New MS3X Build and install - Car runs very rough and wont idle unless I give it gas
 

So as the title states, I built a MS3X with DIYBOB. I have done the following:

• Unplugged stock ECU, plugged in MS3X, mounted it in the glove box, DIYBOB mounted underneath glovebox with zip ties.
• Ran vacuum line from T-fitting between fuel regulator and manifold barb.
• 1K Ohm resistors on unplugged TPS between signal & ground
• 1K Ohm resistor on diagnostic plug between IG+ and B-
• removed ST_SIGN fuse from fuse box
• loaded up https://trubokitty.com/assets/tunes/...93%201.4.1.msq base map
• changed fueling from Sequential to Off as I still have my harness wired for batch injection.
• made sure O2 was set to narrowband instead of Wideband since I am currently running with my OEM o2 sensor

After this, I plugged in the laptop, turned the key to on, and the device was seen, and the gauges showed their default values. I tried then to start the car and it took a bit of cranking. It eventually fired up but immediately died. Cranked again, and it would fire up after a bit, but then it wouldn't stay idling unless I gave it gas. It smelled like it was running really rich, but that could just be because I have a test pipe and I dont often idle for long. The car SOUNDS like its running on all 4 cylinders, and doesnt sound too bad, but it is very rough and wont idle on its own. Any ideas here guys? I know almost nothing about tuning this thing, and I just wanna get it running so i can install everything else.

I've done a short data log which I attached below, but I didnt wanna run it long because I dont wanna blow up my engine.

good job getting it running. now you just have to tune it..

Its not running... and I know it needs to be tuned, I am just trying to get it to idle properly so that I can do that.

curious why you removed ST_SIGN fuse...

double check the req_fuel value makes sense for your injectors.

All of the documentation I was looking at for MSPNP said to do so otherwise you can fry the MS on 1990 - 1993/1997 (EU Cars with 1.6L) miatas. Diyautotune's documentation somewhere basically said that because of hijacking some of the wiring of the miata to do some things.

here's an example:https://trackspeedengineering.com/files/Instructions%20for%20the%20Basic%20MS3%20V3%20140. pdf see #6

do you have a MSPNP or MS3 Basic from MS Labs?
:likecat:

you built to my docs, and my docs dont say touch the sg_sign.


not that this will solve your issue, but if these things are custom built, and can be built various ways -- hijacking potential damaging wiring is one way.

If you are building a standard MS3X you need to only look at the documentation on turbokitty.com. Stop looking at documentation for other MS's made from other people. Things are done differently and what they say to do is not what you need to do.

Right-o i will put it back in and then try again, though as you say, i am doubtful this will get the idle to run well.

Hey man, i was just trying to be safe, I didnt want to blow up my $800 computer. I will replace the ST SIGN fuse.

Hey Scott, quick question: I dont have my AFM removed, if I still have my AFM connected, does the AFM do the fuel pump jumping as outlined on trubokitty, like this:

90-93 (NA) Air Flow Meter (AFM) Connector

Using a small length of wire leftover from the installation kit, jump the two pins for the fuel pump. This connects the MS to the fuel pump relay so it can control the fuel pump properly.

The EBC solenoid must be connected to a switched power supply inside the engine bay. The blue power connector near the driver's headlight is the ideal choice.

https://trubokitty.com/assets/images...el_Jumpers.png

Even though I'm kind of at the same level you are on this, I got my super-stock 90 started yesterday with a DIYMS2 and I did NOT remove the fuse mentioned for it to start.
I have stock EVERYTHING, AFM, sensors, etc, it starts right up, altough it doesn't idle for shit.

I initially loaded some random basemap for a 90 but it seems like it was for an older version and after going through most of the settings today at work I came to the realization that it was pretty much useless, it's better to load the shit by hand, matching the information online. Not saying the MS3 one is going to be the same, perhaps yours is better according to trubokitty instructions, but in my particular setup I had to do it by hand, and to be honest, you get a better understanding of what's going on when shit goes south because you know what's there.

For example I found out that the map loaded had the injection set as if I had it setup for sequential, and it kinda ran like that but I think it makes sense what it was doing now. (Basically it was revving up, then dying, and up and down in a loop, I guess that if the engine was expecting 2 cylinders to have fuel but they didn't it would do that)

So, long story short, the AFM activates the fuel pump for you. DONT run the purple wire from the DB37 to enable the fuelpump until you have the AFM disconnected.

I am now working towards a very very basic setup to get the ECU to act almost as a carburetor. I don't care about advanced stuff for now. OverRun disabled, IAC Open Loop, fixed timing, I'm not giving it gas until I get it to idle. All the basic shit so I can understand where I'm standing at and what's affecting what. You gotta start removing variables from the equation because then you change A but B, C, D, E...Z gets out of whack.

I learned this shit while trying to get power on karts on the dyno and tuning carbs, now I have to translate all that to tunerstudio. Once I'm standing kinda OK, I'll get playing with other shit.

My AEM X series arrives on Wed and I'll make an adapter for the BMW TPS on Saturday so we can keep in touch and come up with something good.

no, the opening of the afm jumps the fuel pump like normal.

can you post your msq, and composite log and datalog of cranking for a good 10-15 sec?

Yeah, I wired up my injectors for sequential, rx8 injectors and am in the process of modding for vTPS then I will try again with those settings enabled/changed. I had also not seen the unlockable tables for thermistor and AFR tables which i hadn't messed with. I also noticed my fan thermistor switch on the front thermostat was cracked off, not sure if that would affect idle at all but I ordered a replacement.

Hey guys, finally got around to getting the miata running, it seems alright so far, runs and drives, no real breaking up or anything, but the AFR's run rich when idling and when I stab the throttle, as well as a bit during acceleration, it runs a bit lean.

Some changes since I last tried starting:
1.) 420cc RX-8 injectors
2.) Wired up injectors in sequential
3.) vTPS (wells 4185) wired up and mounted
4.) Fans both wired together
5.) AC + PS Deleted

Ive attached a log so anyone interested in helping can take a look for me. Any help would be much appreciated.

Thanks guys!
-Gary

MS controls your fan now... Typically you unplug this with a MS install, otherwise you have two devices trying to control the fans...

sounds like you need to tune the idle to be leaner, and tune your acceleration enrichments...

Cool, thats what I thought. Will VEAL tune that completely, or should I do it by hand? The AFR's I have set in the AFR table I havent touched much, I pretty much loaded up your MS3X tune on trubokitty, made the appropriate changes for sequential injection, vTPS and changed my req'd fuel value for the larger injectors. I did run VE Analyze on the datalogs I have logged, and it changed quite a few values on the VE table, so maybe I will keep trying that but with VEAL instead since I have the full software now.

veal won't be able to tune either of those. that's all manual adjustment

Yeah you don't want to use VEAL for idle tuning. I also prefer to use the IDLE VE table over the standard VE table for idle cells.

idle is stable, it's easy and preferred to hand tune. i dont even think you can VEAL the idle ve table.

Hmm it was super lean at first like 19:1 until I did a VE analyze in megalogviewer on the datalog. It then seemed to make idle really rich. I also switched from open-loop warmup idle to closed loop because the car was idling at like 2400 rpms at that mode. Seems to idle normally now, so idk if the VE analyzed and changed the table for idle somehow, or if switching to closed loop idle fixed that. My AFR table is right where I want it (I think) it's basically what braineack has set in his startup tune on trubokitty, so I only wanna mess with the VE table to get things closer to Target AFR correct?