MoonPopGun

Cant find anyone doing a TBI turbo setup on anything. If I can get this sorted out I will start by getting a RB manifold, and then by the end of next month getting the TBI kit.


This is the item I am looking at.

Go EFI 4 600 HP Power Adder ? FiTech Fuel Injection

What I ask Fitech:

"Is what I need to know is this.


1. All the Power Adder needs to run is impute from the exhaust, MAP sensor, and the ignition RPM.

- How are you/what are you hooking up to in ignition system to tell ECU what's going on.

2. I can time my engine as normal with a timing light, and this is all I need. The Power Adds ECU does not need to adjust the timing to run yes?

3. Running a turbo the ECU will be able to sort out boost?


SO I need to know if I bolt up this EFI body up will it work. If my timing is set, there is a O2 sensor, and the MAP can read boost... will it sort everything out?"

What Fitech said back:

"The 600 power adder will support up to 25 pounds of boost, does come with a wide band O2 sensor. You can program target air fuel ratio's at various load in RPM points including boost. If you have the timing control side covered all the FiTech would need is a four-cylinder RPM input, 6 or 8 will work to.

Could even use a crank trigger if need be. You did want to try some timing control, all you would need is a locked out distributor with a 2 wire magnetic pickup.

Any other questions let me know"

Ok so the kit has the Exhaust Sensor, MAP Sensor, and it can read Ignition RPM. All the imputes we need to run but....

How do we translate a rotary ignition signal to the ECU? I have a DLIDFIS set up on the car right now. I was told/digging around in "search" the 13b is more like a 2.6l four banger? That seems wrong. Would running the DLIDFIS be more like a V6 signal? Going off of one coil it be 3 fires per half eccentric shaft rotation, so six fires would be a full RPM like a V6?

Could I set the ECU to try, and run a 2.6/3 liter V6? In theory it should be able to sort its self out... yes?

wankel=awesome

Good luck.

Qingdao

iTrader: (3)

Why not run the factory S4 or S5 turbo system that Mazda created?

Seems like you're gonna end up spending all kinds of money for a system that was engineered with half the budget that Mazda put out.


That RB manifold won't work with a factory turbo you're gonna have to find some other turbo.


To answer your question. No, you're gonna have to figure out your timing separate from the FItech system. If you use the stock ignition you need to limit the travel of the distributor.


My suggestion for you if you're really set on an aftermarket fuel management system is to use a known system that works well with rotaries. IE power FC. I've heard its pretty good, all you would have to source is the intake manifolds and the ignition components.

At any rate good luck

MoonPopGun

The turbo will fit just fine. I have no problem doing some plumbing, or some spacers off the manifold.

The block I have is a old 70's block that will not handle EFI like off the FC's. I already have a locked distributor.

Qingdao

iTrader: (3)

Agh, locked? Free it up a little; your drivability will suffer with a locked distributer. I run about 8#s and I've got no problems running about 5 or 6*s of advance vs the factory 15 or 20.

Why not just run the "Jeff style" Nikki? Just follow the instructions on the how too thread. Pretty strait forward operation.


What turbo are you running?

MoonPopGun

I copied Jeff's style of distributer. Works like a champ as is. Cant remember what I locked it at though. Jeff has a lot of write ups, and has a lot of good builds. But what his threads showed me was I don't want to spend the next five years playing with carburetor jets, and all the other story's that are out there about carb turbos.

This is why I was wanting to go with the 600 Power Adder. It knows what's going out the exhaust, and the MAP sensor will take care of what's going in. On top of this the ECU can take care of boost via target air fuel ratio's at different loads in RPM's. I can time my engine as is, and the ECU dose not need to change anything about it to run. The ECU just needs to know what RPM its at. That's why I was thinking the DLIDFIS would maybe to the ECU look like a V6. (once more 3 fire's per half crankshaft revolution)

Qingdao

iTrader: (3)

Well I don't know what that ecu can control as far as ignition is concerned.

BUT if you semi locked your distributer like Jeff did you'll be A ok. I can attest to that. If I were you I wouldn't muck with the ignition at all. Just keep them as separate entities.


