BlackWorksInc

Hey guys, I'm taking a CarQuest Techincal Institute course right now on Pressure Waveform Analysis. I'm not sure if any of you guys in the field are using this stuff to diagnose cars, but its really killer stuff. It can let you even pin-point carbon on a valve without removing the engine.

Which got me automatically thinking; "How do I apply this for a rotary?"
Which is the purpose of this thread, right now I need to crunch numbers, and work out soem theories, but I lack sufficent data. I really only trust you guys since I never leave this section of the forum really.

But off the top of my head I've come up with:

-3 firing events occuring every 120 degrees of rotation + another 3 occuring 20 degrees after primary firing event.
-Resulting in 6 actual spark events in one 360 degree revolution
-Unsure how much of the exhaust or intake stroke you could tell
-Need to calculate how to tell "TDC Compression & BDC Exhaust strokes"

I know alot of you guys have this stuff for breakfest so I'd love yoru input, I am trying to make a comprehensive theoretical chart for the events on the Rotary Engine so that I can then test it on a live engine and write-up documentation on it. I think that this type of diagnostic tool would be very helpful to us.

Thanks guys,

Paul

j9fd3s

iTrader: (3)

in a rotary the rotors are geared 3:1 vs the eshaft. with 3 chambers per rotor this means there is one firing event/power stroke per rotor every 360 degrees.

so on a 2 rotor, since the rotors are opposite (180 degrees) from each other, there is a firing event/power stroke every 180 degrees, just like a 4 cylinder piston engine.

on a 3 rotor the rotors are 120 degrees apart, like a 6 cylinder piston.

4 rotor is 90 like a v8...

a 4 stroke piston engine takes 720 degrees to fire all of its cylinders. the rotary takes 3 to fire ALL the chambers.

or even another way, the piston engine has 180 degrees of crank rotation for each stroke, the rotary has 270.

BlackWorksInc

What I was going by my numbers was treating each chamber individually as only one "Cylinder."

So for a piston engine: 720 degrees of revolution = 4 strokes, 2 TDC, 1 firing event.

So in theory for a rotary: 360* of rotation = 3 TDC events, 6 ignition events (Counting leading and trailing)

I'm trying to get a make-up of the single combustion chamber at this point before expanding into the rest of the engine, I hope I am explaining my idea right...

t_g_farrell

iTrader: (1)

I would count the leading and trailing as a single ignition event since its still for the same rotor face combustion event.

DivinDriver

iTrader: (15)

Interesting; how exactly is the "pressure waveform" detected? Off the exhaust pulses? Driveshaft?

On a 2-rotor, you get one combustion event, normally originating with two flame fronts 20 deg apart*, for every 180 degrees of shaft rotation, alternating between front and rear rotors.

You get replication of combustion events from the same specific rotor face & seals ("working chamber") every 3 rotations.

*On some year models, the trailing ignition is automatically shut down under certain conditions to reduce emissions. And the 20-degree separation is nominal; since both ignitions are controlled by mutual mechanical advance but independent vacuum advances, the separation angle varies.

j9fd3s

iTrader: (3)

the rotor spins once, and the eshaft spins 3 times. so 360 degrees of shaft rotation gets you ONE firing event, per rotor.

the leading and trailing both contribute to the same flame front/combustion event, so even though there are 2 sparks, it may not show on the scope.

BlackWorksInc

You use a pressure transducer to test the engine like you would a compression gauge, but the digital read-out on your scope allows you much greater detail and resolution of what's going on in the cylinder. You can actually watch the piston go TDC, valves open and close, and everything on a waveform that you can analyze. Its pretty awesome stuff.

KansasCityREPU

iTrader: (9)

The person that makes the cool compression tester and was selling them here on the board has a digital readout. Their is an AD to D converter inside the unit to get the digital reading. The transducers and data aquisition unit has much more data then what is displayed. I'm not sure if it would need a more sensitive transducer or not.

BlackWorksInc

The pressure transducer normally used is the 0-300psi and/or 0-30in/hg tranducer. It provides a digital readout that you can apply to a Digital Ociliscope, hmm let me see if I can explain it a bit here... remember doing ignition analysis on secondary and primary coils and examining the waveform of the ignition system to diagnose problems? This is the same ideea, the tranducers give you an accurate readout of pressure changes down to miliseconds that you can use to diagnose issues inside the engine. The purpose of the timing (in terms of Degrees) is so that you map out your waveform to show where TDC occurs, when the valves open, where the piston is at what time, how the engine breathes, ect. Its a cool diagnostic tool and I want to see if I can bring the benifits of it from the boinger end to our end. I think it would be awesome to diagnose an engine without ever having to tear it apart and know exactly how efficently it is running.

DivinDriver

iTrader: (15)

You basically need a phase diagram, then.

There's a tech design manual on Sgt. Fox's site that may have exactly what you need in it...

Right here:

http://foxed.ca/rx7manual/manuals/RE...amoto-1981.pdf

I seem to remember that it has much to say about pressure propagation and phasics in the rotary. Written by the chief design engineer, I believe.

I bet using a vector rather than linear scope display would be revealing, too.

BlackWorksInc

Oh okay I see, I'm basing my messurements off the rotor rotation rather than e-shaft rotation (which now that I look at it I should have done from the beginning...). Funny how a headache makes you think clearly... in any case, I was wondering where do we consider TDC and BDC? Is it counted for each rotor face or when the face reaches the ignition event? I am at work right now so I can't really look over the link you posted Divin. I remember I have that book saved somewhere on my lappytop.

So now the math is starting to make some sense, but I still have yet to hook an engine up to a scope to check it out. Unfortunetly I probablty will not be able to do this for a couple of weeks since my car still is not running and I'd have to catch one of the other rotorheads at the college to lend me their cars.

By the way, DivinDiver, what did you mean by vector scope? We're using a Vetronix MTS 5100 at the college, is there a way to set that scope up for that or is it a seperate piece of technology?

DivinDriver

iTrader: (15)

TDC in the rotary is considered to be when a rotor face (any of the three) on #1 rotor (front) is parallel to the plane of the spark plug holes. It really should be called "Left Dead Center."

Looked at another way, it's when the apex seal between any two faces on the front rotor is exactly at the crest of the bulge in the rotor housing on the intake/exhaust side.

The e-shaft is in the same position every time any face of the #1 rotor is in that postion.

#2 rotor is 180 degrees out of phase from that, so at engine TDC, an apex seal on #2 rotor will be exactly between the two plug holes.

That's why you can use a small wire or a small-bore depth guage to verify TDC by probing gently thru the plug holes; so long as you can recognise when the apex seal is between the plug holes, TDC is when the wire will extend equal distance into the combustion chambers.

Superior? No, I was just think that an old piece of TV diagnostic technology would be fun to use in this case.

A vector scope is an o-scope that is set up to display phase relationships. It reads out in a circular display of degrees and shows the relationship of intensity & frequency between two waveforms. In the analog TV signal, color info is actually sent as phase data, referenced to the 3.54MHz color burst waveform - - so a vector scope was used to examine color signal quality.




I was just thinking that, if you could synchronize a vector scope to shaft rotation, and then display the pressure waveform as radius from center at specific degrees of shaft rotation, you'd have a clear picture of cyclic pressures, and variations from rotor face to rotor face would show up as instability in the display.