Ok....here goes. 12a's pp usualy make hp about 500rpms higher than a 13b would of the same porting. You are correct mazdaspeed, Most pp engines make peak power around 8,500 to 8,800. However depending apon port shape, you can move the power band up by quite a bit. Square ports are usually for "sprints" and the D ports (bolth normal and inveted) are better for low end. If thats possiable on a pp engine My old 12a pp engine was like a toggle switch. 7500 and it was all you could do to shift the thing before 11k. My EFI 13B pp started making realy good power at about 6,800. Part of this was the efi, and part because of porting. Bolth motors would make lots of usuable hp past there peak hp point. Just like a stock port engine/with mods. It was also important to spin the motors high enough to not fall out of the power band when you up shifted. Just like the 5.5k point in a 89+ Rx-7s. To be honest i've never dyno'd a 89+ Rx-7 in the trim i have now. However, you can feel the power start to take a turn down around 7,500. Its still making good power though. This March, i plan on taking it and have it dyno'd. I'll post the results then. CJ

 

Alright alright,
originally I had intended to start some discussion on counterproductivity of tying off the 6pI ports so they are open at all times. My point was that power is actually lost, not gained until the original designed opening rev of the port ie either 5k or 3800 or whatever. I am still looking into this. Now this has progressed into a shift point melee. Which I don't mind, as it's still the sharing of info... but not my initial intention. I shall bow to the masses though and play along... Attached is the dyno graph by Mazdatrix I scanned, as well as some of their findings in text re: the enlarging of the aux ports (note that they state lower power band losses and torque losses) which cross references my claim that wiring them will lead to a decrease.

Strike that, I just tried to scan Mazdatrix '97 catalogue pg R8 but the graph comes out super blurry and too small. I am horrid with computers, and don't know how to fix it (sorry); so you'll just have to trust me with these no's. I cannot find the graph on their website, but if you have this catalogue, please look it up.

Hp is at the rear wheels (stock car) [I am assuming this is pre '89 version]
2000 rpm 45hp
6000 rpm 96hp < this is plotted as the peak power output
7500 rpm 82hp < I think that drop is significant enough to reassess the notion that churning it to the buzzer is always the way

Hp " " " " " (modded with header, dual exhaust, racing only unit)
2000 rpm 45hp
6000 rpm 112hp
7000 rpm 123hp < this is plotted as the peak power output
7500 rpm 119hp

The 6-port non-turbo engines (made from '84-'92) are a totally different story for porting. The front and rear side housings have the auxiliary ports in them, and the center housing has very small ports. Additionally, there is not a lot of casting thickness behind the upper part of the center port. This means if you go too far, you hit the water jacket, and the housing is now pretty much junk. At a later time, we will cover the method of adding epoxy to the back side, and filling in the EGR ports (86-88) to allow more porting for race engines (not recommended for street engines where longer term reliability is needed).

If the separate ports (in the end housings) are simply made into one large port, the bottom and mid-range torque suffer a lot, and the port velocity is compromised as the intake flows from the manifold into the intake runner. The intake port timing of this configuration is: later on opening than a peripheral port, but roughly equal to intake closing on a peripheral port! That's a LOT of timing and port volume for a street car. There are MANY combos that could be tried here. We have not been (nor have we heard of anyone) willing to build/port probably 8-10 different engines/port combos, and spend a few hundred hours on the dyno, to optimize these engines. We are talking big $$$ to do all that - in order to sell porting templates ???

Anyway - this page is written assuming the 6-ports are going to remain active (rotating sleeve installed). We tend to tell phone customers that 6-port engines are "not able to be ported" -- we frequently do not have the time (sorry - this is a confession) to go through the VERY long speech explaining how "that's not really true, they CAN be ported, gain is not very much, might hit water, epoxy behind the center ports, full explanation of the rotating valve, exhaust back pressure activation, port timing and volume problems with connecting the ports, intake runner cross section limitations, etc. etc. etc.

 

Ok. this is how it works. In order to determine the precise time to shift, you need to know your wheel TORQUE numbers, and your tranny gear ratios. You shift when your torque * current gear becomes LOWER than torque (at new, lower rpm) * next gear ratio.

