PDA

View Full Version : Pneumatic power unit Rev1.1



Evan
03-17-2006, 07:57 PM
I made a few changes to my design. Thanks for all the suggestions. I have incorporated some of them. Any comments are welcome. Maybe I have overlooked something else. I'll explain a few of the features later but it's quitting time at the store so I'm outta here.

http://vts.bc.ca/pics/cylinder2.jpg

Main features:

Stroke is adjustable from zero to maximum.
End of stroke dampening works for all stroke lengths.
Stroke speed is adjustable.
An accumulator is provided with cooling.




[This message has been edited by Evan (edited 03-17-2006).]

david_r
03-17-2006, 11:08 PM
Evan,
Can you explain exactly what the check valves are doing? If they are just there to allow the one needle valve to control both directions, I think you could get around that without check valves.

As you have it drawn, your hydraulic piston is the limiting stop on your forward stroke. That may be what you intended.

Design the hydraulic seals as if it was dead-headed. We all have a friend that can't keep his hands off of stuff. My friend would definitely turn that valve closed :O Not maliciously but he can't help touching stuff.

Rustybolt
03-17-2006, 11:51 PM
It's not gonna get hot enough to need fins.

sdeering
03-18-2006, 12:55 AM
What is the hole in the top of the oil res. for? Fill?
The big needle valve, I can’t see it doing anything but being a mechanical stop?

On the oil side, if you are just using it for controlling the speed of the cut stroke (slow) and the speed of the return stroke(fast) you could use a hydraulic cyl. piston arraignment with a line coming off of each end. Use a needle valve to limit the cut stroke and have a parallel line with a check to let it return fast.

On the air side you may be able to find an air operated diaphragm liquid pump to use for the power part or to copy the control or copy a steam powered piston water pump control. Just some ideas. Good luck.
Stephen

Evan
03-18-2006, 02:39 AM
http://vts.bc.ca/pics/cylinder3.jpg


1: This side of the air piston will touch left on the out stroke before the hydraulic damper bottoms, just barely.

2: The relief around the shaft here allows oil flow from check valve at 4 when the cylinder starts moving right on the return stroke.

3: Note the sightly angled cut on the left side of the left oil port. This produces a gradual cutoff of oil trapped on the left of the oil piston on the out stroke providing a cushion at the last 1/8" of travel. Check valve 4 prevents it from escaping through the piston rod to the right.

4: This check valve forces oil on the left side of the oil piston to flow through the needle valve on the power stroke (moving left). On the return stroke (moving right) it allows oil to flow from the right side of the piston to the left side freely. As the volume on the right is greater than the volume on the left (no shaft on the right) the net oil flow will always be through the needle valve (8) on both strokes. The needle valve controls speed on both strokes.

5: The large needle valve is the stroke length adjuster. It also provides damping action on the return stroke. Since on the return stroke the oil is forced through check valve 4 and blocked by check valve 6 it must flow through the center bore of the piston rod to the left side on the return stroke. Near the end of the return stroke, regardless of the stroke length, the needle cone on the stroke adjuster gradually blocks the flow through check valve 4 as the opening around the needle decreases in the last 1/8" of travel. It seals just at the end of stroke. This provides a cushion action on the return stroke.

6: This check valve allows the right side to fill from the accumulator freely during the power stroke. During the return stroke it forced the oil on the right to flow through the needle valve 8 by passing through check valve 4. The ball of check valve 6 is retained by a small piece of music wire below it. Pressure on the return stroke close the valve.

7: This is a vent for the tank.

8: Flow control needle valve for both strokes. Speed will be approximately the same in both directions but slightly slower on the return as 18% more oil must pass through it on the return stroke. A fast return is not required as the cycle time will be fairly fast anyway.

The unit will be cycled at a rate of about one stroke per two seconds so considerable heat will be generated in the oil. Cooling fins are probably a good idea and besides, they look neat.

david_r
03-18-2006, 01:19 PM
Evan,
Thank you for the detailed explanation.

On your extension stroke, your damping is going to drop off when the right side of your hydraulic piston reaches the front port. The hydraulic fluid will bypass the piston as the port is wider than the piston. I don't know if this is a concern for you.

Your delta volume is the same in both directions, right? If you can tie the left and right side together and control the flow of fluid into the accumulator, you can achieve equivalent damping in both directions. Look back at your original design and add an accumulator with a needle valve between it and the horizontal passage.

