Surgical: Low pressure open, High pressure closed valve 3

[MeritRD]

I'm a newbie here and not an engineer, so please forgive the naivete. We have a sensitive pressure transducer used in a medical procedure that blows at over 120 psi and it's useful range is obviously below that. However, the system it is on can generate pressures up to 1200 psi. Currently, protocol calls for the physician to close the stopcock on the manifold going to the pressure transducer when the system would generate the higher pressures and then open the stopcock to the pressure transducer after to be able to measure pressures again.

I'm wondering if a style of valve exists (I don't need to buy one, I just need to see a design we can replicate with tubing IDs of 3 mm) that will be open at lower pressures, but then close with high pressures. That part is easy. The hard part is that on the other side of the closed valve, the pressure cannot go above the lower threshold or the pressure transducer will blow. For example, if it is something like a spring loaded plunger where the spring must be overcome by the high pressure, then the pressure generated "behind" the valve will go beyond the lower allowed pressure because of the mechanical movement of the plunger (or trap door, or whatever) and blow the transducer.

so, to summarize, it needs to be low pressure open, high pressure closed and protect the low pressure side from pressures above the threshold without a vent or blowoff (the system needs to be able to attain and maintain the higher pressures for use) and then return to open for the lower pressure. Seems like everything I've seen does the opposite.

I thought a solution might be to have an open-system blow off valve where excess pressure is relieved out a hole, yet this will not suffice because of the complications of dealing with the waste. However, if the system could bleed off only the tiny bit of volume moved by the mechanical movement of the plunger (or whatever) into an on-board reservoir, that might work. Since the volumes are so small, this would only be a drop or so of liquid. This is a crude solution and you guys probably have something that is out there for a different application that will solve my problem.

another possible solution i thought of is a plunger where one end is open to pressure and the other is open to atmosphere, but is supported by a spring that compresses only when plunger receives over 100 psi. the plunger contains a transverse fluid pathway that is open to the measured pressure system, but when the plunger moves with higher pressures, it closes the transverse pathway because it is protected by the cylinder wall.

Thanks for putting up with my non-engineer-speak and the long post. urls with pictures of internals of examples would be great.

dan

 

[hydrae]

MeritRD

It looks like you need a pressure reducing valve and a pressure relief valve.

The pressure reducing valve would be on the line to the transducer and would have a set point of ~100 psi. If the supply pressure is below 100 the valve is wide open allowing the transducer to perform its work. After the pressure reducing valve a branch would contain the pressure relief valve. The pressure relief valve would have a setpoint of 110 psi and vent any excess pressure that happened to get by the pressure reducing valve provide protection for the transducer.

Now to your application. These type of active control valves are not something you could slap to gether out of tubing. Inside the valves are springs and sensing chambers that adjust the valve position based upon pressure in chamber against a spring (the spring tension is used to set the adjustment)

To see the inside working of pressure reducing valve

http://www.cla-val.com/pdfs/E-990.pdf

Notice there are multiple springs inside and the controling chamber piston has a large surface area on the downstream side and a small surface area on the upstream side

The pressure relief valve could be something like a clamp held down with a spring over the tubing
hope this helps
Hydrae

 

[iainuts]

There are valves called "gauge protectors" such as from Circle Seal, that operate very much how you describe. Here's a web page:

http://www.circle-seal.com/pdf/1100_series.pdf

My experience with Circle Seal in the past has been marginal. You may or may not get prompt service.

Another avenue would be to essentially make your own device. If there's a gage snubber in the line to the transducer, and there probably should be, then simply putting a low pressure relief device between the gage snubber and transducer would work. I'd suggest a very cheap relief device such as any hardware store might supply at 100 psig. If you'd like a recommendation for one, I'd suggest using a Generant, here (they're around $20):

http://www.generant.com/crvMain.htm

The suggestion above, about using a pressure regulator will also work well, and there are many regulator manufacturer's out there. Of course, the reg needs to be rated for the full system pressure, 1200 psig or above.

