Hydro Differential pressure exchange over unity system.

[see3d]

@see3d: I think, if I am understanding you correctly, that I have actually experienced the effect you describe in the single-layer Zed I have in my PerPump v.2.2. The pumping pressure seems to me to behave like you have described, when I have the starting water levels in the Zed chamber just right (apparently just enough to allow the riser to make a gastight seal between the ringwall and the outer chamber, a 1/4 inch or so.) Even in my crude system, I think I'm seeing both the enhanced pressure and the negative pressure effects, due to the volume changes as things move around. I don't have a way to lock my riser/pod combo except by varying its weight (buoyancy at a given displacement) to make it heavier so it stays down longer, and my range of motion is limited to under an inch, but even so, I see the effects on the water output that correspond to what you are seeing in the sim, if I am understanding your description. I interpret this in my PerPump as a force multiplier, like a hydraulic lever, that trades off increased pressure for shorter pumping time... in other words, less volume pumped at a greater pressure. I reported this effect and demonstrated it in a couple of videos last week sometime, but nobody around here seems to pay much attention to Koalas any more -- even if one of them does have a self-powered table top waterpump demonstrating the active Zed principle in action by pumping 13 1/2 inches of water head above its outlet fountain.

TK, Thanks for reporting that.  Here is the thing that I missed in my first sim that I have included now, which makes it make sense to work this way:

The pod is not going to move up unless the "spring" force between the piston and riser is equal to the output load.  Just like if you had a stick with a spring on the end of it and you had a bowling ball sitting on platform with a hole under it.  When you first push the stick with a spring up under the ball, the ball will not move until your spring has compressed to the force of the weight of the ball.  Then it will move up one-for-one with the stick movement.  Of course, the stick "feels" the whole weight of the ball.

In my ZED sim model, there are two springs.  One spring is the Pod Head as a normal buoyancy type force plus the air PSI, making the head seem higher (or lower if negative PSI).  The other spring is the air PSI x the Pod Head area.  The two interact, but do not have the same mx + b factors.  The water head has more gain than the lift from the air PSI, because it has to move further to make up for the volume expansion of the air as the Pod moves up.

The net result is that for my sample model, I have a peak work transfer efficiency of about 65% at around 1/4 inch of riser movement -- which is well before the air PSI turns negative.  So a short stroke is going to be better.

Playing around with the dimensions, I can see a taller and thinner gap and wall model getting up to 85% efficiency.  Keep in mind that this is yet an unproven sim model, and only one layer.

So far, in my sim, everything is consistent with what Wayne said was the direction for a small table top system design point.

[see3d]

see3d, I do not have a locked pod to riser, but when I run with an oversized pod the risers will sit on it when in the down position, right now I am running the pod very lite and so it takes very little water into the chamber to make it lift, but even set up this way the pod does not stay stuck to the riser until enough water has been added to make it stay there, it will go up then settle for a moment while the risers are still moving and then lift up back into the riser.

Webby, that is interesting.  It may be the same effect of a negative force, but uncoupled to the Pod in the middle of the operation.  That bears watching carefully.

[TinselKoala]

@see3d:
Hmm. OK, so you have a single layer Zed that can probably attain a transfer efficiency of 85 percent.... By transfer efficiency you mean the ratio of work output to work input?  I'm assuming you do for the moment.

So ok... now can the single layer Zed then be treated as a "black box", like an internal pod, which then allows the second layer to be calculated in the same way as the first?

What I'm trying to understand is how a nested five layer system of 85 percent efficient modules can ever attain an efficiency greater than one.

(0.85⁵ = about 0.44)

[see3d]

@see3d:
Hmm. OK, so you have a single layer Zed that can probably attain a transfer efficiency of 85 percent.... By transfer efficiency you mean the ratio of work output to work input?  I'm assuming you do for the moment.

So ok... now can the single layer Zed then be treated as a "black box", like an internal pod, which then allows the second layer to be calculated in the same way as the first?

What I'm trying to understand is how a nested five layer system of 85 percent efficient modules can ever attain an efficiency greater than one.

(0.85⁵ = about 0.44)

TK, a single layer is not the same transfer equation for the next layer.  This is the overhead layer to get things rolling.  If the single layer math pans out in the sim, then I bump it up to a 2 layer sim.  I think 3+ layers will all use the same general function, but I am not sure if 2+ would be the same, so I will play it conservative and just add one more layer as a special case first.

I am not extrapolating my single layer result to a future result, and I don't advise anyone else to do that, because it isn not valid.  I am following my painfully slow step-by-step method to understand the ZED from the ground up.  I will leave it to others to judge the final performance from constructed models.  I am so slow and methodical about my methods, that I am sure others will be way ahead of me by just taking the shortcut of building something first.  When I build something, I want to know that my first model will work well before I start.  The methods I use to build are dependent on the design criteria from a sim model.  That is not to say that there is not real value in doing rapid prototypes to test certain theories -- because I do that too, when needed, but they are pretty ugly... LOL

[fletcher]

@see3d:
Hmm. OK, so you have a single layer Zed that can probably attain a transfer efficiency of 85 percent.... By transfer efficiency you mean the ratio of work output to work input?  I'm assuming you do for the moment.

So ok... now can the single layer Zed then be treated as a "black box", like an internal pod, which then allows the second layer to be calculated in the same way as the first?

What I'm trying to understand is how a nested five layer system of 85 percent efficient modules can ever attain an efficiency greater than one.

(0.85⁵ = about 0.44)

TK .. I don't understand the 'work transfer terminology' either ?  - And I also currently can't see why nesting risers will make any difference & should lead to well over OU ?

......................

See3d .. I would have thought the obvious way to test efficiency in [& integrity of] the sim [assuming it has no losses] is to calculate PE start & top of cycle ? - Energy Input into the piston should equal the PE gain at any vertical height cross section after start - no where should it exceed 100% ? - this applies to releasing the riser to do work as well.


N.B. In the single riser layer ZED sim PE + KE joules gained should never exceed joules of energy input [ f x d ] - if it does at any position then OU is being indicated then & there IINM & gravity would not be conservative.

P.S. for those following the piston raising PE of fluid while riser is locked down is doing the same job of injecting fluid into the riser chamber under pressure I believe.