How about an example.. i'm going to make up random numbers anyway..



RPM... E TQ... RWTQ 1... RWTQ 2... NEW RPM

3000..... 85..... 255......... 170.......... 2000
3500..... 95..... 285......... 190.......... 2333
4000..... 100..... 300......... 200.......... 2667
4500..... 105..... 315......... 210.......... 3000
5000..... 103..... 309......... 206.......... 3333
5500..... 100..... 300......... 200.......... 3667
6000..... 90..... 270......... 180.......... 4000
6500..... 80..... 240......... 160.......... 4333
7000..... 70..... 210......... 140.......... 4667
7500..... 60..... 180......... 120.......... 5000


Okay. From column 2, You can see that the torque PEAK is at 4500 (105 lb*ft), but that doesn't matter. Let's pretend that your in 1st gear, with a ratio of 3:1. You want to know when to shift into 2nd gear (2:1)

Third column is the torque applied to the diff in 1st gear. 4th column is the torque applied to the diff in 2nd gear. You *ALWAYS* want to maximize the torque applied. The last column is the rpm that you would be at, if you shifted into second.

So...hm..if you shift at 4500 in 1st, while putting 315 lb*ft down, you drop to 3000 rpm and put down 170 lb*ft. Not the best idea.

If you shift at 7500 1n 1st, putting down 180 lb*ft, you drop to 5000 rpm, and put down 206 lb*ft. Since there is more torque in 2nd, you should have already shifted by now.

If you have a dyno graph, you can calculate the exact shift point for every gear. Since the gear ratios are not always reduced by the same factor, there is a different shift point in every gear.

For the numbers above, around 7000 looks like the best place to shift.. You go from 210 lb*ft (which is DROPPING) to ~4500rpm: 210 lb*ft (which is RISING).

-Tesla

edit: ugh.. table turned out not looking good...hmm..

 

use tabs instead of periods.

 

Mazda (and about all other manufacturers) invented variable port timing (for us- 6-ports, 89 intake, etc)to spread the usable power/torque over a wider range.

In the race engines we do, the torque usually is peaked at the lowest rpm we are concerned about (meaning actually somewhere BELOW there). A "Pro7" 12A has a torque peak at 5000? (where we START the dyno run), and an HP peak usually about 7000. The E/Production 6-port we are testing now has a torque peak at 6000 (where we START the dyno run!!) and we (so far) can not find the power peak!! By this I mean my dyno tach only goes to 10,000 and the engine is still GAINING HP at that rpm!!

This all means that, in the Rotary (racing) world, we pretty much ignore the torque peak (too low to use), and work only with the HP/rpm curve.

The whole key (in semi-tech/math terms) is to maximize the AREA under the HP/RPM graph for the rpm spread you need for your specific RPM splits/drop between shifts. This gives you the most available/usable HORSEPOWER for the rpm range you are using.

A "Pro7" engine (for example):
RPM Torque HP
5.0k 127 121
5.5k 125 131
6.0k 121 138
7.0k 111 148
8.0k 92 140
8.5k 84 136

With those numbers, and needing a little more than 2700 rpm split between shifts, you shift about 8500 (136hp) and the next gear comes in about 135hp. Because the rules for that class require steel apex seals, we use about 8500 as the upper limit for hoping the engine stays together.

We are not sure what we are going to be doing with the E/P 13B 6-port. If we hope the HP PEAK is (only?) 10,000, and an rpm range of about 2000 (close ratio box) then we would need to shift somewhere about 10,600 ??? and we are not at all confident the engine will stay together at that rpm.
This last writing is basically in answer to the post quoting all my writing (done YEARS ago) on testing 6-port engines. Since we now have a reason to "waste?" all the time and $$ playing with it on the dyno, it is nice to know I was pretty much on the mark all those years ago. -- meaning the intake port timing is SO late, and the port volume so HUGE, the max HP is WAY up there in rpm.

And no,we do not "ignore" the torque curve/peak. By playing with all the variables, we are always trying to raise the peak as high as possible, and "flatten" the torque curve as much as possible.

Dave ---

 

BRAVO

 

Its ok, you meant well. Nobody has argued that fact. We all know this already.

