12A carb to Megasquirt. Has it been done?
 

Hey guys im tossing around ideas for fuel delivery, and was curious if anyone had sucessfully doen a megasquirt conversion on the 12a.

If so what kinds of parts am i looking at needing, apart from the obvious, and what kind of cost for said parts can i expect?

Yes it's been done. I have an '85 that was a bridgeported 12A and now it's a S4 turbo 2 motor with upgraded turbo. Megasquirt is running great except for the problem I'm having with my O2 in the tuning software. It's a little irritating not having that to work as it really helps with the datalogging.

As far as the parts I have the stock gas tank back there with a low pressure pump feeding a surge tank and Bosch 044 pump feeding the motor. I've boosted maybe 20lbs by mistake and the setup is more than enough with 4x1000cc injectors made by RC Engineering and -6 AN lines from tank to motor. Also running a parallel fuel rail setup as shown on the FC3SPro.com site. Turbo is a GT45 and the motor is mildly ported. No intercooler just water injection.

It's been running for maybe two yrs. now with a microtech previously installed and the megasquirt by far is a more flexible ecu in my opinion. If you're willing to learn and have patience it's the best bang for your buck ecu.

As for pricing it all depends what you're willing to spend. The Megasquirt ecu inlcuding wiring and the upgrades I've done to it cost me around $400 more or less. Good set of injectors cost maybe $300 for 4. Fuel setup including pumps, surge tank, lines, and a good cleaning of the old gas tank( I kept clogging filters with that dirty ol thing lol ) maybe $500-$600. Everything for the car is priced just parts only all the work was done by me so I saved a lot.

There's really nothing different about Megasquirting the 12A versus the 13B. Your only complication is that you are going to have to arrange an intake and throttle body setup and what you do will sort of depend on how wild you want to get. ITBs are great for top end but are always going to be a bit funny around town and more difficult to tune (must be done alpha-N). A simple RB lower intake with a throttle body bolted in place of the carb is going to be the best choice for a daily driver but you will need to modify the manifold to accept injectors and adapt whatever throttle body you end up with. The easiest thing would be to use a 2nd gen CAS in place of your dizzy and a set of 2nd gen coils.

For the basic idea on what is involved with setting up the Megasquirt, see my writeup sticky at the top of the forum. It concentrates on the 2nd gen, but building and programming the 'Squirt is the same regardless. And the wiring is mostly the same.

Yea the write up will definitely get you situated. It will help solve a lot of the confusion for building and configuring the Megasquirt for the rotary. Thanks Aaron!:icon_tup:

OK great!! thanks for the feeed back ill eyeball the write-up an see what i can figure out. My only real concern was working out a throttle body and intake, working injectors and such.

Anyway, thanks again for your help!

Im taking on this project as we speak. the main thing to take into account is that the 12A irons have no injector placement. so you will need to work that out. also, the 12a ports and the fc t2 ports line up but the bolt pattern is different, so an adapter plate will need to be made.

you will now be able to use a t2 intake and 2 injectors. now figure out fueling, and you are good to go.

If you use 2 injectors with a TII intake, bungs need to be welded onto the primary runners. The stock throttle bodies are staged so if you just run injectors in the secondary location, the car will run like poop until some throttle is applied. The secondaries are also up really high, which will effect light load throttle response quite a lot.

So the best bet is to run the same primary/secondary injector system that the 2nd gen runs. Weld in some bungs on the lower as close to the primary ports as possible.

did you know where to get the adapter.

The adapter will have to be fabricated. And as I mentioned, the injector information in that post is incorrect.

tweakit.net sells the adaptor plate. I will however make my own.

Huh, well look at that, they do. :) I stand corrected.

However, what I mentioned about the injectors is true. Just running with the secondary position will be fine at wide open throttle, but not at light loads and idle. The manifold or adapter will need to be modified to place a set of primary injectors as close to the ports as possible.

yah, but is this a TB issue or what? because i have seen people run their injectors very far way from the engine ( almost a foot away ) and they idle just fine. i have also seen them run the 2ndary injectors just fine at idle and light cruise as well. so thats what im wondering. is this a TB issue?

