while having issues with the hubs and timing i decided to put together my own test hub for the e-shaft to check base timing with. the thing to remember is that the eccentric shaft keyway is ALWAYS milled directly at the 9 o'clock position when the engine as at proper TDC. even eyeing the position is not even close to accurate as there are some hub and pulley combos that can be off as far as 15* or more..

one more thing to note is the timing pulley on the hub can rotate as much as 3 degrees or so just from the looseness of the bolts, tightening them down without knowing for sure could be off by a few degrees.

these issues are mainly something that might want to be checked when building a high horsepower 2nd generation turbo engine. the CAS gear wear can also allow the CAS to walk as much as 10 degrees of advance... if anything retard your timing to be safe if you plan on upping the boost and do not plan on going through all of these procedures.

i was a bit skeptical that all of these things combined would be an issue but looking at it first hand i now realize that it was something that should always be done for anyone expecting much more than 300 wheel horsepower where ignition timing is critical and can lead to failure for various reasons.

anyways, here i dug out my old Crane Cams timing wheel and fitted it to a stock hub after finding TDC of the engine shown later and translating the marks to the wheel, centering it and drilling the mount holes for the hub.


to properly find TDC you need to measure the gap between each of the pockets in the "tub" to rotor and match the gap with the position of the shaft. measurement should be about 4.85mm at both pockets rotor housing wall to face.


once you've found TDC you can build the degree wheel and position it to the hub, i indexed TDC, 5*ADC and 20*ADC as the 5 and 20* ADC marks will correlate to the stock timing marks on your pulley. lining up your TCS mark stamped from placing the engine at TDC and marking the hub after gently sliding it onto the shaft.


now when you place the pulley to the pointer at the right 5*ADC mark the rotor will rotate towards the exhaust port opening the lower gap and closing the top.


then at 20*ADC you will see it open up quite a bit giving you an idea if your marks are even close or not without even the degree wheel in place.


at any rate, the pulley i grabbed for this little experiment was a matched set, it was 3* retarded in timing so i elongated the holes to give it a proper TDC setting and correlate the 5 and 20 marks properly.

i will toss up some more pics soon of other hubs and hub combos to show how far off the timing really can be unknowingly.

as a side note, the pointer on the front covers are located at about the 6.5* advanced position so keep that in mind if you thought the pointer was at the 12 o'clock position. noted by the first picture which is set at TDC and you can see the location offset by the degree wheel.

timing walk of the rotor to gear to e-shaft is minimal at about .5 degrees.

 

all 3 of these hubs/pulleys came off of complete engines, unmolested as far as i know and assuming they are original. one thing seems consistent at least, that the timing is retarded on most original setups.

#1 not too great, not too bad. still nearly 10 degrees of timing lost.


#2 perfect!


#3 eh, not too far off but close enough..

 

wow, excellent post!

 

when playing musical pulleys with each of those 3 hubs i noticed one thing, it isn't the pulleys that are mismachined but the hubs themselves. all 3 pulleys netted the same results when switching them. when switching hubs the marks appeared as they did above every time, in some cases resulting in an advanced timing mark due to the slop of the bolt holes.

not saying this is the case for ALL timing pullies as those were all non turbo as far as i know. there is 2 different sets of hub and pulley combos and i have to dig up something to give more drastic results.

thanks, it has always bothered me that the hubs weren't trued. i figured eventually i would have to have devoted a day to figuring out a method of getting the true answer rather than using an assumed good hub and pulley.

 

hmm so it the HUB that's not accurate? interesting.

the PDF is from the JC cosmo new model highlights, i don't read japanese, but notice the CAS is +/- 8 degrees on the G signal and +/-1 on the NE signal?

 

i think i mistook that, one pulley is a bit more off than the other 2. it looks like it is the pulley that is mismachined. the slop in the bolt holes does make it a little bit difficult to tell.


i can't read japanese either but i would assume it is based off the tooth count. 24 G teeth to 2 NE teeth. i believe the NE is what is used for ignition timing anyways and doesn't account for the gear slop, rather just inaccuracy of the pickup sensors. even the FD pickups probably have the same variance.

