Pneumatic Cylinder

[Physics_Kid]

hi all, somewhat related to physics. i embark on a new project, to build my own TV lift using pneumatic lift.
anyone have any links to places that carry thin OD (~3/4") that has about 24" stroke and can lift ~25-30lbs?

i find spending $600 on a electric TV lift absurd, with my metal stock i already have i can build the lift components but need the lifting item. single action is all that is needed, weight of TV will return it to down position. i'll probably need a psi controlled switch to allow enough PSI into the cylinder to keep the TV fully extended, and a good 1way valve for ingress air, and then a solenoid valve to release the air (i thinking a 555 timer in one shot to energize the solenoid for a period of time to allow TV to come down slowly, etc).

all this will be built into a fully wireless rolling credenza. will have 4 12vdc Li-ion batts in series, that 48vdc will power a real sine inverter to run the 120vac 60hz items, so i can build some things in DC and will also have 120vac available.

i welcome any ideas.

thanks.

 

[billy_joule]

Why pneumatics? they probably aren't the route to save money:
http://www.mcmaster.com/#standard-air-cylinders/=11zl0ly
You could probably find cheaper ones elsewhere. Or make your own with PVC pipe, If you use 3/4" pipe you'll run at about 70psi , well under the 380 psi rating for SCH40 PVC pipe.

I'd go with threaded rod (cheap stuff, not ACME) and geared motors, like many cheap 3D printers use for linear motion (except they use stepper motors). You'd end up with something like the electric linear actuators most (all?) TV lifts use but much cheaper.
You could use linear bearings like 3d printers use (~$1 each for LM8UU, though you may want bigger ones) or something like drawer slides from a hardware store.

Much more power efficient and simpler controls than pneumatics too. How were you planning on keeping your cylinders at equal stroke to prevent jamming?
Will your TV be powered from the Inverter?

NB: links are examples only and aren't necessarily to spec.

 

[Physics_Kid]

hi billy_j
i was planning just one air tube, center lift. the led only weighs 22lbs. i will fab the lift mechanics out of aluminum (TIG is my toy). a single action air tube only needs a few small components to operate. i figure a small 12vdc air pump, a one-way valve, and a solenoid valve to do bleed down. i then allow the air tube to pressurize up to say 80psi (enough psi to lift TV to full extend), pump turns off and the two valves are closed holding the air in. after watching TV the power gets turned off, solenoid de-energizers and opens (its a NO valve, etc), then air escapes via a small orifice allowing TV to slowly return to down position via gravity. if the air tube connection develop a small leak the pump would kick back on at a defined min threshold to keep the TV in up position.

i wonder if the inside of std sched-40 PVC is smooth enough for me to make a piston that has dual o-ring. i can do CAD and i have a CNC guy who will machine for me, so i can make a piston and rod, etc. but, was 1st looking to see if COTS was available.

and yes, the power available to the cart is 12/24/36/48 vDC (four 12v batts), and 120vac60Hz via inverter.

 

[billy_joule]

Your plans sounds good. Google will find lots of tutorials on PVC pipe pneumatics, http://www.adamjdahl.com/PDFs/Low-Cost%20Long-Throw...pdfdesign claims "The cylinder described here has passed a 1,000 cycle test with less than 0.001" wear on the seals and components. The PVC pipe has been tested at pressures up to 125 psi."
Or, shorter stroke OTS cylinders are much cheaper so you could design a pulley or lever system to get the required motion from a shorter cylinder.
I would probably design a mechanical catch to keep the TV up (which could be released by a servo or manual override), to avoid the air pump firing up during the movie, and that way, small leaks don't matter and you may save a tiny bit of battery life.

 

[Physics_Kid]

thanks billy_joule for some feedback.
a mechanical catch sounds good up front, but my design relies on the psi as a cushion to slowly allow the TV to fall back into down position. i think i can accomplish this using a NO solenoid valve and experiment with various size bleeder holes which will govern how fast it will take to return back to down position. in general, i flip the switch and that turns on the inverter, turns on air pump, and energizes the solenoid valve. air pump slowly pumps through a small 1-way air valve up to the needed psi and then turns off by way of pressure switch. turn switch off and that turns inverter off and solenoid goes back to NO and allows air to slowly escape.

if i can find a small quiet air pump that can make it up to 100psi(max) (will likely only need about 50psi to lift) i think i can make the pneumatic using just a few parts. i have already sent a CAD file to my machine guy to see what it will cost to make the piston for me (dual o-ring design). i used Apple Rubber o-ring gland calculator to obtain proper dimensions of the piston for use in 3/4" sched-40 PVC. i might however switch over to ~1" aluminum tube because ID of PVC is far from being smooth or round. if the piston part only costs me say $35 i will try it on PVC to see if it will hold pressure.

 

[Physics_Kid]

well, i got my cnc'd piston today, tested it out, got well into 60lbs of force w/o blowing the seals. i only need about ~25-30lbs to lift the TV
this was crude testing of just the cylinder, now need to test if i can pressurize and hold it there.

