PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2020-11-25, 14:32:17
News: A feature is available which provides a place all members can chat, either publicly or privately.
There is also a "Shout" feature on each page. Only available to members.

Pages: [1]
Author Topic: Hydrostatic Displacement engine build  (Read 418 times)
Group: Moderator
Jr. Member
*****

Posts: 57
The purpose of this thread is to document my attempt to build a self-running ‘Hydrostatic Displacement’ engine.

My goal from day-1 has been to present an open source self-running engine that can be built by a layman with some fabrication skills, common tools, and easily obtainable materials.

First of all I want to provide some background and make it clear this is not an over-unity machine. Without exception it is based on the principles of established physics, and introduces no changes to them. Although I named it a hydrostatic displacement engine, the motive force is supplied by gravity. It is by utilizing the external force of gravity that energy will be brought into the system.

The premise the engine is based on is Pascal’s Principle.
This is stated as:
“A change in pressure at any point in an enclosed incompressible fluid at rest is transmitted undiminished to all parts of the fluid and acts at a right angle to the enclosing walls.”
Look it up if you wish and you will most likely find an example of a hydraulic lift along with it.
Although the fluid used is not incompressible, it is nearly so and may be considered as such for practical purposes here.

I’ve attached a 2 page excerpt from a book on hydrostatics from Cambridge university first published in 1895 about a hydrostatic bellows invented by Pascal. Please review it and take note of this section near the end of the pdf.
"Hence, if A is large compared with a, W will be large compared with w.
By making the area of the board considerable and that of the tube small,
a large weight W can be supported by a small weight w of water.
This fact is sometimes described as the hydrostatic paradox:..."

Therefore it stands to reason the weight of the liquid and the force that weight can develop with a piston are two separate properties, or quantities if you will.


« Last Edit: 2020-10-25, 19:48:20 by Cadman »


---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
Re-posting this here for continuity.

The material chosen for the cylinders is 6” schedule 40 pvc, which has a pressure rating of 120 psi. If you attempt a build of your own with this material make sure you do not get pipe made with pvc foam. Schedule 40 pvc will have the pressure rating printed on it.

The interior finish of the pvc is not smooth enough to provide a seal and needs some work to make it smooth. I made a hone out of a wooden closet rod and split on one end to hold several sheets of sandpaper. The paper is held with 2 bolts through the rod. The other end was fitted with a 1/4” steel rod so a variable speed drill could operate it.

I gave the interior a light coat of flat black spray paint then made a few passes with the hone.  You can see how bad the factory wall finish is. It took an hour’s work, 4 sheets of 60 grit followed by 2 sheets of 120 grit to get a decent finish. The wall still has some slight dips in the circumference but these are handled by the flexibility of the leather piston seals.

« Last Edit: 2020-10-27, 00:44:38 by Cadman »


---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
I do not own a lathe so most of the circular parts are created with a 3D printer using PLA. The software used to model the parts and create the .stl part files is FreeCAD v0.18.4.

The most time, by far, has been spent developing a valve for this engine. For this build I have settled on a large spool valve made with 2” sch. 40 pvc along with 3D printed spool pistons, valve seat, end caps and seal retainers. For now, the mechanism to shift the valve is going to be a small air powered cylinder attached to one end of the spool shaft. Each spool piston will have a rubber disk seal for the seat and a leather u-cup seal at the opposite end. A pressure balance tube will connect the outer ends of the spool housing together. Care will have to be been taken to provide a relatively tight fit between the spool pistons and housing while not allowing the piston to touch the housing during it’s travel. The housing will likely need a little honing too. The pistons will be waterproofed by a 10 second immersion in hot candle wax (200 F) followed by a careful trim back to the outer dimensions after completely cooling. This will leave a skim coat of wax to act as a lubricant in case the piston contacts the housing during run-time. The seals will be held against the spool pistons with a disk of PLA and the lock collars that hold the piston on the shaft.

The seat is a short inner tube with a port opening in one side located in the center to act as the valve seats. The rubber disk on the end of each spool piston will press against the end of the seat to seal off one port, directing the flow from one side to the other. The seat itself will be held in place with a 1/4” dia. Bolt and nut.



