Proportional valve for water, low pressure 4

[johannestoger]

Hi,

We're designing a flow experiment where we need to produce a controlled pulse of water in a pipe. Our current design idea is based on a constant head tank, a flow meter, a proportional valve and a control system as follows:

Image from Gharib et al. J Fluid Mech 1998.

External factors (room size, imaging equipment etc) limits this design to about 1.4 psi pressure in total. Previous researchers have used ASCO 8202, which seems like a perfect fit. However, the 8202 has a too low Cv - 1.1 gpm in the largest configuration - and we need to go to about 6 gpm for the flow rates we need, at this low pressure.

Does the valve forum know of a valve similar to the 8202, but with a higher Cv?

Thanks in advance,

Johannes

 

[DRWeig]

If you need solenoid type, Burkert goes up to 1" size Cv 5.8. Asco seems to stop at 1/2" and Danfoss stops at 3/4". Neither will get you to Cv 6.0.

http://www.thevalveshop.com/menu/auto/burkert/6223.html

If there's not a reason to stick with a solenoid valve, there are tons of choices.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[johannestoger]

OK, thanks, I'll look into the Burkert valve!

The Cv value is not set in stone - my rough calculations tell me that Cv around 5-6 should be appropriate.

The reason that I want to use a solenoid is that we need a fast reaction speed - we need to open the valve and close it again in about 300 ms.

Do you know of any other type of valve that can be used?

 

[johannestoger]

Oh and another thing - the Burkert and Danfoss valves have a stated minimum pressure of 7 psi (0.5 bar). The highest we can go in the current design is about 1.5-1.7 psi.

So I'm thinking: Will these valve not open at all at lower pressures? Or will they work, but not optimally?

 

[DRWeig]

These are pilot-operated valves. In my experience, they will not open at lower than the minimum pressure.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[xnuke]

I know from experience also that the Burkert 6223 will not open with less than 7 psi upstream.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[steamdog]

You may need to use a small valve with pneumatic actuator, with a large solenoid on the air line that can open and close the valve VERY quickly

 

[btrueblood]

Might work with a pneumatic actuated butterfly valve, but the fastest actuators I know of would not respond much faster than 1 second for a close-open-close cycle. I could diagram a way to make a simple one-way plunger that could deliver the flow impulses you want at any volume/speed/stroke you could want, but you ought to be able to figure it out yourself: put a check valve on the head of the plunger, so it draws from the inlet tank on the return stroke of the plunger, and discharges though the outlet check valve on the forward stroke. Drive it with a regulated air supply, or from a hydraulic pump and variable speed motor, or via a motor and drive screw.

 

[johannestoger]

Thanks a lot for very good answers!

@btrueblood: The plunger idea sounds interesting. We have been discussing putting a piston in a pipe, as in the diagram below, is this what you are thinking of?

This design has some nice advantages, including that we can get very exact and repeatable pulses with the right driver. Our main issues with this design are 1) getting a good seal between the piston and the pipe when pressurized and 2) making it doesn't wear down over the several thousand cycles we need. I'd be very happy to hear what you think.

 

[gerhardl]

You do not exactly describe your 'controlled' pulse of water. Your flowmeter seems to indicate a controlled amount, variable length, and variable (from pulse to pulse) cv.

Suggestion: If you separate the flow regulation from the opening/closing regulation you might come better out with commercial available components.

In series : a) A standard, but fast, on/off solenoid valve with good repeatability, large enough cv, operating from about zero delta P, and fast opening/closing time (soft opening/closing or not?). See for instance attached data-sheet.

http://www.buschjost.com/pdf/E82090.pdf

(Kv 8 at 1", available also in NPT). Your problem might then be the abrupt opening/closing, but other direct acting types or brands might give faster opening/closing times.

b) Regulating valve, hand operated valve to give correct combined cv.

An idea?

 

[johannestoger]

@gerhardl - Thanks, I'll look into it.

By controlled pulse I mean that I'd like the flow to vary as v = V*sin[sup]2[/sup](pi*t/T). Here T is the pulse time, about 300 ms and V is the peak flow, about 0.6 m/s.

 

[btrueblood]

Johanne, yes that is what I meant. A good quality stainless steel hydraulic cylinder will likely work, but you may also get good results from some of the cheaper aluminum ones. Making your own is not that tough - piston seals can be simple o-rings for the low pressures you are talking about. Finding a good source of smooth, corrosion resistant alloy tubing might be the toughest part. We once built simple low pressure cylinders from standard drawn copper tubing (water piping), the i.d. finish is typically mirror-smooth.

 

[moltenmetal]

A stainless pneumatic cylinder might do the job.

The proportional solenoid valves don't work at low pressures as already mentioned.

 

[johannestoger]

Thanks everyone for great answers!

 

[johannestoger]

@btrueblood: Would plastic tubing, e.g. PVC work? Or is polished metal a better choice to minimize wear on the O-ring?

I submitted the cylinder/piston idea to the fluid mechanics forum, feel free to chip in there:

http://www.eng-tips.com/viewthread.cfm?qid=342215

 

[btrueblood]

Don't know about pvc pipe - the i.d. finish can be pretty rough on larger size pipes, but I've never really looked at the i.d. of say 4" pipe that closely. Glass tubing certainly works (doesn't get any smoother than that), but end caps are tricky, you'd want to make up a tie-rod cylinder and put a good cushion on the ends of the tube, or use a fairly soft plastic for the end caps.

 

[johannestoger]

@btrueblood: Thanks, I'll keep the cylinder roughness in mind.