Dunno about 5 years. A month or two sure, but you're looking at that for any tuning. Now I will say that a fuel injection system might be dynamic enough to accept more changes than a static carbureted setup. For instance I recently went with a much larger intercooler. I had to change some stuff, but only took a day to swap out some jets. I think a fuel injection program would have been able to account for this change in intercooler and I never would have known the difference.

j9fd3s

iTrader: (3)

if it wants a tach signal give it a tach signal

MoonPopGun

I am sorry I miss said that. Yes its a semi locked distributer. I don't need the ECU to control anything with the ignition. They can be run as separate entities, but the ECU still needs to know what RPM the engine is running at to set target fuel ratios.

This is what I am wondering when setting up the ECU for a base starting out. What do I tell the ECU to think? As the FiTech Tech said I would need a four, 6 or 8 cylinder RPM input for it to work.

Jeff20B

iTrader: (13)

Five years? I don't spend more than five hours tuning a carb. Usually far less time than that.

MoonPopGun

I was going off of when you started doing these carbs good sir. And you are still adding info of experiences as you find them. Its getting near to 5 years over all?

Jeff20B

iTrader: (13)

I started in Feb 2014 and it's almost Feb 2017. So almost three years.

MoonPopGun

Tomato tomahto.

wankel=awesome

I just typed up like 2 pages of problems you're going to have, and that FITECH knows about, because I was to test this system and help develop it for the rotary, and they for the most part backed out.

It isn't going to work like you think. It actually doesn't work like anyone thinks. You input a startup map, (none for rotary at all, and I'm not writing them one anytime soon, seeing as they failed to ship the product) then you physically set "targets" along different simple parameters. It acquires target AFR'S by trial and error, starting with the base map.

It has a % of richness that is hard set for when the engine is cold based on that map. As the coolant temp increases, it leans it out. Simple.

Here's the biggie though: It adjusts idle mixture by 2 inputs. Vacuum reading, and tach output "speed".

So if you tell it that you want 900 rpm, it will adjust to that speed, but it will not smooth out. Here's why: Rotary vacuum is not smooth. Even on a stockport, UNLESS both rotors are pulling on the same plenum, and your sensor is in said plenum.

The fitech pulls vacuum from a single throat of the TB. This means you're only getting a strong vacuum signal from one runner, and it's corresponding rotor. Bit of an issue.

It fine tunes the entire map with the vacuum readings, so unless you either a) modify your intake manifold so that the runners cross over (performance decrease, unless it's a well designed plenum) or b) change the way it reads vacuum, which you're pretty screwed on with an RB intake manifold.

I knew that going in, and have a few ideas to fix it, but it's not my project lol.

Btw, if you actually plan to send them money, do it through summit racing so they can refund you without questions if for some reason you don't receive the part...

miasmicmonky

iTrader: (1)

@wankel=awesome, Maybe you should contact holley about their sniper or terminator kit since FiTech didn't go through with it. Holley is a larger company and may have more time and money to be able to provide you with a product as long as you can provide them with some fuel maps. If the sniper or FiTech worked for rotary, I probably would have already bought one!

wankel=awesome

I'm extremely invested in a couple different carb/turbo setups for now, or else I'd consider it. Maybe in the future, but the time I had to play with it has passed.

wankel=awesome

Btw @OP I want you to watch the second episode of "Engine Masters" on Youtube.

Note that the Holley terminator EFI WOULD NOT function normally, or be overwritten when their engine had a dual plane intake. It had to be switched to a single plane for the computer to function normally.

Your RB individual runner manifold would be 200% worse.

Just because parts physically bolt up to each other, doesn't necessarily mean they'll work.

MoonPopGun

[QUOTE=wankel=awesome;12143378]
So if you tell it that you want 900 rpm, it will adjust to that speed, but it will not smooth out. Here's why: Rotary vacuum is not smooth. Even on a stockport, UNLESS both rotors are pulling on the same plenum, and your sensor is in said plenum.

