Idea for 4 stroke rotary valve.

[Rayanth]

if your intake port and exhaust port are open at the same time, it's either during the end of the exhaust stroke (intake opened too early) or intake stroke (exhaust closed too late)

Why valve overlap is bad:

If it's during the exhaust stroke, then your intake port is open while you're pushing exhaust out of the cylinder. In a combustion engine, this means you're backfeeding 'bad' air into the carb, or other air supply system, and will make the engine run REALLY rough if it is forced to try and burn air it's already burned, or not at all if the backpressure prevents it from getting fuel at all (more likely.)

If it's during the intake stroke, then you're sucking in both your properly metered fuel/air mixture from the intake valve, AND residual exhaust air from the exhaust valve, thus offsetting your carefully metered fuel/air mixture, and again making the engine very rough as it tries to burn already burnt air.

Either case is bad - and mixing them both will almost surely lead to a nonrunning engine. And even in compressed air or steam engines, you end up with significant performance loss as valves try to handle air in a way they weren't meant to.

The issue isn't necessarily with the valves themselves. in most cases they're mirror images of each other and couldn't care less. The issue is more with the systems attached to the other sides of those valves - intake systems aren't designed to handle exhaust, and exhaust systems aren't designed to be feeding exhaust back IN to the chamber.

Note, in most poppet valve combustion engines, the exhaust valve IS kept open past BDC on the fourth stroke. This is carefully designed to ensure that the exhaust has plenty of time to get out of the chamber (after combustion you have a net gain in pressure, this pressure will briefly prevent any fuel/air from coming into the chamber, so leaving the exhaust valve open lets the pressure balance itself). In these engines, the intake valve is often starting to open just before BDC on the fourth stroke as well. this intentional overlap ensures that the maximum amount of fuel/air can get into the chamber, for fuel efficiency. It is not necessary to have any overlap, and it's far easier to have none than a carefully designed forced overlap.

Hope this clears things up

- Ryan

 

[mike4517]

Ryan
You have a few things wrong in your last post.
Should be TDC not BDC.
And overlap is an advantage the exhaust system draws mixture through the open inlet valve.

 

[radfordc]

Intake/exhaust overlap is usual for high performance...

http://www.compcams.com/Pages/416/valve-timing-tutorial.aspx
Now something quite unique begins to take place. Just before the piston reaches the top, the intake valve begins to open and the exhaust valve is not yet fully closed. This doesn't sound right, does it? Let's try to figure out what is happening.

The exhaust stroke of the piston has pushed out just about all of the spent charge and as the piston approaches the top and the intake valve begins to open slowly, there begins a siphon or "scavenge" effect in the chamber. The rush of the gases out into the exhaust port will draw in the start of the intake charge. This is how the engine flushes out all of the used charge. Even some of the new gases escape into the exhaust. Once the piston passes through Top Dead Center and starts back down, the intake charge is being pulled in quickly so the exhaust valve must close at precisely the right point after the top to keep any burnt gas from reentering. This area around Top Dead Center with both valves open is referred to as "overlap". This is one of the most critical moments in the running cycle, and all points must be positioned correctly with the Top Dead Center of the piston. We'll look at this much more closely later.

 

[mike4517]

Radfordc

You are right.
Some of the charge does go into the exhaust but if the tuned lenth of the pipe is right you will get a return wave that pushes it back in, this is at certain revs.
I like to set my cams up equal overlap 5 deg before TDC.

 

[radfordc]

You can see it here http://modelenginenews.org/gallery/images/pascoe_v3.jpg
Valve driven by shaft at 1/4 crank speed.
This means ports can be smaller and helps with sealing
He has used chain drive as well.

That is exactly what I was envisioning. I knew there had to be one somewhere.

 

[Rayanth]

Ryan
You have a few things wrong in your last post.
Should be TDC not BDC.
And overlap is an advantage the exhaust system draws mixture through the open inlet valve.

Mike,

Thanks for the correction.

I did try to pay homage to the advantages of intentional, and carefully calculated overlap but should have spent more time there perhaps.

While there is a distinct advantage to overlap, as Redford explained in his followup, there's a bit of a trade-off involved. For every day use, the advantages of an overlap are minuscule enough that it doesn't really make much of aa difference. Properly done, performance is boosted on the scale of about 5 to 10 %, and fuel economy is improved by a bit more than that, as each 'bang' is put to its full potential.

However for a non-performance engine designed by a nonphysicist, the potential for getting it wrong is potentially fatal, as in my worst case scenarios mentioned previously. A difference of just a few degrees of overlap, or an open-stack exhaust system used on an engine designed for a tuned exhaust...that few degrees could go from a 10% net boost in performance, to a loss of easily 25% or more. A few more degrees, and the engine may not run at all, especially if intake is opening too early and the residual pressure from combustion stalls the intake stream.

I did not intend to scare anyone away from any design which has valve overlap...I was trying to explain why it might be a bad thing. My personal recommendation, especially if you're unfamiliar with the mathematics and physics behind combustion cycle engines and designing something from the ground up, is to have zero to minimal valve overlap at first, and gradually add a little overlap at a time, running the engine to see how it reacts.