One word of caution, you also should consider, "What will happen if this device fails?" In the case of the snubber and relief valve, if the relief valve fails open (ie: it leaks either slightly or completely open) and the pressure drops, what are the reprocussions? What if the transducer thinks the pressure is 50 psig, when you actually want 100 psig? or 50 psig, when it's actually 1000 psig?

Don't add any old valve to the system without running it by your safety folks at least. A management of change, or whatever may be needed.

 

[MeritRD]

thanks for the input. the cutaways are helpful. the only problem is that we have to keep in mind these are disposable and have to be dirt cheap. i came up with a design that i think would work, but it would be too expensive to be viable. see:

http://hoopes.net/dan/merit/overpressure_beta_open.jpg

then, when i began to brainstorm the cheapest ways, and came up with this:

http://hoopes.net/dan/merit/overpressure_beta2_open_closed.jpg

however, that one probably wouldn't work because of the gymnastics i'm asking this piece of rubber to do. lastly, i came up with this one and we'll probably prototype something like this:

http://hoopes.net/dan/merit/overpressure_open.JPG

and closed:

http://hoopes.net/dan/merit/overpressure_closed.JPG

the rubber "bulb" in the middle would have atmospheric pressure on the inside, like a regular rubber ball, and under high pressures would want to decrease in volume, thereby "squatting" down and compressing against the sealing edge. additional pressure would increase the seal because of the "cup effect" seal lip around the bulb. the little corners inside the bulb are to protect against over-crushing under very high pressures and might need to be modified/enlarged.

if it works, it will be 4 cheap pieces of plastic and rubber and fill the need. anyone see anything that might not work about this kind of design? remember, we're not talking about flows here, this is static pressure measurement of basically a water (no air bubbles) system that is connected to human blood pressure. the overpressure protection needs to come on when fluids are forcefully injected through the manifold/tubing into the bloodstream at the high (up to 1200 psi) pressures mentioned.

comments on these possible solutions from you pros would be greatly appreciated.

dan

 

[iainuts]

Hi Dan, I took a look around your web site. Congrats on getting admitted to UCI. What are you going for?

From looking at these sketches, I'd say you're going to have a few problems...

The rubber bulb for example, isn't going to simply move into the flow passage and block the port. It'll shrink because of the low air pressure inside. It'll just sit there like a grape on a vine, whithering up into a raison. LOL

And the bi-metal thing I simply can't imagine how it might work. I assume you're thinking the thin plate will get pushed in by the pressure, which will cause the part to bend over and into the flow stream. But I don't see that happening actually. Bellows assemblies can do what you're looking for but they're very expensive and subject to damage if the pressure gets too high. The cost of making that part in the middle is prohibitive.

The first one is the only one that might stand a chance of working, but it's overly complex with a bunch of extra ports. And the sensing element needs proper sizing, the O-ring seat isn't captured... lots to go wrong there. Actually, with a few minor changes, #1 can be made to work and if purchased in quantities of about 50 or so, the cost might be as low as $50 each. Larger quantities will obviously be less expensive.

If only a handfull of parts are needed, I'd suggest using one of the suggestions given in the previous posts. It's much less expensive to spend a few hundred or even a few thousand dollars for a handfull of parts than to have one specially made. Tooling, set up, etc... are going to eat your shorts if you go for plastic or rubber parts in low quantities.

If you need larger quantities, or if what's available doesn't suit your need for any reason, contact valve manufacturers and see what they can provide. Many times, they have valves that can be made to work, or parts of assemblies from mass produced valves can be used to cut costs. If you go that route, I'd be interested in seeing what you are looking for.

 

[PUMPDESIGNER]

I'd sure like to be the guy selling you the pressure transducers. I am not in the biomed field, but I cannot recall many industrial designs where the system relied upon someone to close a valve to protect instruments except in extreme cases.

Lucas Chavez used to sell a transducer that protected itself against mammoth overpressure by means of a sealing piston that was spring loaded. As pressure increased the piston was pushed upwards eventually sealing itself and the sensor off from the system.