The fitech pulls vacuum from a single throat of the TB. This means you're only getting a strong vacuum signal from one runner, and it's corresponding rotor. Bit of an issue.

It fine tunes the entire map with the vacuum readings, so unless you either a) modify your intake manifold so that the runners cross over (performance decrease, unless it's a well designed plenum) or b) change the way it reads vacuum, which you're pretty screwed on with an RB intake manifold.[QUOTE]

[QUOTE=wankel=awesome;12143452]
Note that the Holley terminator EFI WOULD NOT function normally, or be overwritten when their engine had a dual plane intake. It had to be switched to a single plane for the computer to function normally.[QUOTE]


Thank you vary much for the book. Vary helpful! Odds are the Holley works the same way then, or the Fitech copied Holley's blueprint.

Now what about just using a single plane spacer, and then knife edge the RB manifolds plenum?

wankel=awesome

A single plane intake on the 12A makes very poor torque, and it would be especially flat because of how large that throttle body is.

A small TB+a well designed plenum would work very well, and I know this because I've done it.

You could try making a plenum and then running it that way, but you'll probably suffer some performance loss off boost.

I doubled the carburetion on a 12A already by adding a .550" plenum. I used a totally retuned nikki on top of it, and the bottleneck is no longer the carburetor in any way. As a matter of fact, it was actually over carbureted.

Food for thought.

chuyler1

iTrader: (1)

I saw that same episode of Engine Masters a while back and just watched it again the other day and it got my brain churning about what could be done with that RB Holley intake.

I'd love for there to be a bolt-up EFI solution for my 12a that looks as slick as these all-in-one solutions but everything wankel mentioned makes complete sense. I just don't see it working unless they upgrade the unit to support dual plane and ITB configurations.

So for the time being, I'm between just grabbing the RB Holley kit or diving into the Nikki mods.

wankel=awesome

Trust me, I did my research on this whole thing while I was preparing to test it. The fitech would need modified, a spacer under it, and to be rotated 90 degrees for proper distribution.

As for the holley junk, avoid it. It's just heartache and headaches until you finally give up on it and go either nikki, or a proper multi port injection system.

MoonPopGun

Turning it 90 would not be that bad. I have seen that done with the superchargers, and well any time a Holley was used. Also wonder if there could be a fly by wire set up maybe so no odd brackets/throttle cabling needs to be done?

But how would the Fitech need to be modified? If the spacer is basically going to make it a single plane plenum that should make it a constant vacuum? Also what would it do with a lumpy v8 idle because that would not be a constant vacuum?

Also reading on a MegaSquirt system ""MegaSquirt® EFI Controller, the Mazda 13B rotary looks just like a 4 cylinder engine, except for having a 2nd set of plugs delayed by a few degrees. Each rotor has 3 chambers and executes an Otto cycle (intake-compression-power-exhaust) in 1080° of eccentric shaft rotation.
Therefore, setting up a rotary as a 4 stroke cycle, 4-cylinder (since it gives 4 tach pulses per 720°), and 2600cc (displacement x 2), works out fine."" I wonder if that will work too?

wankel=awesome

The problem with the FiTech is that you can't over ride it's self tuning capability, which would theoretically fix the whole issue.

Being able to set the injector timing up, and staging them with the engine needs, being able to change to the idle mixture where the engine runs best, regardless of what the vacuum readings are telling some otherwise pointless sensor.

It would also benefit from being able to load a good map for a 12a, or 13b to start with, and not massaging a different map for weeks to get it to run okay on an engine it doesn't know exists...

It having the ability to incorporate rotary ignition via an FC crank sensor or even a crank trigger would also help lots.

Oh, and being able to move the injectors down to the intake manifold for port injection would be yet another improvement.

And flipping the tb 90* on an open plenum would run pretty super terrible at low RPM with a 900 cfm throttle body sitting on top of it.

Air speed would be slow and throttle response would be lousy. Your whole low RPM map would be flooded with gas just so it engaged smoothly...

but why modify it this much? A product that does all of these things already exists? And it's called Megasquirt.