Ok, it's time to go standalone. This morning I learned of a standalone called MicroSquirt that can be had for under $400 with the harness. It got my attention.

I have a few questions that hopefully I can get answers to and not discourage me from getting this ECU.

The car is an '87 TII. Never had issues with the stock fuel injection (Rtek7 chipped ECU) until recently when it began to hesitate/buck at low RPM @ around 10 inHG vacuum. I've given up trying to isolate the culprit.

So I see that the MicroSquirt batch-fires the injectors. Is that correct?

Are all the spark plugs fired like stock? Can't figure that out looking at the diagram.

The ECU will be controlling 4 X ID1050 injectors. Ignition will be with smart coils. The engine is currently configured to support over $300 WHP (turbo, FMIC, etc.).

How is batch-fire injection in terms of being able to support 400 WHP (my target) dependably and reliably? How about in terms of fuel consumption vs sequential? Drivability?

I had been leaning towards the Link Monsoon (yes it has more capability) but if the MicroSquirt can handle the job, I'll b quite happy with the savings.

Thx in advance

 

yes but you have limit ins and outs I have done plenty.

 

Hi Clubuser, thanks for reaching out.

So to answer some of your questions:
- Yes, the Micro batch fires the primary and secondaries. This is going to amount to a very marginal efficiency loss over sequential. Completely negligible.
- Batch fire is going to support 400hp, no problem. It would support 1000HP.
-I can't comment for certain whether running 1050CC injectors is going to present a drivability problem on the primaries vs the stock 550s, but I suspect it will be ok.
- For spark you can either run the stock coils (FC Mode) or Individual coils per plug (FD Mode).
If you want to go FC mode, you need to read my thread on MSExtra.com:
https://www.msextra.com/forums/viewt...581658#p581658
because there is a huge glaring issue in the manual that gives you the wrong pinout for the trailing coil.
However, you mentioned you want to use smart coils, so that won't be a concern for you. You will be running the Leading pair of smart coils with the same trigger, and individual triggers for T1 and T2.
- I had some problems with my CAS signal, and I ultimately resolved those by enabling the VR1 shunt, which requires you to open up the Micro case, and bridge a solder pad on the board, very simple. ONLY the internal shunt worked, using an external shunt made no difference for me. This was all covered in the MSExtra thread I linked above.
-You are probably going to want to spend the 60$ to get the full version of TunerStudio to enable to autotune feature.

Details of my installation:
-Micro controls E Fan, Boost Solenoid, and the Twin-Scroll Solenoid
-I installed GM IAT, CLT, and TPS. Had to do a bit of fab to mate the TPS to the throttle body.
-I'm not completely tuned yet, I have been working on another standalone project that has taken all my spare time. My off-boost areas are pretty close, but I haven't taken the time to tune the boost regions. That being said, I'm fully confident the Micro is running properly and it will do the job just fine.

If you have any more questions, fire away

 

If you have a drivability issue now, you really should fix that before going to a new computer setup. That way you are not chasing a problem while trying to get a computer set up properly and tuned.

Batch fire will require a richer mixture than ideal with the stock manifolding because one rotor will starve the other of fuel.

I have 2xID1050x injectors on my setup, which uses an MS2/Extra running fuel only. I bought them because I did not want to screw around, and get the proper injector characteristic data. What I got is injectors that heat soak and the engine runs progressively leaner on road trips. Where I would be at 13.5:1 at a 70mph cruise would be 16-17:1 after three hours and the car is barely drivable. Fix it to run well hot and it is pig rich when not heat soaked. This is not a CLT or IAT related issue.

 

I completely agree with Peejay that you want to start with an engine in good running order, or at least figure out what your issue is beforehand, if at all possible.

How would batch fire cause one rotor to starve the other of fuel? Each rotor has it's own intake runner with it's own injectors in it.

 

The runners share a plenum. Say the injectors fire when rotor 1 is at TDC. Rotor 1 gets fuel injected during the middle of the intake stroke, rotor 2 gets fuel injected during the overlap period, when airflow is a little backwards at light throttle/low RPM for various factors, like the end of the previous intake cycle, the exhaust pressure pushing flow upwards, and rotor 1 pulling a strong vacuum on its intake port, which is connected to rotor 2's intake ports via the shared plenum. So the fuel gets pulled up where some of it wets out, some of it gets pulled into the plenum, some of it gets into rotor 1.

You are NOT going to be firing 1050cc injectors twice per revolution. The pulsewidths are already going to be crazy-low as it is, like in the 1.5-2.0ms range, firing once per revolution.

Mind you, I have been running two injectors batch fire since I first ran MS in 2008 or so. I have over 100k miles of Megasquirted rotary. Obviously, it does work, but it isn't ideal. Having an ITB style manifold helps a LOT compared to the Mazda "ping pong" manifolds.

 

Thank you all much for the feedback. I had a hunch batch firing would not be all roses with the stock intake.