If system has no energy storage (non-compressible fluid with no storage devices or accumulators), such that pressure fluctuations are very fast, so fast that no device can react, then you are looking for a solution that does not exist except by your present means of manually protecting the transducer.

If your system pressure fluctuations are slower because you have some energy storage (compressible fluid and/or accumulators), then you can skin that cat in numerous ways, take my wife's cats if you want, both of them.

PUMPDESIGNER

 

[MeritRD]

iainuts: thanks for laughing at my possible solutions. It only proves that i'm on the right track to solving a problem that most people think has no easy solution. Hey, they laughed at Columbus and Edison.

about the rubber bulb withering up, i agree. however, if the dimensions and bracing of the bulb were such that it wanted to flatten (thicker at lateral edges and thinner at top and botton corners). while it would take a lot of fine-tuning, i believe it would work.

about the first one working, sure it would work, but in this application, all of the inside corners and nooks and crannies will create noise in the pressure readings. also, i presents complications for clearing all air bubbles before use.

These have to be cleared of air bubbles to get good readings and they are disconnected and reconnected and flushed maybe 10 times a day, before each procedure.

we manufacture all of this stuff, so buying big valves and parts is not the best solution for us. we would make a plastic mold and sell tens of thousands of these a month. that's why i'm trying to see how it might be done in larger or more common applications and then replicate the paradigm in a small, molded part.

for example, something that uses springs and steel parts in another application could probably be done with a little compliant mechanism or carefully crafted piece of rubber.

btw, i've contacted a couple of valve manufacturers and they have the same reaction i did: "this is the opposite of what a self-actuated valve should do."

 

[MeritRD]

PUMPDESIGNER: actually, we make and sell the transducers to clinics doing these procedures. if a doc forgets to turn the stopcock to protect the transducer its toast and they get to buy another one. i agree that it's a pretty lame way to do it, and that's why i'm working on a solution so the docs can just go for it and never have to think about it.

if you have ANY info on the Lucas Chavez transducer you mention, i would be very interested. that's exactly what we're trying to do.

about not having any energy storage: an accumulator is a possibility and thanks for bringing up that point, but there is another way, i think. between this overpressure valve and the transducer, i could put in an over-pressure blow-off cuff. it's basically just a hole in the tube with a rubber collar around it. when the pressure gets too high, it can escape through the hole by pushing the collar outward slightly. for this application, the collar would have to be very stiff to resist up to about 500 psi, but that's just a matter of finding the right dimensions and durometers.

that way, when the valve needs to close there can be flow that acts as the energy the close the valve. the pressure will not damage the transducer because any extreme pressures will be bled off by the overpressure cuff.

note that i can't just use a cuff to solve the whole problem because i need to maintain those high pressures in the rest of the system to get them to inject correctly. however, the cuff would allow probably a few drops out, enough to provide flow and close the valve that protects the transducer.

so, the problem is really just how do i make a valve that closes itself under high-pressure flow.

i don't know what exactly you mean by slow and fast pressure fluctuations. they usually hook the system up to a power injector that ramps up the pressure pretty quick to get this media into the heart so they can take their x-ray picture.

about skinning the cat: 1. what are the numerous ways? and 2. do you kill the cat first or just start right in?

 

[iainuts]

Hi Dan,
Sorry if I gave you the wrong impression. I'm not laughing at your solutions. The principal is there. In #2 and #3, your pressure sensing element is a flexible rubber or metal part which uses some pressure inside the element to compare to the system pressure. The pressure on the inside is your reference pressure. Metal bellows use this principal. They often are evacuated in order to produce a consistent reference pressure. In theory, they could be made to work.

Here's a few thoughts about using rubber or plastic that need to be overcome with any design using these materials as the sensing element as you've shown. First, there needs to be some way of preventing the bulb from extruding into the flow passage since rubber is so soft, and the pressure (1200 psi) is so high. Second, with rubber or any plastic being used as a sensing element is that the material will leak, even when brand new. Rubber is permeable to gas, and to a limited extent liquid too. That's why balloons loose air, even at the fractional psi levels they experience. The rubber bulb might work the very first time, but loosing or gaining gas or liquid internally will be an issue over time. Another issue is adjustability. There really needs to be some way of adjusting the set pressure of something like this. Getting any accuracy on set pressure may not be possible without some way of adjusting. The solution is to use ambient pressure as the reference pressure as you show in #1.