 

actually you could be right about that... but you can see why the FD setup is better

 

no gears to give backlash to. hiccups like too much fuelling during tuning can cause the timing to jump forward. the gear is always under load so timing the engine the CAS is actually retarded but can walk to an advanced state if the engine stumbles for any reason, like transitioning injectors on a high compression turbo setup...

too many small things i think about that can cause issues but at least gives people ideas for what to expect. at least most EMS systems can adjust the injector staging point to turn on at lower boost levels to hopefully not cause a lean spike during transition in conjunction with an unwanted timing advance from the CAS.

the FFE trigger wheel kit is a good alternative for those who want to avoid the possibility of CAS walk ignition advance.

 

Archive!

 

It is commonly said that the racing beat unit is a superior unit. do you have one you can test this theory with?

 

great info. thanks

 

i do and i tossed it on to see. it's about 1 degree within spec.

importantly with the racing beat pulley there is no markings for 5* ATDC to set the FC timing properly so you would have to ballpark it between the left and 2nd timing marks. the trailing would line up outside of the left most mark.

the racing beat pulley timing marks are from left to right, 10*ATDC, TDC, 10*BTDC and 20*BTDC.

 

Nice work.

 

Interesting thread...I'm assuming this is all S4 correct?

 

Please Archive, this will be of great help when it comes to my rebuild.

 

applies to virtually any rotary engine.

the basis model was a series 4 turbo II engine but most later model mazda engines are timed with the 5*ATDC marking and the timing marks are in the same locations.

 

Karack,

Any chance you checked keyway position with the degree wheel once you got tdc on the rotor. Im wondering for assembled engines.

Thinking one could use the keyway, bottom of the front cover and a degree wheel to find true tdc(really just a reference to find true 5 and 20).

Wish I had done this while my engine was apart, course, I never thought Id have a different pully at that time.


Thoughts?

 

you can use the deck of the engine as a reference once you have the degree wheel centered at the 9'oclock key with TDC at 12o'clock(well not quite if you reference it off the 6.5 degree advanced timing mark).

but yes, the keyway will be at the 9 o'clock position based off the bottom deck of the engine.

for this tool i put it together to use on various combinations of engines so it is based off the pointer position on the front cover which isn't a true 12 o'clock position so the degree wheel won't match the deck for a proper measurement.

what you would want to do is make a pointer that bolts to the bottom of the front cover, set the engine to TDC via rotor measurements and set your degree wheel to the hub based on that position. i would dedicate a hub to the degree wheel so that it never loses accuracy as the bolt patterns on the hubs are slightly off one to the next as well as slop in the bolts playing a role in being a few degrees off depending how you drill the degree wheel holes.

keep in mind the various engines have different oil pan bolt patterns so this wouldn't work for some configurations unless you make various pointers for the bottom deck.

 

Great stuff man! Makes me want to make a mess in my garage tonight. Ive got a tote full of pullys in my attic.

Ive got two things to do before I start upping my boost, 1)make sure timing is dead on 2) fix the huge dent in my fuel supply line.

I want to try the Lynn E. Hanover method too:







Remove the plugs from the front rotor housing.

With a mirror and strong light turn the engine clockwise until an apex seal is centered in the trailing plug hole (the top one).

Mark the pulley under the wire pointer.

Turn the engine backwards until an apex seal appears centered in the leading plug hole. (the bottom hole).

Mark the pulley under the wire pointer.

Measure the distance between the marks.

Calculate 1/2 of that distance.

Measure the 1/2 distance from each of the first two marks toward each other.

If both measurements end up at the factory pulley mark, then that mark is TDC.

If the pulley is blank where your measurements predict the TDC mark should be, the factory mark is an unknown. Make your TDC mark on the pulley.

Measure the pulley diameter.

Find the circumference of the pulley. Calculate it. Don't try to measure it.

C=3.1416 X the diameter.

Divide the circumference into 360 to find the degrees per inch.

