Z axis counterbalance, pneumatic cylinders, air reservor, fittings, etc.

[1Jumper10]

Not a double acting cylinder. A single acting , spring return, air cylinder with 2 ports. Most spring return cylinders have only one port but Norgren and SMC make a model with 2 ports. There are probably others, I just did a cursory Google search. Spring return cylinders have a spring inside them, they extend when air pressure is applied to the rear port. They retract when the air is exhausted. But if you connected the two ports together, the pressure on each side of the piston would be equal and cancel each other out and your Net force then would only be the force of the spring. As the piston traveled through its stroke, air would flow from one port to the other. If you put a flow controller there, you could control how fast it reacts by restricting the flow from one port to the other. I think the effect might be that small Z movements happen quickly but longer Z travel would take longer to equalize pressure. It might take a pretty large spring to counterbalance your Z axis. Maybe more than one. The spring return cylinders I've used have been on the small side. I've never tried this, only theorizing here.

[1Jumper10]

Just to add a little more info: The flow controllers I've used are adjustable one-way valves. You would want to orient it so air would flow through it unrestricted when the Z moves up, and restrict it when it moves downward. They can be adjusted very finely. And come to think of it, a spring extend cylinder might work better,... a standard double acting cylinder might work to since the area on the two sides of the piston isn't equal due to the rod on one side of it. Probably wouldn't be much net positive lift in normal size cylinders.

[NIC 77]

Not a double acting cylinder. A single acting , spring return, air cylinder with 2 ports. Most spring return cylinders have only one port but Norgren and SMC make a model with 2 ports. There are probably others, I just did a cursory Google search. Spring return cylinders have a spring inside them, they extend when air pressure is applied to the rear port. They retract when the air is exhausted. But if you connected the two ports together, the pressure on each side of the piston would be equal and cancel each other out and your Net force then would only be the force of the spring. As the piston traveled through its stroke, air would flow from one port to the other. If you put a flow controller there, you could control how fast it reacts by restricting the flow from one port to the other. I think the effect might be that small Z movements happen quickly but longer Z travel would take longer to equalize pressure. It might take a pretty large spring to counterbalance your Z axis. Maybe more than one. The spring return cylinders I've used have been on the small side. I've never tried this, only theorizing here.

I do appreciate your input. Always good to think of new ideas with other people.

The force on both sides of the piston wouldn't be equal, it would be different by the rod diameter times the pressure. At lower PSI's with small rods, it would be small, so not a huge factor for our practical purposes with this problem. Traditional gas struts use this, but are loaded to higher PSI's and incorporate additional dampening features.

The problem is, now you are relying on a traditional spring to counterbalance the Z axis. In a traditional spring Force = K x X where K is the spring constant and X is the distance of elongation or compression. So the force varies a large amount between the top and bottom of the stroke. I assume that the springs are not extremely strong on these, just enough force to reliably return the piston when pressure is removed. I actually don't know, I haven't seen the spring specs listed in the Amazon ads I looked at. But whatever it is, it's not adjustable, you can't change it by changing the pressure.

This would be a good application perhaps for someone who wants a traditional spring with a rod and guide attached, that can have customized dampening added through an adjustable flow restriction valve as you mentioned.

Unfortunately, unless I missed something big, which is entirely possible, it won't work for my application.

I do enjoy and value your input, certainly I thank you for the info about the push in pneumatic connectors. So please keep posting.

[1Jumper10]

No, it probably wont. I've got enough parts at home I can try it out. I might need a similar system for my router. The Z axis stays put without power but just barely.

[NIC 77]

No, it probably wont. I've got enough parts at home I can try it out. I might need a similar system for my router. The Z axis stays put without power but just barely.

Yes, it could help. If you could find out the spring constant you could do a couple of quick calculations to see. If the spring force is such that at the bottom of the stroke where the spring force is greatest it provides more force than the weight of the Z axis, and at the top of the stroke it provides less force than the weight of the Z axis, then at some mid point it would be neutrally balanced.

Overall, the motor would have to work less over the entire operating range even though it would work more in some areas than others.

Or if at the bottom of the stroke the spring force equaled the weight of the Z, then it would mean that the Z motor would have to work least when near full extension, and progressively more as you got higher up, but still less than it did before.

