Hydraulic or Pneumatic Cylinder as a Liquid Pump? 1

[Helepolis]

Hi all,

I need to replace a plastic 3ml syringe on an experimental setup to something more robust.
The syringe acts as a miniature pump, and the pumping action is at a very high frequancy which results in leakeges of air in to the system.
The substance that the syringe is pumping is water mixed with glicerol.

I was wondering if a small pneumatic cylinder will do the job, or I have to find an hydraulic cylinder that can handle liquids less viscous than oil?
The reason I'm asking is that I have little experience with hydraulic and pneumatic cylinders.

 

[MintJulep]

It might work for a while.

If you need a pump why don't you use......an actual pump?

Perhaps

https://en.wikipedia.org/wiki/Peristaltic_pump

Or

http://www.micropump.ca/

Or something else from the entire universe of products that are actually intended to do what you need to do.

 

[Helepolis]

A regular pump wont do the job in the parameters we need it to work.

The system setup is a closed loop, and the syringe is reciprocating the liquid back and forth, at very high frequency of up to 5Hz.

 

[btrueblood]

A diaphragm pump might do a better job at those frequencies. No sliding seal is going to not allow some air to slip through (unless your system pressure was higher than the amplitude of dynamic pressure at the piston face). I'd think a piezo driver immersed in your fluid would give a better frequency response than a syringe - are you actually seeing that frequency at points in the system distant from the "pump"?

 

[SwinnyGG]

When you say "the system is a closed loop" do you mean you have the same fluid on both sides of the piston, and you're actually oscillating flow?

Depending on your fluid, you might be able to build a linear peristaltic pump that would create oscillating flow. The key variable would be whether or not there is a tube material which is flexible enough to use in a peristaltic pump and compatible with your fluid.

This pump would consist of a length of compatible tubing, a few machined parts, a linear actuator (either mechanical or pneumatic) and the controls for said actuator. Which you may already have, to drive this syringe arrangement.

 

[rothers]

Bellows syringe ?

 

[Helepolis]

A diaphragm pump might do a better job at those frequencies. No sliding seal is going to not allow some air to slip through (unless your system pressure was higher than the amplitude of dynamic pressure at the piston face). I'd think a piezo driver immersed in your fluid would give a better frequency response than a syringe

Any type of pump is out of the question, the actuation is precisely controlled by an actuator with precisely tuned pumping profiles.
And yes,

are you actually seeing that frequency at points in the system distant from the "pump"?

Yes, the loop is very small.
It consists of a small reservoir, optical silicone model of an infant lung air ways and the syringe-actuator assembly.

When you say "the system is a closed loop" do you mean you have the same fluid on both sides of the piston, and you're actually oscillating flow?

With the syringe, there are a set of valves which directs the flow; reservoir->syringe->lung model->reservoir.
In case of a double acting cylinder, the system fluid lines can be easily rearranged.

Bellows syringe ?

Thanks, wasn't aware of this type of syringes.

 

[John2025]

Piston or Plunger pump with the check valve(s) removed to just move the fluid back and forth?

 

[drawoh]

Helepolis,

Is a syringe going to operate reliably for any length of time at 5Hz? Do nurses and physicians with syringes operate at 5Hz?

--
JHG

 

[btrueblood]

Sorry, I thought I read 5 kHz and was wondering why in the heck you'd be beeping noise into an infant's lungs. Even 5 Hz seems way to high a frequency, unless you are just pushing to get some kind of cycle testing done?

"Any type of pump is out of the question" -- no. Your syringes are a type of pump too, the way you are using them. So "some" type of pump is apparently acceptable, but you aren't saying much about why. You mentioned valves directing flow, which is what check valves in a plunger pump would also do. If you want to control the valves yourself, fine. Still a pump.

I think a diaphragm (pump) is still the right way to go. You can drive a diaphragm with an actuator as easily as a cylinder. A bellows syringe would work too, as would a rolling diaphragm cylinder. If air entrainment is a problem for you, then you need to design out any sliding seals, imo.

 

[John2025]

Wow, I was sure it was 5kHz, too. I guess when I saw "very high frequency" my mind just jumped to kHz.

 

[drawoh]

btrueblood,

I was thinking 5kHz too, but I re-read the OP.

--
JHG

 

[JStephen]

It occurs to me that you could use something like the peristaltic pump mentioned, only don't make full revolutions with it- just back and forth in a 90 degree arc or so.