Use that data to lay out timing marks on the pulley.

Do it several times to be sure you understand what you are doing.

Use a fine 3 corner file to score the timing marks on the pulley engine side flange.

Fill the marks with different colored paint to indicate which mark it is.

You can now measure from the TDC mark to determine what that existing mark is. 5 degrees 10 degrees etc.


This will pretty accurate. But you should know that Mazda moved the plug holes up and down over the years to gain some advantage in polution control.

So, your TDC mark may turn out to be off of dead on accurate by a degree or so one direction or the other. This will have no effect on performance.

There is a very accurate method to measure for TDC if the engine is apart shown elswhere on this site.

Lynn E. Hanover

 

yep, noting the differences in the spark plug holes i opted away from using that as a method for checking TDC as both plug positions are both advanced in the housing. look at a bare rotor housing and you can see both plugs sit higher up in the housing and not very equally positioned in relation to TDC(the "hump").

i will check it a bit later to see how close it is but i have a feeling it's far from the 1 degree that the article mentions.

 

after checking the article method i noted that the trailing apex seal lines up 105 degrees left of TDC, the leading apex seal lines up 120 degrees right of TDC. giving a TDC which is 7.5 degrees offcenter from actual. this was with series 5 rotor housings with the narrower split. series 4 would likely be closer but still off by 5 degrees with a guess off the top of my head.

so i guess correctly i assumed there really is no way to properly time the engine outside of having it apart without knowing exactly the offset of the method used in the article. which i suppose i can set something up to get the timing offset for series 4 rotor housings in a bit.

granted most of the time the pulley marks are going to be on the safe side, but not always. using the method in the article with series 5 housings the TDC mark would be 7.5* advanced however. i believe i actually used that method on a recent engine which has been having issues holding the rear plate on a 9.4:1 turbo engine, which would make sense because once you get preignition with these engines in boosted applications they are overly touchy. they take much better to having timing retarded to even a hair slightly too far advanced.

 

Yeah, Lynn is a 12A guru, some things transfer well foreward, some dont.

Kicking myself for not doing this when the engine was apart. I had the original pully, transfered those marks to a gilmer, then to a RE setup. Now I question the original. Gawd rotaries are fun

 

well, i did check it against series 4 housings and the result was the opposite what i though. trailing lines up at the standard 105* left of TDC. leading lines up 135* right of TDC...

series 4 article method = 15* advanced TDC marking. bad
series 5 article method = 7.5* advanced TDC marking. still bad but not as bad


to correct myself the best method i can gather without all this fuss would be to get an accurate pulley, assuming the racing beat and others have good accuracy for their markings.

not to knock Lynn's method but unless the 12A rotor housings had symmetrical plug locations then the method does not work. i am fairly sure even 12As had advanced plug positions but i don't have a 12A housing to check against.

it's close enough for naturally aspirated engines, also considering most pulleys are retarded in their markings but for boosted applications i wouldn't advise anyone to use the method. or if so use the method and then move your markings retarded by the number of degrees i mentioned in this exact post.

 

a little possibility is that the series 4 and series 5 pulleys differ for the plug timing variance. looking at the first picture of the examples you can see the pulley is about 7.5* retarded which would make up for the difference between series 4 and 5 plug timing locations.

unfortunately all my pulleys are unknown and it will be a while before i tear apart our S5 GXL with original motor to compare if that theory is true.

mazda never changed the specs and continually called it 5ATDC and 20ATDC for all the 13B engines, wondering how true that statement is based off the true TDC of the engines and how they stamped the pulleys from the factory.

hopefully not just adding further confusion as this is even starting to make my head spin, without knowing the exact timing that the ECU uses, accurate pulleys offset for the plug timing variances and all combined. technically the spark plug is just in a different location so i am going to say ignore this single post as it shouldn't affect the timing of the engine but moreso just affects the results used in the post above this one, which are still accurate if you want to check your pulley marks.

 

That is one of the reason why you don't just tune every engine by X amount of advance..

Nice post.