The first drawing you show uses ambient pressure as a reference pressure. Here, you show a piston exposed to ambient on one side, and the process on the other. The spring makes it adjustable and the whole thing works. This is exactly the principal used for the mechanism on the Lucas Chevez transducer that PUMPDESIGNER suggested, and it's fundamentally the same mechanism a pressure regulator uses. It's simply a sensing element (diaphragm or piston) exposed to process pressure on one side and a reference pressure on the other. The sensing element is comparing these two pressures. A spring is used for adjustment, and a seat is used to close off the process. So don't worry about corners, bends or other internal features creating noise, that doesn't happen.

For 10,000 pieces a month, I'd expect the cost for something like this to be around $5 each, give or take. Maybe less if the design can be incorporated as a cartridge into the transducer.

 

[MeritRD]

iainuts: i just went back and reread your first post and realized that i didn't understand what you were saying on the first read through.

actually, you're right, the claims of the gauge protector are exactly right on what i am looking for. however, i can't understand from their pictures what is where. i tried calling them but just got a machine and left a message. where is the gauge in the diagram? also, where is the reference for the check valve?

about the second solution, i don't know if it would work. after researching the gauge snubbers, i found out that they just smooth out pressure spikes and still allow the full pressure to reach the gauge after a short time. placing a regulator after it would allow the fluid to escape to atmosphere, just more slowly. this might allow the ~10 second injection to take place, but it would probably create too much of a mess and bleed off too much fluid to be a usable solution.

also, a snubber is not currently used because the sensor has to be very sensitive and respond quickly with no smoothing at the low pressures (up to about 100 psi) and then everything above about 120 is known to be garbage and it doesn't matter what the reading is. that's why they just shut it off with a stopcock currently and don't even get readings during that time.

i am very interested in understanding the theory behind the "gauge protectors" and possibly seeing a diagram of a Lucas Chavez transducer.

thanks for the valuable info, iainuts. also, thanks for the congrats on UCI, I had just put that up the night before I'm thinking about surgery for now. however, i've heard that when you get into 3rd year and start seeing what the different lifestyles are really like it starts to skew your desires back to reality (I still want a life after my residency). however, i have to do something that will allow me to take part in innovation of tools and techniques and use my IT master's for something. but every specialty has room for more tech and innovation

Dan
Merit Medical
Research and Development

 

[iainuts]

Hi Dan,
Regarding the gauge protector from Circle Seal, I see you managed to get in touch with their fantastic customer service department. LOL It's not one of their strong points. The gauge protector is oriented with the relief valve closest to the transducer being protected. The whole valve works by popping the relief valve, which suddenly creates a large amount of flow through the check valve which closes the valve due to the dynamic pressure drop across the valve. This works just like an excess flow valve. Once the check valve shuts, the valve is held shut by the high pressure as long as it remains on the inlet of the check valve. Note that the back side of the valve has been bled off by the relief valve which shuts once the check shuts. Once the high pressure on the inlet to the check valve drops, the spring in the check valve allows it to unseat and open.

For your application, if you're looking at large quantities, I'd say none of the suggestions above really give you a cost effective solution. I believe the least expensive solution is still around $20 - $40.

The easy solution is to have a piston with a seal on it as the sensing element just like you show it in the first version you drew. I'm sure that's all the Lucas valve is. The only issues with what you show are essentially minor design changes to minimize size, cost, and to ensure proper function. My guess is that valve should be on the order of $5 give or take, given quantities of 10,000 although there is an open question as to how large it has to be.

I had a friend from college that went on to surgery. He spent at least 12 years after undergrad studies doing that. It really does suck up huge amounts of time. Best of luck.

Dave.