 

So using a stock car, ( see graphs above ) what would be the best shift point for each gear? What is the formula again for hp/rpm= most power per gear? i see the one that uses the torque method, but Mazdatrix was saying to use hp figures. I'm asking if i hasd a dyno sheet how could i tell where by best shift point for each gear would be....

great info here

 

Optimum shift point is determined by 1) gear ratio differences in RPM and 2) the power curve. You want to use most effectively the power your car is making.

When your car would be at the same HP level in the next gear (if you upshifted), SHIFT! That means that you would shift just after the peak and end up in the next gear just before the peak. The optimum shift point varies by gear ratio because some of them are not spaced as wide across RPM's. The difference in RPM when you shift from second to third, is greater than the RPM difference shifting from 3rd to 4th.


FYI: I heard that the 7speed gearboxes in some F1 cars use as little as 50rpm between top gears on some tracks! No wonder they don't shift manually anymore.

 

Yikes, this topic has grown legs and tan away from me. I have decided to wash my hands of this. Anyway, everybody has great insight but is comparing apples and oranges here. Race prep, street applications, horsepower and torque etc. Aye. I am truly flattered that on only his sixth post, Dave from Mazdatrix has kindly found my thread worthy of his time. Obviously if there is any factual evidence on the forum, he would be the person to look to. Cheers to Dave

I will be posting some drawings I am going to scan in soon about some crazy internal mods that I dreamt up one night. I am wondering about direct injection and an altered recessed area each of the sides of the rotor. I am no engineer but having no preconceived ideas about what is possible and not may work to my advantage. I am very much looking foreward to your thoughts. Hopefully within the next week or so.

 

Mazdatrix, I have a question for you. Your site is quoted above, and I have read it there too, about grinding the 6 ports into one big port. You said this is just too much area, and port flow is severely compromised. I assume this was done with a stock manifold? IF you had a manifold custom made, and optimized for the port timing/area, would that porting not make impressive power? How would that compare to a PP, which DOES have an optimized runner cross section for the port timing/area?

 

Thats what i thought and have been doing, thxs for clearing that up

 

Been there, done that, DOING it right now ---
Big end ports, custom manifold (long runners, MANY hours to fabricate - didn't work), etc etc.
Literally just finished a run (1/2 hour ago) with SHORT primaries, BIG pipe, comp muffler (didn't work quite as well as LONG primaries, but not all that bad)--- there was no hp "curve" just a straight line -- starting at 6000rpm it simply just kept gaining power till I stopped the run at 10,000rpm!!

Oh well, I'll invent some more things to try in the next few days. We are trying to get it to peak about 9-9.5k with as much as possible about 7k.

Btw, this is all with a 52mm Weber IDA downdraft, with 38mm chokes (as E/P carb rules require - it is SO much simpler than the injection).

That last run actually HAD a torque peak -- 7k instead of the 6k we usually see where we start the "pulls". Nice find out about where the torque was.

Dave --

 

Off subject...but is there any update on the 3 rotor drag car? Or is it on the back burner? CJ

 

I've just been too busy to keep any "updates" current on the website.

Drag Car: Engine etc is in the car (fresh rebuild and broken in on the dyno- see movie at: http://www.mazdatrix.com/dragcar.htm )

12" supercharger on it. The adapter plate between the carb and the blower is at NOS having the spray bar installed - I should have it back next week. Then I have to re-plumb the system, finish some re-wiring, put all the bodywork back on, put the ignition back in (using it on the E/P dyno), get the chassis certified, go get a physical, go test at Palmdale to get my "license", and we should then be ready to show up somewhere.

(Meanwhile getting the Pro7 car ready for the year, building the E/P Convertible car, and developing the E/P engine on the dyno).
Piece of cake.

Dave ---

 

Do i get discounts for having your mazdatrix sticker on my 89GXL.

 

I just removed the sleeves, installed a custom ported TII intake manifold to fit the 6th port hole, intercooler, turbo, throtle cable, fuel rails, Blue 12-a turbo as secondary injectors and stock n/a injectors for the low end. Ported the exhaust and I am running on a stock gxl N/A cpu just fine!!!!.