Non-Return ‘Pea’ Type Valve

I could be completely wrong.... but

I think inertia will prevent a 'pea' valve reacting at the rate you are after, it will probably bounce and oscillate.

I'm a bit confused by the upto 22bar vs the 1bar mentioned 3 lines later...

Del

Yes but it may be possible to tune the bounce oscillation [size/weight/material] to aid its operation as this design is meant to operate at 75-100% so its fairly constant.

If you compress 1 bar 22 x it becomes 22 bar [just the way I explained it]

Regards, JD

Can you tell me why this type of valve is needed, as opposed to some other means of backflow prevention?

It is for a kind if IC engine [but not reciprocating] Do you know of any other means of back flow prevention under those conditions?

Regards, JD

How do you want to control the valve? By the pressure difference or via an external system?

It's a passive valve... when a huge sudden increase in pressure from the 'out' side occurs it will slam the 'pea' against the in hole [sealing it]

Regards, JD

I really do think the reed valve would be the way to go! As posted, They have been used in outboard engines, motor bikes, etc, for a long time and show good reliability! My hunch is, if you went for the 'Pea' type NRV, it may need a bit of thinking about due to frequency problems, and also the 'Pea' type NRV will not be as accurate as the reed due to the pea searching for the hole! If you can imagine the pea in an open position, when it is time to change state, the ball will not directly close, it may also bounce round the hole a few times creating a small delay in closing! I'm only talking about maybe 0.01 seconds but by then, you would want to be on the next cycle.

I think I'm right in what I say here, but this is only my personal opinion!

Hi, Thanks for your opinions. When I read up on reed valves it said that they work well at low to medium revs but can't handle very frequencies as the reeds bounce causing leakage and they can shake themselves to bits. Also they can take up a fair bit of room due to the way they need to be mounted. It was suggested that only rotary valves can handle very high operation frequencies. A 'pea' can have guides in the chamber that only allow it to move in and out [not up and down]. Thx, JD

Del's got a point [but luckly his hat covers it]

the cycles 10000/min is in the range of fuel injectors or pnumatic operated valves from auto motive applications, but then you're stuck trying to design control circuits.

simple reed valves would also work & could be obtained from materials to withstand the conditions such valves are used for air conditioner compressors, the trubulence could be could be accounted for through proper shaping of the out let side.

LOL And good answer!

I agree, read valves are a good suggestion, and work well in 2-stroke engines running at 15,000 rpm. The pressure is quite high, but could be managed with a smallish flow hole and local reinforcement of the read perhaps.

To the original poster: what you describe as a pea valve is ordinarily called a check valve.

Thanks for your input. Got any examples/links for reinforced small reed valves? Obviously small holes sound like they will restrict air/gas flow but it's worth checking out. Do you know why they are called 'Check Valves'? Regards, JD

re the derivation of "check valves": I dunno. I suspect the word "check" is being use in its sense to stop or to control. But as you can see here there are many meanings for the word.

No, I don't have any links for reed valves -- although I know there are zillions of possibilities... loads of manufacturers for reed valves for two strokes.

In one of my previous lives, I taught a course in modifying two strokes, an essential component of which was to have the students understand how to design an expansion chamber for a racing bike. The exhaust port in a two stroke is thought (in simplified terms) to be closed by the piston as it comes up -- and of course in a machanical sense that is correct. But the flow direction, at the time the piston closes the port is not out the port into the exhaust pipe -- it is into the cylinder from the exhaust port. So in effect, the exhaust port is closed (to flow out) by sound waves in the exhaust system. This effect is essential for racing engines, and changing the tuning of a pipe can alter the hp of a two stroke by a factor of 2:1: in other words, you can make a 15 hp engine out of 30hp engine by changing the pipe, while leaving the piston porting the same.*

So, that is a roundabout way of saying that perhaps you can use inertia and sound to help your valve work. I can imagine a case in which at very high engine speeds, the inertia in the intake track is sufficient to keep the read valve (of a two stroke engine) open much longer than you'd expect it to be. You can imagine the the reed may never fully close, given enough inertia, but at least in practice, high speed two strokes run well with stiff carbon fiber reeds. (It easy to think of 15,000 - 20,000 rpm as ski high, but of course ordinary guitar string vibrate much faster than that most of the time.)

You could calculate the deflection of the reed from back pressure, and then divide the opening size required for flow into as many cells as necessary to reduce the deflection to a level that would prevent destruction.

Then there are poppet valves, as in the head of an ordinary four stroke. These withstand peak pressures much higher than 22 atm (in fact in diesels, they withstand this pressure even before the bang). The spring required to close these quickly is many times too strong to be overcome by intake differential, but perhaps you could use a positive opening/closing system (desmodromic?). But I suppose you are trying to stay away from such a thing, in the interests of simplicity.

*This relates to the reason why older personal watercraft (Jet Skis, Waverunners, etc) would put about 1/3 of the gas that went into them right into the water, unburned -- something that made environmentalists more than a little upset. The transfer ports and exhaust port are open at the same time, and in fact the exhaust port is open for the entire time that the transfer ports are open. So it is easy to imagine that practically all the new charge would go right out the exhaust pipe -- and in fact, a great deal of it does. A properly tuned expansion chamber stuffs this wayward mixture back into the cylinder just before the exhaust port closes, dramatically improving effiency... but only at the tuned rpm =/-. Thus, the famous peakiness of highly tuned two strokes.

Hi, I always like a cover up if you get my point ;) Have you ever heard of a fuel injector that can have high pressure air blasted through it to mix with the fuel and not let it return? That would be the ultimate but alas I think it's wishful thinking [at the moment]… if you know different please forward the maker and model No. etc. Could you make me a link compressor reed valves and pneumatically operated valves that can handle these conditions as I would like to check them out? Thx JD

A lifetime of 200 million operations at 10,000 times a minute means that running for 24/7 it will only last about 10 days...

Is this the lifetime you will be needing?

John.

Yep! Looks like my maths is as bad as Bush's interpretation of the Bible ;) In some applications it could run continuously for a couple of years but usually it would run between 1–4 hours/day and exceptionally up to 8 hours a day… so 4 [hours/day] x 10,000 [operations/minute] x 60 [per hour] x 365 [days in a year] x 8 [years] = 7,008,000,000

It must be remembered that it's not only number of operations but extreme temperature changes and intermittent operation that causes premature failure.

Regards, JD

If you come up with a suitible valve, please let us know, as I could use a few, too.

Looks like I have avoided the need for a seperate valve(s) and Im very excited about the system which should have at least two patentable devices

Nice one! Hope it all turns out well.

Mr. Brain