 

[Physics_Kid]

just thought i would keep this alive, to share some physics at work in application.
i ran test #2, to see if i can keep psi in w/o leak down. more challenging than i expected due to the threaded fittings (thank you soapy water). had to redo the threads more precisely with teflon tape and paste. the piston seems to be holding ok. its been sitting at ~75psi for last 15min or so, need to watch it over next few hrs. a very slow leak down is acceptable but hopefully i can get a few hrs of hold psi to keep the TV in up position.

math and testing is about on par with each other. 75psi with a pipe ID of ~0.800" is about 37.5lbs, my shipping scale said it was 43lbs, which falls into the range of gauge and scale inaccuracies.i think the is feasible as long as the psi can hold there. i will need to choose the peripheral components wisely to avoid dreaded leak down.

i do need to investigate further the use of Nitrile (buna-n) o-rings and PVC. nitrile and chlorinated hydrocarbons do not like each other (if my mem/chemistry serves me correctly, its the chlorine that is the issue). the cylinder is lubricated with silicone grease, but me needs to verify nitrile & pvc.

 

[Physics_Kid]

update - its been hrs and its holding at 75psi

 

[billy_joule]

update - its been hrs and its holding at 75psi

Nice. Looks like it's going to work well.

 

[Physics_Kid]

held at 75psi for ~16hrs, until i manually let the air out.
my design so far has proven success. if i need more force in lift i can up-size the PVC and easily mod my piston CAD file to accommodate. i can likely lift any TV that is less is about 40lbs or less.

the piston design was something i have never done before, had to read a bunch about piston applications and o-ring types, then i used a o-ring groove calculator from Apple Rubber Co. using PVC can also be dicey because ID is not smooth and is way out-of-round, but under 100psi PVC seems to be ok.

i found a cheap air pump on amzon, has set feature with memory, i just not sure yet if i can use that alone because i will have a inline 1-way valve, and when the pump turns off it itself will likely have some leak down, so the pressure in pump will drop and turn back on again for a brief sec as the hose fills with psi, so a nasty feedback cycle i need to avoid (i not sure yet if the pump requires a power cycle to reset the set point, will need to test it). i'll likely use pump set feature as a safety feature, and use a cylinder psi switch as the main on/off pump switch, etc.

 

[Physics_Kid]

update,
the initial pump did not work as needed, replaced with another and much smaller unit. it can pump the psi w/o issue. it is however fairly loud. i might try some PWM control to see if i can slow it down some to reduce noise. i have moved into next phase of testing, a real setup to lift weights to see by experiment psi needed to lift weight. i have a good understanding of what is needed, but now need to see it real-world, etc. i also have the other components to do full control of the lift, but will be tested in phase-3 of testing.

 

[Physics_Kid]

update,
made some more progress. new pump that is plenty for the need.
also made full scale pneumatic. a few changes. piston was redesigned to be about 1/2" taller and just a hair bigger in diameter to provide some extra stability, re-did the CAD and had my CNC guy run me 10 of them, they work great. i also switch to 304 SS rod, much better than the 1018 stuff i was using, SS rod allows me to make the rod hole in cap a 64th" smaller in dia, so some extra stability there. i know it can lift, now getting ready to build a test lift jig to see it lift some mass. i had to switch the upper cap to a 1" pvc item to make the cylinder serviceable (the piston would not fit through 3/4" fitting). in the pic the rod fully down sticks out about 3/4" with a full 24" stroke.

i also have the pneumatic components needed for automation, which contains a flow valve, one way input valve, a digital hi/lo controller, a safety switch, and a solenoid valve for letting the air out. all this will require 9 discrete electronic components to run (plus my inverter, batt, and batt charger). i need to test the pneumatics (2nd pic) for leak down.

 

[Physics_Kid]

arrrgghhhh! testing the pneumatics (less piston because that's already been tested) for leak down and it will not hold pressure. solenoid valve will not hold any psi, so i have to find a better item.

 

[JBA]

Are you sure you have the solenoid valve installed with the pressure connected to the "inlet" side? Most solenoid valves are pretty simple and work satisfactorily but are dependent upon the correct connections to do so..

 

[Physics_Kid]

Are you sure you have the solenoid valve installed with the pressure connected to the "inlet" side? Most solenoid valves are pretty simple and work satisfactorily but are dependent upon the correct connections to do so..

i believe so. the valve itself has no markings but the picture the seller has of it online shows an arrow for flow from the wires side to the no-wires side. it seems like an internal leak on the inlet side because i do not get air coming out on the exit port.

 

[JBA]

You are correct, the arrow indicates the direction of flow and the inlet is on the arrow base end of the valve body. The majority of solenoid valves I have used have a (nonmagnetic) tube enclosing the actuating piston rod in the valve; so, if you are so inclined and can remove the solenoid, you can leak test the connection of that tube to the valve body for leakage (I am just trying to insure that the valve is the leak source before replacing it).

 

[Physics_Kid]

i simply bought another valve to try. it needs to be a working cots item as i don't have time to do the forensics beyond just pinning the issue down to the valve. the problem valve is a low cost MIC item shipped from china, which apparently has quality issue. i'll try new valve and hopefully that fixes my problem.