---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
I spent a lot of time trying to find a good source for leather. Forget about calling leather suppliers, they think leather is worth more than gold. A good inexpensive source is Amazon or your nearby welding supplier where you can pick up a leather welding apron.
Like this one:
https://www.amazon.com/Safety-Shop-Protective-Blacksmith-Improvement/dp/B07NZVFDKD/

The aprons are roughly 1mm thick, can be cut with scissors, and make good piston seals with soft and flexible edges. One apron will make all the seals you need with plenty left over.

The next thing you need to do is make the seal forming tools. Those are just a set dies. A small one can be made from some 1/2” PVC trim board, a piece of the pipe the seal is to fit, a threaded rod the same size as your piston rod, some nuts and flat washers You’ll need a forming disk (punch) that’s at least as thick as the raised lip at the edge of your finished seal, with a clearance between the pipe the same as your leather thickness. You need to be fairly accurate with the clearance, the leather needs to be compressed a little against the pipe wall. You can see I had to add a little diameter to the large punch with some blue masking tape. The large die set I made from a scrap piece of 2 x 8 treated lumber with a coat of poly. The pvc ring has a small radius sanded on the inner edge and the punch has one on the bottom edge. I used a 2mm radius on the small punch and a 6mm radius for the large one. You also need a way to push or pull the punch out of the seal.

You’ll need an Exacto knife or sharp box knife for this next step. The leather should be cut the diameter of the cylinder it’s to fit + 2 times the height of the seal lip + about a 1/4” (6mm) extra. Before you begin cutting, punch or cut a hole in the center, the same size as the piston rod. Then center your outer diameter on that. Then you can cut your blank.

To form the leather, soak it in hot tap water until it darkens all over on both sides. Rub the surface to remove all the tiny air bubbles that cling to it, then soak it some more. 10 minutes total will do. Push the wet leather onto the threaded rod followed by the punch disk, a flat washer and the hex nut. Tighten the nut snug enough to bottom the punch into the die and completely flatten the bottom of the leather. Take your knife and neatly trim the excess leather away while it’s still wet.

Let each seal air dry for a day then pull the punch and carefully remove the leather.

The next step is to melt your wax or paraffin by heating to F 200. That can take hours so be patient and allow plenty of time. Don’t go any hotter or you’ll cook the leather. Drop your seal into the hot wax for about 5 seconds, take it out until the wax ‘skins’ then put it back in until it turns a milk chocolate brown. That can take a minute or two but not longer. If the leather starts bubbling like a french fry in hot oil when you drop it in the wax, yank it out immediately or it will cook into a hard useless chunk of hide.

After the seal cools, flex the raised edge between your fingers until it’s fairly flexible. You can see it turn a lighter brown as the wax coat fractures.

That’s about all there is to it.



---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
The valve housing is cut and assembled. The port holes were cut with a 2” hole saw and the short port tubes were notched with a 2-3/8” hole saw. 4-bolt flanges were added to all the ports so the valve can be removed for maintenance or modification if need be. The flanges are cut from the 1/2” pvc trim board. I simplified the balance tube with elbows instead of half-couplings too. On a side note, the type of pvc glue to use is the heavy duty kind that has a thicker body to it, not the kind that's just liquid acetone and MEK. It has to be able to fill any small gaps in the fit.

Also printed the hollow spool pistons and the corresponding piston seal retaining disks. The pistons have been wax coated and trimmed. The flat seat seals are cut from a 6” square of rubber plumbing gasket material from Menards. By the way, make sure the gasket material says ‘rubber’ on the package and not ‘rubberized’ which looks identical but is junk.

The valve seat is finished and a retaining bolt boss printed and attached with super glue. Printed the housing end inserts with nylon bushings and waterproofed them with poly, and finished the bolt-on end caps with buna-N shaft seals.

Shaft
https://www.mcmaster.com/8934K26
Nylon bushings
https://www.mcmaster.com/6389K627
Shaft seals
https://www.mcmaster.com/9691K51
Set screw shaft collars
https://www.mcmaster.com/9691K51



---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
Fabricated the air cylinder. Waiting on some 1/4” NPT fittings and air line to be delivered.