 

[btrueblood]

Yeah, but I just don't like the fatigue issues with peristaltic pumps so usually don't consider them. For lab testing it's probably a pretty good idea, though. And tubing is fairly cheap, and controls for stepper motors to drive the pump likewise common and fairly cheap. Presumably there is some type of waveform modulation going on requiring a 5hz capability, and that's quickly achievable with that setup.

 

[Helepolis]

Piston or Plunger pump with the check valve(s) removed to just move the fluid back and forth?

Thanks, I'll look into that.

Is a syringe going to operate reliably for any length of time at 5Hz? Do nurses and physicians with syringes operate at 5Hz?

I'm guessing that it wont hold up its integrity for long, but i don't care.
This setup is meant to simulate the flow field of HFOV - High Frequency Oscillation Ventilation.
Once everything is running as it should, the flow will be imaged using a method called TomoPIV, where an output of 4 high speed,strategically placed, cameras is combined into a 3D flow filed.
So i need no more than a couple of seconds of operation to capture the flow.

"Any type of pump is out of the question" -- no. Your syringes are a type of pump too, the way you are using them. So "some" type of pump is apparently acceptable

I meant that i can't use any commercially available pump, as the setup uses a very precise magnetic linear actuator, so i need to work with it.
The high precision is paramount for the experiment, the total transported volume is about 1.5-2 ml of liquid at high speeds and the actuator delivers exactly that.
And if, i will be able to find a compact, high precision, low volume, high frequency oscillating pump it'll probably cost way to much.

Wow, I was sure it was 5kHz, too. I guess when I saw "very high frequency" my mind just jumped to kHz.

btrueblood,

I was thinking 5kHz too, but I re-read the OP.

Yeah, i shouldn't have written "very high" when it lacks context, but when it concerns ventilating preterm birth, 5Hz is a lot (link)

It occurs to me that you could use something like the peristaltic pump mentioned, only don't make full revolutions with it- just back and forth in a 90 degree arc or so.

Can a peristaltic pump operate at 5Hz?

 

[1gibson]

Set up two bladders, hydraulically one on either side of the component, totally sealed system. Drive a lever arm so when one side goes up, the other goes down, and presses on a bladder. Use a small % of the bladder capacity as the working volume to avoid fatigue problems. When you press down on one, it refills the other.

Drive system could be as simple or complicated as you want, using up/down lever mostly just to illustrate the concept. Pneumatic, linear actuator, rotating cam, reciprocating piston, etc.

For that small volume, you could probably just squish some rubber tubing, the bladder is mostly conceptual.

 

[moltenmetal]

A diaphragm pump with check valves will do this well.

A piston/plunger pump with check valves removed will also do this very well.

A syringe pump also exists, but will not like running at 5 Hz forward/backward.

 

[btrueblood]

"Can a peristaltic pump operate at 5Hz?"

Yes. Finding one off the shelf to do that is probably going to be tricky. I'd purpose build one to do it with a stepper motor and controller, and a bit of software. But you have to use a linear actuator, so what's the point of asking?

 

[gerhardl]

This post has gone from answering the original question to suggestions on how to solve this with other mechanical devices or pumps.

On use of cylinder built for air in general: I have successful (and with permit from the producer) used pure water directly to operate air cylinders of aluminium. Answer: a standard air cylinder built for normal pneumatic pressure, and with parts compatible with water/glycol will in general function.

Quite another question is if you will find one cylinder on the commercial market that will suit and function within your requirements.

What is the size and size restrictions? How does it operate? Suck in as a syringe in a single stroke, and pump out in one longstroke or several smaller? Or another way? One way in (port one) and anotherway out (port two). Time limits in the sequence?

More information is necessary for design. Checking the internet: cylinders with 5 mm cylinder boring seems to be about the smallest available? Too large?

Good Luck!

 

[EngineerTex]

Hope I'm not too late to the party.

You could possibly get away without making many changes at all. If the problem is only that you are getting leakages of air into the system, simply rearrange your existing setup such that the syringe is arranged vertically and then simply fill the back side of the syringe cavity with your working fluid. If you get any leakage, then what is leaking in and out is the same working fluid you are already using. Or leave the syringe horizontal and use silicone to mold a case to keep the entire pumping end submerged in your working fluid. Or if you want to be really cheap and clever, use tape and a plastic bag. I like cheap and clever.

Second alternative is to machine a new plunger out of UHMW or aluminum and put 2-3 seal grooves and seals in it. Assemble the barrel and plunger submerged in the working fluid and you should be able to prevent any fluid from passing out of or into your loop.

Engineering is not the science behind building. It is the science behind not building.