I wonder if spring induced oscillations could be an issue? I've never actually seen anyone use a traditional spring on a CNC Z. I wonder if the motor might not like it for some reason? IDK, it might work as an assist on a machine that could get by without anything, but a little assist is nice.

For my purposes, with a 192 lb monster, I am looking at an air spring solution. Routing the tubing through the cable chain and putting the reservoir outside simplify things. I used a few pressure loss calculators and even using a single 6.5mm ID tube and a 15 foot hose, at the flow rates we're talking about, less than one cubic foot per minute, its not an issue. Even with a 5mm ID tube or even less it's not an issue. So I'm glad Jim Dawson talked me into doing it that way.

The issue for me is leakage and if this can be done with a closed system, that is perhaps only topped up once a week. That's the big question mark. If that's not a problem, then the rest is a straight forward matter of a few calculations, and picking the right fittings, and finding a cheap reservoir.

Another thing I've just been looking is the effect of temperature on a closed system.

So let's say that the lowest temperature for my machine is 15 degrees C, and the highest is 30 degrees C.

That's 288.15 Kelvin and 303.15 Kelvin.

P2 = P1 x (T2 / T1)

So at 90 PSI and 15 degrees C, increasing to 30 degrees C, P2 = 90 x (303.15 / 288.15) = 94.7 PSI, so that is acceptable for my purposes.

[1Jumper10]

Thinking a little more about this for my system, I'm not going to need an assist for my z axis. My servo is about 5x stronger than it needs to be. I'm just going to need some type of brake to keep it from descending when power is removed.

Sounds like you're going to get a good solution in place for your build. Looking forward to seeing it.

Sent from my Pixel XL using Tapatalk

[bobmagnuson]

Did you ever get the pneumatic assist working for your z axis???

[NIC 77]

Did you ever get the pneumatic assist working for your z axis???

I'm still building the machine.......My Z axis is a monster. Perhaps a great idea, perhaps a foolish one. I'm committed to seeing it through.

I ended up ordering the pneumatic cylinders from a different company than the link I posted on Aliexpress, I think because the sales person I asked was helpful with helping me sort out the fitting threads. The ones I have are stainless steel tubes, but the tops and bottoms are aluminum. I have them pretty much fitted now, one of them has a small alignment issue which I will try to sort out today actually, not sure what's causing it yet. I also got some silencers that will be used on the open ports to act as filters so debris can't get into them.

They came labelled as 0.7MPa max which is less than the advert said (1.0 MPa), so I was a bit bummed, but I still gave a good review to the store because I had specifically told the salesman I would use them at less than 100 PSI, so my advice there is to make sure the sales person knows what you need. The quality also looks great and the fittings look good too, but I don't yet know how well they will work under pressure and it will be a while before that happens.

The hole in each bottom, and for the swivel fittings I purchased for the tops are 10mm. So I am using 10mm bolts to mount them. I've taken longer bolts that have an unthreaded section and used a die to add threads up to where I need, so that the holes in the bottom and the top are free to rotate on an unthreaded portion of the bolt. I haven't designed them to rotate when extended / retracted, but if the alignment isn't perfect then there may be a very small amount of rotation that you don't want to restrain. You can also buy premade brackets to do this, but they take up more real estate.

So the answer to your question is that I'm still working on it. I can let you know when I'm done. I believe it will work.

[bobmagnuson]

Yes, please post when you're done!

Do you have any pics of your current progress??

[Dan911]

This thread brings back some memories' , I experimented with this a few years back when I purchase a 3kw spindle and after realized it exceeded the weight limit of the K2 Z I purchased 2 weeks before...lol.

I picked up 3 used pneumatic cylinders on ebay supposedly from an aircraft, with a precision regulator, and have to say worked well for some1 like myself who didn't know what I was doing. I had a video of me lifting 3kw spindle with a finger with cylinder attached but couldn't find. I eventually upgraded Z with profile rails and ballscrew and eliminated.

Here's the thread....


http://www.cnczone.com/forums/diy-cn...w-spindle.html

[bobmagnuson]

Thanks Dan911, that's exactly what I'm hoping to do with the beast of a spindle I bought to replace my 3.7kw unit.

