Exhaust valve forces

[tigtorch]

A few weeks ago I posted concerning an idea I have to build my first IC engine, somewhat along the lines of Henry Ford's first engine but with a modern twist: solenoid actuated valve(s) triggered by an Arduino, with Hall effect and/or optical sensors.

I am well on my way to doing this. I am using valves in the cylinder head itself (as opposed to in a valve body adjacent to the cylinder), the bore is 0.875" and I decided to make the valves as big as possible, primarily so as to be easier to make: they are 0.375" diameter and nearly (0.050") touching each other.

I have procurred 2 small solenoids ( https://www.sparkfun.com/products/11015 )that I have determined are too small. My question: how do I determine the actuation force needed for the exhaust valve? I can figure the valve and spring forces but how do I determine the internal gas force on the valve face at the moment that the valve is cracked open?

If it makes a difference in the calculations: in an effort to reduce friction (I want a very slow running hit and miss operation with a big flywheel) I am using no rings, just a close slip fit (this might be a mistake and is mostly because I know nothing about making and using rings). I did make the piston long (1.75") in order to reduce leakage.

 

[Till]

I think this way is a bad way, because you have no experience at all.
A good idea is to machine the engine exactly according to a succesfully established plan.
If you got it running perfektly (compression, carburetion, ignition, everything in perfekt order) then convert it to your own concept.

how do I determine the internal gas force on the valve face at the moment that the valve is cracked open?

Easy way: Use rather conservative timing and compression ratio, so you can assume a pressure of 3-5bar. Then apply to the valve surface to calculate the force needed.

If it makes a difference in the calculations: in an effort to reduce friction (I want a very slow running hit and miss operation with a big flywheel) I am using no rings, just a close slip fit (this might be a mistake and is mostly because I know nothing about making and using rings). I did make the piston long (1.75") in order to reduce leakage.

That's a bad idea. Choose a bore common with rc toys or chainsaws, then buy commericially available pistonrings. One possible error ruled out, cheap and easy.

 

[tigtorch]

Thank you for your response and of course you are right about following an established plan. I, however, derive most of the pleasure of this project from doing it myself, even if I fail. In the process of failing I learn infinitely more than building a previously successful design!

 

[Till]

Trial and error is the most ineffcient way to go,
as without any reference you'll sooner or later make false assumptions regarding complex connections. It turns out to run somehow on accident, not because you understood what the cause was or how it really works.
Countless hours wasted, small knowledge gained, no motivating success for a very long time (probably no success at all).

Break the problem down into smaller bits and pieces.
Simplify the problem by ruling out as many sources of errors as possible, solve one single task after another. Step by step.
Methodic, structured, thats the way it goes... quick, efficient and the perfect solid basis for successfully designing your own parts and components.

In the process of failing I learn infinitely more than building a previously successful design!

I don't think so. This is the best way to waste time, become frustrated und will likely end up with the conclusion, that certain things are too difficult or rather impossible to solve... leaving the project unfinished.

 

[stevehuckss396]

I think this way is a bad way, because you have no experience at all.
Choose a bore common with rc toys or chainsaws, then buy commericially available piston rings.

He did!

http://shop.ajreeves.com/cast-iron-piston-rings-78-x-332-1-6833-p.asp

 

[tigtorch]

Till, I apreciate you concern but my question (the amount of force necessary to open a 0.375 exhaust valve in a low compression IC engine) would not be answered by my using someone elses design unless there is an established design that uses solenoid actuation of valves instead of a camshaft, I don't believe there is.

I very much want to do something more or less unique. If I just wanted a running engine I would go buy one.

 

[MCRIPPPer]

pi x .1875^2 = surface aria in inches^2. amount of force is psi times the number of square inches of aria.

for example, 100 psi applied to 10sq inches would make 1000 lb of force.

 

[tigtorch]

What I need, and probably doesn't exist, is an estimate of combustion chamber pressure right at the commencement of exhaust in a low compression engine.

 

[Ripcrow]

I have thought about electronically timing an engine and my thoughts would be to cannibalise Jacobs brake activation solenoids.if you don't know a Jacobs brake is found on most American truck engines and operates by opening a oil path which expands a piston arrangement to open the exhaust at injection so they are strong and designed for high speed and frequency which I think most off the shelf solenoids could not handle.if your engine is pressure oil fed it should be easy to fit but even if it is not you may be able to use just the solenoids. You should be able to source second hand units from a truck wrecker because these hardly ever fil or perhaps the solenoids of fuel injectors but I am not sure of there setup

 

[tigtorch]

Thank you for your answer. It is the first one (other than the equation which is rather obvious) that actually addressed my question to some degree.

The reason I included the unconventional (and possibly unworkable) details of my engine wasn't to invite commentary on their probability of success, it was to provide all of the information necessary to answer my question regarding exhaust valve force.

I like your idea about the brake solenoids and understand your concern about speed of response. My goal on this engine is to make it as friction free as possible, have a very big fly wheel, operate it in "hit and miss" fashion and hopefully achieve very slow operation. If I succeed in this, the response of the solenoid is not as important as it would be in a conventional design. That said I realize I should not ignore the speed of response.

 

[canadianhorsepower]

Hi any good engine simulator would answer your question
here is a link to one of them

http://www.proracingsim.com/

 

[Lakc]

Easy way: Use rather conservative timing and compression ratio, so you can assume a pressure of 3-5bar. Then apply to the valve surface to calculate the force needed.

I think Till actually answered the question on the first reply, Figure 75psi x area of exhaust valve seat.
Even the professional engine builders have to estimate at this stage. Once its built you can take cylinder pressure readings to determine the actual value, which will heavily depend on volumetric efficiency and cam timing.

 

[mu38&Bg#]

You may do some studying here.

http://rbowes1.11net.com/dbowes/

Greg

 

[tigtorch]

Thank you all for the great information. Perhaps Till give me the correct info up front, if so I apologize to him. I was put off by his negative slant, there are far too many negative things around me and I do much to avoid them, sometimes at the expense of tact.

 

[old-and-broken]

A less powerful solenoid may work for you.

Most solenoids have a 'stroke' distance much longer than you would need move the valve. Using the lever and fulcrum principle, you could arrange a lever system that will sacrifice the stroke length of the solenoid and multiply the force available for a shorter valve movement. 3 to 1, or 4 to 1 should be easily attainable.

 

[tigtorch]

Old and Broken: Thanks for the suggestion, I had come to that conclusion myself. In fact I have already made two rockers and gotten two bigger pull type solenoids that I plan on mounting so that they pull on the top of the rocker and only travel about 0.050 to maximize the force.

I will post pictures of everything soon.

 

[Forestgnome]

I found a reference that indicates typical peak cylinder pressure at WOT is 2-5 MPa (see link). That would equate to 32-80 lbs of pressure on the valves you specified.
http://www.tech.plymouth.ac.uk/sme/desnotes/icepress.htm

 

[mu38&Bg#]

The maximum value won't matter, it's the pressure when the valve opens that will determine the force required.

Greg

 

[Forestgnome]

The maximum value won't matter, it's the pressure when the valve opens that will determine the force required.

Greg

Point is that if the valve is calculated to open at peak pressure you're guaranteed it can open 100% of the time. Otherwise it's a very complex calculation using only estimations of pressure anyway. Make sense?

 

[mu38&Bg#]

The valve would never open at peak pressure. Seriously, the solenoid would be 5-10 times oversize. Even a random guess from any old four stroke cylinder pressure plot would be close engouh and still give a large safety margin for the engine proposed. I'd assume 5 bar at EVO.

Greg