I still don't see how you are sealing your glands. Hydraulic fluid will find a way to leak. It looks easy enough to add an o-ring to the front one. There are many different designs out there that use o-rings. I haven't seen any that use face gaskets.

I'm still not clear how you are mounting the accumulator to the cylinder. I think you mentioned face mounting in your previous thread but your drawing looks like you are piping it. If I may suggest, use face mounting with o-rings if you have the capability to machine the grooves.


Thanks for letting us be part of your design process.

Evan
03-18-2006, 01:49 PM
"On your extension stroke, your damping is going to drop off when the right side of your hydraulic piston reaches the front port. The hydraulic fluid will bypass the piston as the port is wider than the piston. I don't know if this is a concern for you."

On the extension stroke when the hydraulic piston left edge passes the port the only escape for the remaining fluid is through the wedge cut at #3. This takes over from the needle valve for motion control.

"Your delta volume is the same in both directions, right?"

No, it isn't. That was the mistake I made in the first design. The volume is greater on the out stroke than on the in stroke because of the space occupied by the shaft.

"I still don't see how you are sealing your glands. Hydraulic fluid will find a way to leak. It looks easy enough to add an o-ring to the front one. "

I'm not quite sure what you are referring to. I do show an O-ring on the left side of the hydraulic cylinder between it and the air cylinder (just to the left of 2). I don't mind a bit of leakage there as it will lubricate the air cylinder and just be exhausted with the air as with an air tool. The right side I omitted the seal on the stroke adjuster/damper which will probably be a compression seal which will also act as a lock for the adjustment.


The accumulator will be mounted with screws (not shown) and the pipes connecting the assemblies will have O-rings on the pipe ODs (also not shown) in compression between the tank and the cylinder.


Thanks for your comments and questions. It makes me think about the design. The first design looked good to me but was completely unworkable.

david_r
03-18-2006, 03:30 PM
<font face="Verdana, Arial" size="2">On the extension stroke when the hydraulic piston left edge passes the port the only escape for the remaining fluid is through the wedge cut at #3. This takes over from the needle valve for motion control.

3: ...Check valve 4 prevents it from escaping through the piston rod to the right.</font>
In order for that check valve to function as above, your piston needs to be thick enough to cover the port at full extension. If it isn't, that fluid will bypass via the port into the back of the cylinder as soon as the right side of the piston passes the port edge.


<font face="Verdana, Arial" size="2">No, it isn't. That was the mistake I made in the first design. The volume is greater on ....</font>
Of course. That was a good catch in your other thread but I didn't say total volume, I said delta volume. Move the rod one inch and the volume change is X. One direction the volume reduces by X the other direction the volume increases by X.

Let me try another tack. Take your first design and add an open standpipe (call it an accumulator) to the horizontal passage. When you retract the cylinder, the fluid will rise a certain distance in the standpipe. When you extend the cylinder, the fluid will fall the same distance. Delta V is the same. So if you can throttle the fluid entering and exiting that stand pipe, you will have the same damping force in both directions and not need any check valves. If you'd like to add a check valve, you could end up with a fast retract but you'd need to redesign your stop screw to provide more damping force. Maybe a parallel section with a small port?


<font face="Verdana, Arial" size="2">I'm not quite sure what you are referring to. I do show an O-ring ....</font>
You show SHCS holding the hydraulic cylinder together. At these points, how are you sealing? The left side shows a boss that could take an o-ring. The right side doesn't look like there is enough material anywhere to put an o-ring to keep the fluid leaking through the cylinder/endcap junction.


<font face="Verdana, Arial" size="2">The accumulator will be mounted with screws (not shown) and the pipes connecting ...</font>
If you have the tooling to mill a flat on the cylinder and to make o-ring face grooves, you can eliminate the pipes. This is how hydraulic valves are commonly stacked. You buy a port adapter, connect it to a ported plate and connect that to a valve body. They are sealed with face o-rings and bolted together. Very robust.

[This message has been edited by david_r (edited 03-18-2006).]

Evan
03-18-2006, 06:11 PM
"In order for that check valve to function as above, your piston needs to be thick enough to cover the port at full extension. If it isn't, that fluid will bypass via the port into the back of the cylinder as soon as the right side of the piston passes the port edge."