The stock system has staged secondary throttle plates, so at very low throttle and at idle, the secondary runners are cut of. I supposed that the throttle body can be adjusted to open up the secondary runners, but I don't know what the result is.

It may also depend on an individual's perception. Tuned rich, injectors high up in the intake would probably work fine. But personally, I don't consider 11s at idle and 13s at light load to be "tuned" at all. Now a single big throttle body that opens all the runners may make this work a little better, and the EAE in the MS2 would help throttle response a lot. But you are still fighting runner volume and low velocity.

are you talking about the stock TB or the modified ones? my current one is modified and works realy well. i am also toying with the idea of adding a 65mm tb ( per rotary gods recommendations ) instead of the stock one.

Stock throttle body. It shuts off airflow to the secondary runners at idle and low throttle.

o ok. ill be using the completely modified TB.

This isn't really the case anymore. You can tune a hybrid speed density/alpha-n setup. Just tune speed density down at low load, and alpha-n above that.

The 3.0.3 alpha code also has ITB mode, which more or less does the same thing for you, but with only 1 table (and a couple of curves to set the switchpoint and the desired load at that switchpoint).

On my 20v 4age (comes with ITBs from the factory), the 2 table method proved to give the best behavior of speed density (good fuel economy and low-load drivability without popping in the exhaust) and alpha-n (finer tunability at mid to high load).

I have not yet tested ITB mode in a car, but intend to very soon along with sequential on the MS3.

Ken

With the ambiguous vacuum signal of a rotary with ITBs, I'd think that speed density would be more trouble then it is worth. From the (minimal) experience I have with tuning Sam's car, there were so many load points that were in use during both medium and high throttle that the map looked like a mountain with a valley in the middle. The car would go from 20" to 0" at the touch of the throttle. It was like tuning a bridgeport, but worse.

The new MAP sampling code might help there. It actually times when it samples, and then samples for a certain number of degrees, returning the lowest MAP point it saw.

This has taken engines that were previously unusable on speed density, and made them usable, and has taken engines that were previously fine, and made them better.

Ken

Sweet! Now that gives him another excuse to get off the ancient firmware he is using! Sam, are you listening? ;)

uh oh.... seems like the impossible isnt so impossible any more mr cake.

I didn't say anything was impossible. In fact, I rarely say that something is impossible because almost anything is possible (except those things that violate the known laws of physics, etc.) with enough time, money and talent.

There are two conversations in this thread:

1. Injectors way up in the intake. If the only injectors feeding the engine are the secondary injectors, there will be issues as I have already mentioned. Many of these issues can probably be tuned out or at least minimized, but having injectors way up in the airstream is always going to be detrimental to light load, throttle transitions and idle. That is just a fact based on the length and volume of the runners. Mazda knows this, which is why they place the primaries as close to the intake ports as they possibly can. Look at most other manufacturers, and they place their injectors as close to the intake port (on the cylinder head) as they can.

2. ITBs and their vacuum signal. In the past, Alpha-N would have been the way to tune this. But apparently, the new MSII code (which is still in Beta, from what I know) will be able to work around this.

You will still probably want alpha-n for higher load, but this code makes SD very tunable at idle and low load and cruise.

The 2.1.1 beta (which will not be updated again) supports this feature, and the 3.0.3 code also supports it.

3.0.3 is currently in alpha state, but I'm going to bring it into beta state very soon, and release shortly after that.

I'm currently using it on the 20v 4age with 2.1.1, and I'm using code that is mostly the same on the ms3 for the rx7. In both cases, I was able to get the engine to run smoother and more responsive by tuning the MAP sampling.

One other report I have had on this code is that people can now increase their MAP lag factors (which makes them do less ironically) which makes response better.

One of the benefits of MS in general is that since I write the code, and I own an rx7, I make sure that the code I write works on the rotary, and benefits it as well (as I've done for the EAE mods I made to make EAE usable on the rotary... currently only done in ms3 but will be backported). This MAP sampling code helped my more or less stock S4 NA get a better MAP signal, so I'm sure it'll help others.

Ken

so what if i dont have a msII and just a ms1??

MS1 is either speed density or Alpha-N only, according to the settings I just checked in Megatune.

You can always just get an MSII module and plug it in.