 

[PUMPDESIGNER]

I sorry about taking so long to get back, was out chasing a cat.

By fast pressure changes I refer to pressure transients, extremely fast virtually instaneous changes in pressue common in systems with no energy storage, i.e., no flex in the tubing, non-compressible fluid, no accumulator, etc. Anything that fast is well beyond the ability of any device to react or respond quickly enough.

I am not experienced in your field, so I am sure my priorities are not correct, but I step out to stimulate my mind I suppose, as long you you do not mind.

Energy storage could be as simple as a coil of tubing that can provide some flex and stretch, that would tend to moderate pressure transients so that they do not occur.

I will look for that Lucas Chavez transducer info, been many years. Simple concept though, a spring loaded piston partially blocked the port leading to the strain gauge. The back side of the piston (towards strain gauge) mated to a seat with an "O" ring. If pressure increased the piston backed up onto the seat sealing off the strain guage so that it was no longer exposed to incoming pressure. Was used on the space shuttle originally, and I think it was developed for the shuttle. Simple, really simple.

PUMPDESIGNER

 

[MeritRD]

------------------------------
sorry this is so long, but i thought that you "valve people" would be interested, especially considering that a few people in the industry told me this was impossible with all our requirements. both work and i'm going to demo them for our CEO to see if he wants to move forward and produce them.
------------------------------
Well, I've got something that might be interesting. I created a couple of prototypes and they do what i want them to do. here's links to some images.

A CAD image of the intended design:
(this one has two linkages, but actual doesn't, for now)

The smaller, white-handle one:
(for scale, the syringe part of the model has an ID of 8 mm)

-clear view of the linkage

-(3) shows the functionality the best. The high pressure (system) side is the bottom and the protected side where the transducer would be is on the top. Note the straight fluid path (the plunger end rests against the normal fluid pathway so that air bubbles are easily cleared during setup.

The larger, blue-handle one:
(scale: the syringe body ID is 13 mm)

The reason none of the solutions you guys mentioned would work is that we need an uninterrupted and non-tortuous fluid pathway. The other valves I was able to find after looking into all the ones mentioned are the following (they may be helpful to someone in the future):

Circle Seal Controls, Inc. Gauge Protectors: 1100 Series
problem: lots of snubbing, tortuous fluid pway, releases fluid to environment

Chemiquip Pressure Limiting Valve: PLV 5460, 5500, 6430
good: exactly what i am trying to accomplish. this is pretty much what i had drawn in my first drawing (see above links) without ever having seen one.
problem: too much snubbing from tiny fluid orifices and a tortuous fluid pathway

Wika Adjustable Gauge Cutout: Series 680
problem: very tortuous fluid pway using needle valves and tiny, tiny orifices

Basically, since we have a bunch of surgical stuff around, I just started cutting, machining and gluing and finally got something that works. There are two prototypes. Both use the same basic idea: a syringe body and plunger is the sensing element with some kind of spring resists the movement and sets the activation point. A wire linkage mounts to a screw post in the syringe plunger and connects it to the stopcock. The stopcock is a standard one and therefore needs no o-rings for sealing and has a completely unobstructed and straight-through fluid pathway.

The stopcocks are different pressure ratings. The blue one is rated to 1200 psi and the white one to about 300 psi, but higher when it is a short pulse. This also affects how hard it is to turn the handle. The blue one is much harder to turn (about 4 lb-in torque max!) and therefore needs a stronger spring to return it to open/neutral position. This means that the sensing element (plunger) must be a larger diameter to provide the needed force at the specified pressure. The white handle one is much easier to turn and the spring and therefore the plunger diameter can both be smaller.

The white one uses a spring inside the barrel of the syringe body and the blue one uses a torsional spring around the stopcock body. this is mostly just to see if both ways would work.

both prototypes work exactly as needed. the blue one actually works faster, which is important, because the spring acts directly on the stopcock which requires the higher forces to move. the white one is a little slower because the spring must act through the linkage and all associated slop to activate the stopcock handle.

thanks,

dan