Ryan

 

[mu38&Bg#]

Model airplane engines have had significant overlap since their inception. 40-80° overlap period is common. As far as tuned exhaust, just how long of an exhaust are you putting on your model engine?

Anyhow, as we all know engine design is a compromise of many concerns. If performance is not an issue, anything works. One doesn't need to be familiar with design specific of engines to build a model engine, even one that performs fairly well, copying existing designs works just fine. And, existing designs have a lot of overlap.

 

[Rayanth]

In retrospect (what is it they say about hindsight?) I realize I neglected to take scale into account. I have been buried in the nitpicky details of the largest piston powered aircraft engine ever to go into production, and things work much differently when you are running a 5.75 inch bore and 6 inch stroke on 28 cylinders =)

As you scale down a cylinder, the combustion process doesn't necessarily scale with it. For the same fuel/air mix, the flamefront still travels the same speed regardless of size, so in a smaller cylinder it will burn up all the fuel faster. This leads to higher Rom's through a couple other linked processes but I digress...

With the faster combustion stroke comes a MUCH higher tolerance for valve overlap, and in fact it does almost become required, as the strokes are so short and fast you simply don't have time to clear all your exhaust and intake all your fuel/air without overlap (another thing that doesn't scale well is air movement)

So to go back on my previous responses- in small scale engines, by all means overlap those valves! And I wouldn't be a very good model engineer if I didn't recommend experimenting (on that note, do they make scale dynamometers? How would you be able to tell if one valve design over another is any better?)

Ryan

 

[radfordc]

I worked up some draft drawings for the head and valve. The head will be 1.5" square and the rotary valve will be 3/8" diameter. The cylinder bore will be 1".

 

[radfordc]

Here is the engine design so far. I found a set of drawings for a .60 cu. in. 2 stroke engine and revised them to use a 4 stroke rotary valve head.

View attachment 4 stroke rotary valve engine.pdf

 

[kf2qd]

2 problems - The size of your rotary valve will severely limit the amount of air you can move into that cylinder. I think you will need a bigger valve. You also need to work on your squish area as it will affect ignition and timing.

 

[radfordc]

The intake and exhaust both have a 3/16" diameter orifice at the smallest point. If it turns out to be a problem it will be reasonably easy to make another valve. The ignition and timing will be done with an electronic ignition with a CM-6 sparkplug.

Upon reflection, increasing the size to 1/4" is probably worthwhile.

Charlie

 

[Drei]

how can you seal the ports to avoid leaks, say if you had to apply the design on a small v8 block ( a one piece block).

Thanks
Andrei

 

[mu38&Bg#]

http://www.coatesengine.com/

 

[VVTi]

Hello

My english is not very good, but i want to help you. I know almost all rotary valve sytems.
One rotary valve is better than 2 valves. The diameter of the rotary valve should be relatively large.
A small diameter is too small time cross sections The biggest problem is the seal of the valve. A good material for the seal is crystalline graphite. Crystalline graphite is self-lubricating.


I think that's very interesting and "relatively simple" by MGN W12.


F1 engine MGN W12 with rotary valves. [ame=http://youtu.be/QWb70X0xELA]http://youtu.be/QWb70X0xELA[/ame]
Ilmor F1 rotary valve project http://home.people.net.au/~mrbdesign/PDF/AutoTechBRV.pdf
Patent US 6666 458 B2 rotary valve system http://www.freepatentsonline.com/6666458.pdf
THE DEVELOPMENT OF A VIABLE ROTARY VALVE ENGINE http://ralphwatson.scienceontheweb.net/rotary.html


Best regards

VVTi

 

[d-m]

Interesting topic and one I have been interested in for years. coats international (mentioned above by dieselpilot ) is producing a vary successful rotary valve head used on gen sets. if you dig into the web site there used to be info on early testing of the ideal and its development. As I recall it was a small block Chevy in a impala wagon was there first successful engine . Also at one time they offered heads for the SBC, the key to there success was in the material used in sealing the rotary valve and as i remember the material came from the space program. Its been a few years since I have read there web site, the ideal is old but with the new technology of today the concept is vary exciting and the research to develop it even further is way past do. I for one am going to revisit there site and see whats new, and I will be watching to see how this turns out in the model engineering world. Thm: Thm:

Dave

 

[mu38&Bg#]

I don't see any rotary valve making it in the auto industry. Variable valve timing, duration, and lift, have essentially obsoleted any rotary valve. In stationary single speed apps, like gensets, they could still have a benefit.

Charlie, is that 10cc or larger? The CM-6 plug is fairly large.

 

[picclock]

Just my 2 cents. The fit is fairly critical as you have a large pressure differential. It may be worth considering two rotary valves arranged twin overhead cam style as this would mean the fits would be less critical, easy size optimisation of the inlet and exhaust ports, and simple adjustment of the relative valve timing.

The downside is two extra bearings a timing pulleys (cog ?) and the extra work needed.

Please feel free to ignore.

Best regards

picclock

 

[mu38&Bg#]

That's also been done in the RC engine world. Not many were made. Cipolla 4T-TSV

 

[radfordc]

Charlie, is that 10cc or larger? The CM-6 plug is fairly large.

Yes, 10cc. Here is the head with the plug installed.