That makes no difference. The damping in the last 1/8" of stroke to the left is produced by forcing the remaining fluid through the wedge opening on the side of the port. Where it goes doesn't matter.

"When you retract the cylinder, the fluid will rise a certain distance in the standpipe. When you extend the cylinder, the fluid will fall the same distance. Delta V is the same. So if you can throttle the fluid entering and exiting that stand pipe,"

Maybe I am misunderstanding but trying to pull fluid through a needle valve won't work. It will just pull a vacuum if the fluid can't flow fast enough. I can't see any way to do it without check valves except for extending the shaft out the right. That would make variable stroke very difficult to implement.

Check valves are easy to make anyway. I had to rework the valve body on the pump on my Land Rover and add a check valve. It will hold the blade up for days without drooping.

Incidentally, I always thought that Delta V means "change in velocity".

As for sealing, I will probably use paper gaskets with Permatex silicone sealant. I built a valve body for the engine oil radiator cooling system on my truck that I made years ago and the valve has never leaked one drop.

I can mill a flat no problem and that is a good idea. Scratch the pipes.

john hobdeclipe
03-19-2006, 12:42 PM
I think you're almost there. A couple more thoughts:

The large adjusting screw 5 for the return stroke limit will require a seal to keep oil from working out past the threads, and also a locking device to keep it from backing out on it's own.

The resevoir vent will need a filter.

Move needle valve 8 as low as possible into the oil resevoir so it won't pick up air even if the oil is sloshing around a bit. Perhaps bring it in from the top, with the metering orifice near the bottom of the resevoir.

Do you really need check valve 6 ? I think you can control the return speed well enough with a flow control in the air line, and the cushion effect provided by 4 and 5 will keep it from hammering at the end of the return stroke.

And perhaps check valve 4 can be eliminated also. Give the adjuster 5 a long tapered point (3/4") to push into a matching long tapered cavity in the rod, and see what happens. This should allow a rapid return stroke with ample cushioning.

edit:

On second thought, the check valve at 4 is necessary, although it may work better and provide better flow if it was built into the piston.

I need to take another nap.

[This message has been edited by john hobdeclipe (edited 03-19-2006).]

Evan
03-19-2006, 10:20 PM
"The large adjusting screw 5 for the return stroke limit will require a seal to keep oil from working out past the threads, and also a locking device to keep it from backing out on it's own.

The resevoir vent will need a filter.

Move needle valve 8 as low as possible into the oil resevoir so it won't pick up air even if the oil is sloshing around a bit. Perhaps bring it in from the top, with the metering orifice near the bottom of the resevoir.

Do you really need check valve 6 ? I think you can control the return speed well enough with a flow control in the air line, and the cushion effect provided by 4 and 5 will keep it from hammering at the end of the return stroke."


John,

You must have missed this comment of mine. "The right side I omitted the seal on the stroke adjuster/damper which will probably be a compression seal which will also act as a lock for the adjustment."

Yep, filter on the vent.

The needle valve won't pick up air. The hydraulic damper system is essentially a hydraulic pump and with very minor modification this entire unit could be used as an air powered hydraulic pump with a pressure capability of up to 2000 psi at low volume when operating on 120 psi air. Both check valves are required for the pumping action to work as the flow is out throught the needle valve on both strokes. Valve four provides the seal for the leftward stroke and valve six provides the seal for the rightward stroke. Better oil seals would be required but that isn't too hard.

All that would be needed to make this a real hydraulic pump is an air shuttle valve actuated by the piston rod to provide fully automatic operation. Pressure oil would be tapped just before the needle valve and would return to the accumulator tank. The needle valve would control oil pressure.

john hobdeclipe
03-20-2006, 02:46 PM
Yeah, I did miss the statement about the seal on the threads of the large adjuster screw.

I still think that you can eliminate check valve 6. You'll still get full control on the out stroke, where it is most important. You'll have a faster return stroke, and have to be sure that the return stroke is well cushioned at the end. I'm just thinking of eliminating machining steps, eliminating parts, making everything as simple as possible.

With only a few modifications you can have a fuel pump for an EXTREME pulsating BILLET FLAMETHROWER.

Evan
03-20-2006, 03:00 PM
Yep, it may work without that valve. I wanted control over the return stoke too but it might be OK without it. It certainly wouldn't be hard to try as that valve is simply screwed to the bottom of the tank as a separate unit. One thing I would worry about is possible cavitation on the left side during the return stroke.