Visualizing Rotary Injection Timing

[hugemikeyd]

I want to better understand where in the 0-360 available injection timing for a rotary is visually:

The documentation states:

• On Rotary engines Injection Timing Position can be adjusted from 0 to 360 degrees.

• The units are degrees before top dead centre (BTDC) with reference to top dead centre between the compression and power stroke. This means 360 represents TDC between the exhaust and inlet stroke.

Using the diagram above, would that mean Compression (second stage) represents 0 BTDC and Intake (first stage) is 180 BTDC?

[Adamw]

Perhaps not a great diagram to understand the eccentric shaft position.  The red dot in the middle is showing the eccentric shaft lobe position so picture 1, 3 & 5 all show the eccentric shaft at what most refer to as TDC or "0" (eccentric lobe pointing to middle of sparkplugs).  Pic 2, 4 & 6 all show the shaft at 180BTDC or some might say "bottom dead centre".  I think referring to this as TDC exhaust/intake possibly adds confusion so I will avoid that (I have only seen this terminology used when referring to port timing rather than engine cycle).  

Hopefully my explanation doesnt add to the confusion...

The thing that makes it harder to explain is there are actually 3 combustion chambers in action all at the same time, and the eccentric shaft needs to rotate 3 times for the rotor to rotate once.  In your first picture above for example the eccentric shaft is at 0 or TDC (ie the eccentric lobe is pointing to spark plugs), but if you look at the 3 faces of the rotor, the one that is marked in red is doing its intake phase, the next rotor face clockwise on the rotor is at TDC and the next face around is doing exhaust.   If we only consider one rotor face, then the rotor needs to rotate 360deg for one complete cycle - intake - compression - power - exhaust.  And since the eccentric shaft turns at 3 x the speed of the rotor, there is actually 1080deg of "crank" rotation for a wankel cycle.  The ecu knows this and does work in a 1080deg cycle (important for limiters etc) but in terms of general setup discussion and timing etc it is common to refer to the cycle as if it occurs in 360deg.  

In regards to injector timing, say if it was set to start at "180deg BTDC", this would more accurately be 540BTDC.  Even though the eccentric shaft and a rotor face would be at TDC 180deg after this injection event, the chamber that we actually just squirted fuel into wont be at TDC until 540deg after this injection started.

As a general rule with a rotary you want to start injecting as early as possible in the intake stroke, but after the exhaust port closes (or not too much before). The exhaust port generally closes around 130BTDC.   

 

  

[Pete_89t2]

As another guy with a rotary (FD, 13B-REW with stock ports), I've been trying to understand how Injection Timing relates to the rotary combustion cycle. FWIW, I've been using the default 130* BTDC Injection Timing figure that is provided in Link's base maps for the S6 & S7/S8 RX7s in mine, and it seems to work pretty good. I understand Adam's point that you want injection to start as early as possible on the intake stroke, but after the exhaust port closes. But where I get completely confused is reconciling Adam's statement "the exhaust port generally closes around 130BTDC" with what I'm finding in the FD's factory shop manual, which specifies the exhaust & intake port opening/closing timing metrics as follows:

Intake Ports: Primary Opens @ 45*BTDC; Secondary Opens @ 32*BTDC. Primary Closes @ 50*ABDC; Secondary Closes @ 50* ABDC

Exhaust Port: Opens @ 75* BBDC, Closes @ 48* ATDC

I'm assuming Mazda means "BBDC" = before bottom dead center; "ABDC" = after bottom dead center; BTDC = before top dead center and ATDC = after top dead center.

The math & understanding here would probably be simplified if we had a consistent definition and usage of these metrics relative to the rotary combustion cycle.

[Adamw]

On 7/9/2023 at 3:18 AM, Pete_89t2 said:

The math & understanding here would probably be simplified if we had a consistent definition and usage of these metrics relative to the rotary combustion cycle.

On 7/8/2023 at 12:08 PM, Adamw said:

I think referring to this as TDC exhaust/intake possibly adds confusion so I will avoid that (I have only seen this terminology used when referring to port timing rather than engine cycle).  

God knows why, but for whatever reason, when it comes to port timing Mazda no longer references the common "TDC compression" crank position that you would think makes the most sense.  Since tuners are used to referencing everything from TDC compression and you have a convenient mark on the front pulley referencing TDC compression, then from a tuning and engine control perspective it makes sense for the ignition timing, injector timing, knock window, etc to all reference this common point. 