[bill south]

Hey Guys;
Dan, can you sketch up a schematic of the air circuit you created that worked with the twin cylinders. Also, do you know the brand and model number of your air regulator. I'm having trouble wrapping my tiny little mind around this concept. And as usual, I'm most likely over thinking it!!!
Many thanks.
Bill

[Dan911]

Hey Guys;
Dan, can you sketch up a schematic of the air circuit you created that worked with the twin cylinders. Also, do you know the brand and model number of your air regulator. I'm having trouble wrapping my tiny little mind around this concept. And as usual, I'm most likely over thinking it!!!
Many thanks.
Bill

Hi Bill,

It really wasn't that complicated and I probably got you beat with the tiny mind, the schematic can be seen in pics of thread I linked. All the 1/4" compression fittings and hose I purchased from Automation4less. I used a tee on output of regulator and fed the bottom of each air cylinder and used air filters on top of each cylinder, the input of regulator was connected to a 5 gallon tank compressor.

The regulator reads:

SMC Precision Regulator
IR2010-N02

I couldn't find any markings on the air cylinders.

Dan

[bill south]

Thanks Dan!
I really appreciate the info!
Bill

[NIC 77]

A couple of thoughts.......

Dan's system has the cylinder rods in tension, where mine will be in compression. In tension, you have to subtract the diameter of the rod for the pressure calculations. In tension, you don't have to worry about buckling.

Having both ends fixed (as opposed to pinned) increases the work to make sure everything is aligned properly, but also reduces the chance of buckling if using the rod in compression. In tension, I don't see any advantage to having both ends fixed, unless you find it easier to build that way.

I had a look at some math, I should be fine for buckling, but I am also rethinking if putting the rods in tension may have been a better idea. The 10mm diameter rods do look a bit weensy in comparison to my mega Z.

The big question for me is if these will leak. If I only have to top up the reservoir occasionally, then I will be fine. In a closed system like mine a regulator isn't necessary. If you add a regulator then you will need to have it hooked up to a compressor all the time, as it will constantly be dumping air. If you look through the math in this thread you will see that even with a medium sized reservoir, the changes in pressure at top and bottom of stroke aren't enough to be concerned over. That of course all depends on if it leaks or not. If the systems leaks appreciably, I'm kinda hooped. If it doesn't leak appreciably, the regulator and constantly running compressor aren't necessary.

I will be using a simple T fitting to combine the pneumatic lines from both cylinders to a single line that will run to the reservoir.

[1Jumper10]

Twin 10mm rods are probably a little overkill, but that just means they will be reliable so, good call. I just watched a YouTube video of a Shop Sabre router with a pneumatic counter balance and they used it in tension. And it looks like they also used vacuum, probably along with a spring return cylinder. The air line connected at the top of the cylinder. The only way with that configuration to get it to pull in , and counterbalance the Z is to use vacuum. They're spindle was a big 10hp monster so they're counterbalancing a good size load. Just an observation.

[NIC 77]

Twin 10mm rods are probably a little overkill, but that just means they will be reliable so, good call. I just watched a YouTube video of a Shop Sabre router with a pneumatic counter balance and they used it in tension. And it looks like they also used vacuum, probably along with a spring return cylinder. The air line connected at the top of the cylinder. The only way with that configuration to get it to pull in , and counterbalance the Z is to use vacuum. They're spindle was a big 10hp monster so they're counterbalancing a good size load. Just an observation.

ShopSabre, there's a good example! Yes, they are using it with the counterbalance rod in tension.

Found this video



That one clearly shows an air port at the bottom being used, so compressed air instead of vacuum.

Standard atmosphere is 14.4 PSI, so my thoughts are that this is the biggest pressure differential that you can get with a vacuum system, which typically isn't enough (depends on the bore of the cylinder), so with a large spindle, and one cylinder with a small / medium sized bore, I believe these systems used compressed air, not vacuum.

It certainly looks like they have one connection to the cylinder in the next video, at the top, which would have to be a vacuum (I don't know, perhaps it's hidden at the back?) but 5:00 into the video he mentions they use compressed air. I will add this vid in the next post, as I can only add one vid per post.....

[NIC 77]

[1Jumper10]

The cylinder could have a spring in it. You would have to add it's force the to the vac force (x cylinder bore ). I guess they could be running a low pressure in there if the spring was strong enough. Seems unlikely though. The spring return cylinder springs aren't usually monsters.

[bobmagnuson]

Great posts. The Shop Sabre Z axis is what I've been using in modeling my new Z.

Now if I can figure out exactly how they did it!