Redesigned the lower half of the displacement piston, which is printing as I write this. The last one had a flat surface on the underside which I realized would trap some air. That is a something to avoid as it can kill the engine operation.

Starting on the cylinder to valve assembly tomorrow.



---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
Finished the cylinder to valve assembly. The new displacement piston lower half turned out well. The OD after cooling is 5.945” which leaves plenty of piston to wall clearance. The pistons should not touch the cylinder walls at any point.

Pay close attention here, this is a critical component assembly.

The images below show the old thin displacement piston and valve assembly from the previous prototype.

Image disp_piston_disk_valve-1 shows the piston seat plate with the leather seal and blue plastic tensioner that pre-loads the seal against the cylinder wall. On the right from top to bottom: stepped disk guide, 2” flat washer weight, floating valve, and rubber seal. The seal is the same material used in the valve assembly.

Image disp_piston_seat_plate shows the four raised seats. These prevent the fluid from getting trapped between the rubber disk and the seat plate, and the narrow contact area increases the sealing pressure against the rubber.

Image disp_piston_disk_valve-closed shows everything as it is on the piston rod. The stepped guide and the piston are held snugly between two set-screw lock collars. The rubber seal and weighted floating valve are free to slide up and down about 1/4” (6mm).

The image disp_piston_disk_valve-open is self explanatory.

The last image is what I use to seal the printed PLA parts.




---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
Progress is a little slow right now. Cutting and deburring over 80 pieces of threaded rod and drilling over 30 holes is drudge work. And there’s more to come. But 3 of the 4 cylinder end plates with seals and bushings are done. The fourth one I managed to bugger up. That one was already cemented to the upper cylinder so it will have to be cut off and the cylinder end refinished. All 4 of the 7” square steel end plates are cut and drilled too.

To make the task of accurately centering the end plates to the cylinders easier, a spoked ring was printed to snugly fit in the cylinder bore with a 3/4” hole in the center. My rod is 3/4” chromed cylinder rod stock and the ring slides over it without any play. The end plate was bored and counter sunk for a 3/4” nylon bushing then a ring was attached to it while a short piece of rod aligned everything perfectly. Since the end plates are 1/2” pvc plate, the ring has an inner flange with 4 screw holes to attach the ring. The spokes and center ring will be removed before final assembly. The ring is also glued with pvc cement. I should have used super glue, but, didn’t think to in time. I blame that mental lapse on old-timer’s disease.

And I also printed the 4 housings for the rod seals.

The drawing below shows the cylinder and piston layout. As you can see from everything posted so far, the engine layout is simple. By the way, all of these drawings are not exactly what ends up being built, but they do show the idea. If things work out, and there is any real interest, I’ll modify them to show the ‘as built’ details. That said, this drawing does not show the cushion ends that will stop the piston at the ends of it’s travel.



---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
These images show the flow of the engine and a semi-cutaway view of the finished engine.



---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Group: Moderator
Jr. Member
*****

Posts: 57
Well, I sort of shot myself in the foot.

Originally this engine design was conceived with a 6 inch stroke. Over time, I managed to damage one end of the pvc power cylinder. I thought to myself this wouldn’t be a problem, I’d just shorten the stroke down to 3”. After all, that wouldn’t change the thrust at all.

The problem with that is in the air compressor part of the design. Having solved that earlier, I shelved it and didn’t give it any more thought.

It takes 13.5 cu in of compressed air to shift the valves, plus the air line volume, or about 14 cu in total.
A 2” piston with a 6” stroke can deliver 16.5 cu in of air at 2 psi, and develop about 6.3 lbf of resistance on the power piston. However, with a 3” stroke the piston diameter has to be increased to 3” (working with pvc pipe) and that will deliver about 18.6 cu in at 2 psi, but the resistance jumps to 14.3 lbf. More than double. Obvious once you think about it. And if it turns out I really need 3 psi then the resistance climbs to 21.2 lbf. Which is half the thrust available and this isn’t even allowing for seal friction which is substantial with those buna-n rubber seals.

So it looks like I have to take the valve and cylinder assembly apart, cut and hone another cylinder, rebuild the piping, and put it all back together.





---------------------------
Something isn't false, or fake, just because you don't believe it.
   
Pages: [1]
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2020-11-25, 14:32:17