I generally dont like to add the confusion of port timing since its not really relevant to tuning, but since I seem to have already confused you I will try a quick explanation.  I might do a better drawing showing injector timing etc one day.   When it comes to port timing, rather than use this same "TDC compression" fixed position that we all know, Mazda instead reference the combustion chamber volume that is relevant to the port or phase.  To try to make that clearer, try to think of when a piston engine is at TDC it has the smallest combustion chamber volume, and when the piston is at BDC it has the largest volume.  For a wankel there are 2 opposite points where "smallest chamber volume" (or TDC) occurs and 2 opposite points where largest volume occurs.  So, Mazda use 2 different "TDC's" and 2 different "BDC's" for port timing.  

In reference to the red letters on the 6 pics below:

C is "TDC compression". The chamber under compression (9) is at its smallest volume.  This is what TDC or 0 is in Link ECU.  

F is "TDC Intake/Exhaust".  The chamber that is open to intake and exhaust is at its smallest volume.  Intake port opening and exhaust port closing timing refer to this point.

B is "BDC exhaust".  The chamber in exhaust phase (14) is at its max volume (eccentric lobe pointing up).  Exhaust port opening timing uses this point as zero.  

D is "BDC intake".  The chamber that is in intake phase (4) is at its max volume (eccentric lobe pointing down).  Intake port closing uses this point as zero. 

 

 

[Pete_89t2]

Thanks Adam for the clarification & diagrams, that was very helpful in helping me visualize how Mazda's oddball port timing references relate to the typical TDC compression cycle reference ECU's typically use for tuning. Now I see why 130*BTDC injector timing makes sense for a stock ported 13B-REW.

I've got one of those illustrated GIF animations of a rotary engine on my computer somewhere. If I get ambitious I'm going to see if I can edit it to overlay the above info into it to try to illustrate this for a complete 4 stroke cycle of the 3 rotor faces (i.e., 3 revs of the E-shaft, 1 full rev of the rotor).

[hugemikeyd]

Awesome, I am so glad that I asked. I was confusing F in @Adamws diagram above as the TDC for the Injection timing based on the explanation in the manual: 

• The units are degrees before top dead centre (BTDC) with reference to top dead centre between the compression and power stroke. This means 360 represents TDC between the exhaust and inlet stroke.

But actually, it's C, got it. 

Full disclosure, I was trying to take a dynamic End of Injection table from a competitor and make it work, but perhaps I should stick with the static 130 from the FD base maps...

[hugemikeyd]

Bringing this back from the dead. Rob Dahm made a video where he goes in depth about various topics and one of those is Injector Timing. Here is the point in the video where it starts: 

"You want fuel going in at that 400ish BTDC." 

The above is confusing to me since Link only lets us adjust from 0-360, so is he really talking about 180 here? I assume he is speaking in terms of the degrees of rotation for the e-shaft whereas the Link is speaking in terms of the rotor?

[Adamw]

There is a "TDC" every 360degs, so 400BTDC is effectively the same as 40BTDC. 

[hugemikeyd]

And 400/3 = 133, so he must be speaking in e-shaft rotation degrees. He also mentions something along the lines of "you have 270 degrees to inject fuel", 270/3 = 90 and 90 + 40 = 130!

I think i finally get this about a year later  

[Adamw]

No, I dont think you have got it yet - but I dont blame you when there are popular videos like this floating around.  I dont claim to be a rotary expert myself, but there is so much wrong in that video that I can only roll my eyes.

The Link uses E-shaft degrees and it appears he is also talking in E-shaft degrees.  But why the hell he is saying at 28:10 (same as "position F" in the schematic above) is 460degs is beyond me.  If the E-shaft lobe is pointing to 9o'clock then this is 540BTDC.  460BTDC would have the lobe pointing to about 11:40o'clock.   

[Adamw]

I've just drawn a possibly easier-to-understand visual diagram.  Attached as a PDF as I dont think it would be clear if embedded as an image.

I have plotted a full 1080° cycle for one rotor, showing approx port and combustion cycle events based on stock quoted FD port timing and an example of where the injection events would occur if started at our default of 130BTDC and a duty cycle of 70%.  The top row marks out Link injector timing "BTDC" values.  The "spark" I have just placed at TDC and dont show the trailing spark just to keep it simple.  

The orange region marked overlap means both the intake and exhaust ports are open in that region.  With ported engines the exhaust would close later (and intake opens earlier) so the orange overlap region would get wider. Injection timing would likely need to be a little later (smaller BTDC value).      

Rotary Port and Inj Timing.pdf

[R_Prowess]

Am I understanding this correctly, by default the Link ECU is setup for start of injection at 130BTDC? and not end of injection?

[Adamw]

Correct.  Two reasons:

1. The main scenario you want to avoid is injecting during the overlap period, so using SOI set after that angle will prevent that possibility.
2. Secondary PW is calculated based on the primary calculation, I cant remember the exact sequence of events required to cause the scenario, but if using end of injection timing with staged injection with vastly different PW on pri and sec, it is possible to have a secondary inject into a different cycle than the primary.  

