha...) earlier in my "Has anyone tried this?" thread when I started talking about doughnut-pistons and such. That would be a precharger system (if you pressurized the non-water side), right?

Technically, you could attach a separate air chamber attached by a long tube to add to the air ratio, instead of making the whole thing bigger. If you don't get what I'm talking about, imagine adding a 2L bottle, attached by a tube, to the air side of a precharger, and pressurizing the whole thing. That way, there is 2L more of air you could position elsewhere so that you don't add too much weight to the actual gun portion. Makes sense, right?

Sorry, I know this thread is put more toward informing rather than idea posting, but I just wanted to add that in...

EDIT: didn't see Ben's post, but I guess my idea could be a possible solution...

[Silence]

The "long tube" idea is where Ben and C-A 99 were going with the "over-under" design - the air portion doubles back and passes over the water portion, just to reduce the length. It doesn't matter what shape you use there since the plunger doesn't need to traverse that length of pipe.

[aEx155]

Originally Posted by SilentGuy

The "long tube" idea is where Ben and C-A 99 were going with the "over-under" design - the air portion doubles back and passes over the water portion, just to reduce the length. It doesn't matter what shape you use there since the plunger doesn't need to traverse that length of pipe.

By long tube, I meant an air line to connect the 2L bottle to the precharger air chamber. Does it make sense now?

I know that they were going to try the over-under design, but doing that would add a lot of bulk to the gun, which is why I suggested the other idea previously.

[Specter]

Is that pressure gauge the kind that's used with compressed air or is it a different type?

[Ben]

Yes, but that shouldn't affect the reading, only the durability of the gauge. I'm not sure if the gauge can handle water very well without breaking down. It's still working fine so I don't think there's a problem.

[Specter]

I was just wondering because I might use a pressure gauge on something with water and was wondering if an air gauge worked.

[Ben]

A fill gauge should be easy given the piston after some thought. Simply use clear PVC pipe. You'll be able to see the location of the piston exactly and it'll be a perfect fill gauge.