In regard to the drivability issues, I get no fault codes from any of the sensors (per "Reading Mazda RX7 86-88 Error Codes" method).

I have installed new injector clips and sent the injectors out for service. After the injectors were put back (about 2k miles ago), the problem seemed to have been taken care off. But about 2 weeks ago the symptoms started to reappear and now its back full force. I monitor the leading injectors' pulse while I'm driving and when the hesitation occurs both the duty cycle and tach are rock solid. WTF! So, if it's not the injectors have to think it's the factory ECU. The thing is with the slightest pressure of the pedal the hesitation disappears and when I floor it pulls hard thru redline! Mechanically the engine is 100% mechanically sound. Over 100psi on every face of each rotor. 16 inHG of vacuum at 850 RPM (stock S5 block). I'm 100% confident by eliminating the factory EFI and going standalone with a new harness/sensors this headache/heartache will disappear.

Peejay, regarding the ID1050x, in a sequential application do you believe they are too large? I too questioned it when I was suggested to run them with the Adaptroninc PNP but was assured it would run right. What's your suggestion on injector sizing mix for 400 whp?

 

FWIW I am running TWO 1050s for an engine making about 270hp, at 2.5bar fuel pressure (stock GSL-SE). Four for 400hp should be fine.

Regarding your issue. If you are watching pulsewidths and they are staying steady with your hesitation, clearly it is not a fuel issue, so why do you think it is a computer issue? If the injection is correct then the computer is working correctly.

I would look closely at the trailing ignition, plugs/leads/coils. Trailing is super important at low load. An engine can run at WOT just fine with only leading ignition but run very poorly at light cruise. If the plugs have more than 10k miles on them they are due for replacement.

 

Ah I understand. Is this problem not inherent in most batch fired engines though? Does the problem not more or less average out between runners? From the reading I've done on the subject, it amounts to rather small efficiency losses.

I did some quick napkin math, and with 550cc/min stock primaries, my Reqfuel is 3.95ms. (In MS speak, Reqfuel is the injector pulsewidth required for your engine at 100% Volumetric Efficiency. In MS we adjust VE numbers in the fueling table, not actual pulsewidths like some other systems use) If I change to 1060cc/min primaries my Reqfuel goes to 2.05ms.
So in the primary only region of the table, you're going to see numbers like 30-50%VE. With 1060cc injectors that amounts to 0.62-1.0ms pulsewidths, and that is below the stable region on typical injectors, I believe. I think the minimum on those IDs is something like 0.85ms.

If you need that much fuel, you could go to a huge secondary and leave the primaries alone, or go with like a 750cc primary and a 1300cc secondary.

 

Have you cleaned up the grounds as per this page?
https://www.aaroncake.net/RX-7/grounding.htm

That can be a cause of hesitations.

 

Thank you all for passing on great information!

PeeJay:

Took your suggestion to focus on the trailing coil. I figured no loss taking it off the car and driving it to see what happens. I noticed the TSP wiggling a bit when I moved the throttle, so I secured it down. When I removed the coil, I noticed a connector with wires grounded to the strut tower. Did a disconnect/reconnect. Then took the car for a spin. To my amazement, I could not get the engine to hesitate/buck! Let's see if all's well going forward.

Taranabes:

Thanks for pointing me to Aaron's write up on corrosion and fuel injector sizing. I see Aaron points to the connection I mentioned above. I'll go further and do to it what he suggests. The ECU 2 block ground along, the negative cable 2 frame, and strap from manifold to fire wall I've addressed in the past. And regarding the ID1050's, I'll stick with 750's in the primary and larger on the secondary. Would have preferred ID 750's but I don't think they are available.

 

Well here's hoping that's all it was is a crappy ground!

I guess this thread got me motivated, cause today I drove my RX7 in anger for the first time in like 3 years. I had taken it around the block a few times off boost a few months ago, but today I took it right up to the wastegate and roughed in the fuel, felt great! Still idles a bit rough but otherwise feels smooth.

 

Yes, which is why the OEMs abandoned batch fire as soon as they could fit the necessary hardware for sequential injection into their budgets. Other manufacturers who thrive on being cheap (Toyota) would do a semi sequential thing.

Before Ford and GM ditched batch fire on their V8s, they had intake runners in the 22-24" range which helps tremendously. GM went to sequential injection with the LT1 in 1993 (1992?) because it allowed them to have much shorter runners than those old loopy TPI manifolds. Buick Grand Nationals had sequential injection at least as far back as MY1986, and their intake manifolds looked like 3/4 of an LT1. I tried running one batch fire and it was hilariously bad, like only running on four cylinders bad!

It is also why K-Jetronic was used on luxury cars up until the early 1990s (I think one flavor of 911 Turbo was the last to use it in 1993-1994ish!) because when injection flow is constant instead of pulsed, you can gets some very very good drivability and mixture control.