[R_Prowess]

Good, this makes perfect sense and I follow the logic.

I was just researching Link as another rotary friend turned me onto them. While doing so I came across this thread which reminded me of a similar one on RX7club.

I see little discussion on this topic specifically regarding rotaries and what I have found seems to be primarily from Haltech.

I had my own Eureka moment months ago and to my surprise it aligns with Link's train of thought, so it's nice to know injecting right after exhaust port is closed works well in practice. The only way I could think to do this in a repeatable manner is 'start of injection' which you just confirmed is the case.  

My MS3pro defaults to end of injection but can be changed. Both MS3 and Haltech go by E-shaft angle which I don't believe can be changed. Either way, looking at your PDF it appears 130 degrees BTDC lines up with 230 degrees E-shaft angle. 

This evening I went out, switched the MS3 to start of injection or in Megasquirt's terms "beginning of squirt" and changed the table to reflect the E-shaft angle of exhaust port closing.  It started up well, but unfortunately I am unable to drive it and do some road tuning until finished with the 8.8 IRS swap.

My only concern was whether or not there is enough time with SOI to fit the injection in at higher loads before the port closes, but there seems to be an ample window of time and it has obviously been working for Link all along.




For anyone who wants a look at that similar discussion. 
https://www.rx7club.com/haltech-forum-62/injector-firing-angle-continued-1101069/

[Adamw]

1 hour ago, R_Prowess said:

Both MS3 and Haltech go by E-shaft angle which I don't believe can be changed.

We are all talking in E-shaft angle.  And both MS3 and Haltech injector timing tables represent degrees BTDC.  The mark you have placed on the diagram in your post is actually 230 AFTER TDC.  

It is easier to look at the TDC at the far right-hand end of the diagram and count backwards just like the "Link" angles I have placed in the top row.  The only difference between Haltech and Link is we roll back to zero at 360 whereas Haltech allows values larger than 360.

So a few marks added to the diagram below to clarify.  The start of injection for rotor face 1 would be 490BTDC using Haltech's convention.  This is 130BTDC using Link's convention (ie just add 360 to get the Haltech value).  Note however if you set Haltech to 130BTDC you would get exactly the same result as 490BTDC since the injector fires at the same point every 360deg anyway.  Your MS3 would be the same, the value will be either 130 or 490 depending whether it allows values larger than 360 or not.  

  

[R_Prowess]

Well I sure got it twisted, thanks for clearing that up, hopefully this helps others to understand. 

[Copyninja]

I have read this and a lot of stuff on rotary injection timing and still struggling to really pick this up. It doesnt make any sense to me still...so i've been using the soi 130deg setting.  

Adam - fairplay mate, you gone above and beyond to explain this but i dont get it at all. Is there a way you can dumb this down even more? lol

Let say i'm running the largest street port possible and there is greater overlap than a stock port. Would the 130deg setting still be optimal? or would 140-155 be more optimal window above 5000 rpm to redline?

The reason i ask is because 90deg below 3000rpm actually fees subjectively better than running the oem 130deg on the large streetport. Am i doing something wrong here?

The haltech ecu tuners claim being unable to fire the primary and secondary separately on link is a limitation. Does this make any difference?

[Adamw]

With a ported engine the orange "overlap" region will get wider - the exhaust port will close later so you will have to start injecting later.  This means your start of injection will likely need to move to the right (smaller BTDC numbers).  So yes, I would think 90ish would be the ballpark I would expect. 

 

12 hours ago, Copyninja said:

The haltech ecu tuners claim being unable to fire the primary and secondary separately on link is a limitation. Does this make any difference?

With a typical rotary application I would say no.  Your primary and secondary injectors are usually located a similar distance away from the combustion chamber and the same exhaust port closing limitation applies to both.  If we were talking about a high RPM NA P-Port engine with one set of injectors squirting straight into the chamber and the other set mounted outboard of the throttles for example, then I think the ideal timing for pri and sec would be quite different.  In this case however it is never going to idle or drive well at low RPM anyhow - no matter what you done with the primary injector timing, so you would likely spend your time optimising your injector timing for the secondaries and not care about the primary. 

[Copyninja]

Thank you mate. That is really helpful.

In a large streetport above 5krpm to 9krpm does 150 to 155 seem a reasonable setting in your view over the default 130 setting? Can you confirm that the exhaust port will deffo still be closed or propose a more ideal setting if possible please?

90deg feels good to around 3000rpm. Thanks for confirming this is in the ball park you expect.

130deg again feels good to around 5000rpm? Can i please get your thoughts on this?

Looking for some advice above 5krpm as really dont want to inject when exhaust port is open on the front and rear rotor.