Cosmic_Eraser

iTrader: (1)

I have an absolutely bone stock 1997 (OBD-II) Miata and am trying to complete an MS-1 parallel installation. Currently, the car runs great on the stock ECU. I wouldn’t hesitate to drive it on a long trip tomorrow. When I plug in the MS, it won’t start.

First, I have to stick with the parallel installation. I know it is harder, but this car will be moving to Gwinnett, GA early next year and will have to suffer through the draconian emissions test. Also, parallel keeps the air conditioner operation simple.

I built the MS in July, https://www.miataturbo.net/forum/t23382/, and got a lot of help from this board. Then, I spent forever hunting down the harness components (I have a of spares if anyone is looking). In the mean time, I installed an LC-1 and the IAT sensor. The LC-1 is in a separate bung after I learned that it cannot emulate the stock sensor, contrary to what the advertisements claim. I finally got the harness built in mid-September, https://www.miataturbo.net/forum/t25529/. I plugged it all in and it wouldn’t start. I decided that I’d better walk away for a bit rather than get mad at it. Now, it is November, my wife is coming back from Tokyo in December, and I’d like to have this going before she returns.

What does work:
--On the stock ECU, the car runs great.
--My MS works great with the stim. All the gauges move when I turn the appropriate *****, except for the coolant gauge. The coolant gauge did work until I made the parallel mod. Now it is pegged, but, from what I’ve read, that is normal.
--My harness should work. I put more time into untangling the wiring diagrams and sketching out the harness than I did on assembling the MS. My final harness plans haven’t had any errors pointed out by this board and you guys were pretty good at finding mistakes in the first versions. The only alteration I made past the .pdf file is that I added a switch to be able to toggle between the MS and the stock ECU for idle control.
--My LC-1 works. I can plug my laptop into it and capture a data log for as long as the battery on the laptop will last. I did notice that the stock ECU tends to keep the car a little rich, but I’ll sort that out later.
--My IAT sensor should work. I don’t really see a way to screw that up.
--I am running the hr_10g firmware and I loaded the MSPNP base map. At least, I think I am running the MSPNP base map. Is there any way to definitively assure that the map actually loaded to the MS?

What doesn’t work:
--With the MS installed, the car will not start. The starter turns the engine, but it doesn’t even try to catch. The two outer LEDs on the MS light up. With the laptop on, the gauges in Megatune come to life. I live in a condo where I get a hard time for even opening the hood on my car (yet another reason to move), so it is hard for me to tell if I am getting spark or fuel. I am also very nervous about burning up a coilpack, so I don’t want to solve this with the trial-and-error approach. Is there anything obvious I should check next? Maybe there a is gauge I should watch more closely than the others. How long can I attempt to start the car before I fry the coil?

If anyone is in Jacksonville, FL and has time free one weekend, I'd appreciate another set of eyes on this. I'm even free to drive to the Atlanta area November 14-15 if I can't get any local help.

Chris Dobbins

patsmx5

iTrader: (16)

It would help a ton if you posted many clear, well focused pics of the MS board itself.

Also, what all guages work in megatune? Do you get an RPM reading when spinning the starter?

Post your msq here and we can go through it to make sure it looks good.

jobambo

Im in the exact same position as you, but mine WAS able catch starting but now it wont. I was running the exact setting as you, Hi-Res and the PNP map. I switched to the regular firmware and pnp and now it wont start.
I recommend first checking if your getting spark. I did this by putting a screwdriver into a sparkplug wire and set it near the intake manifold. Try to start the engine and you will hear a spark, turn it off asap.
Also check to make sure it can see your TPS by calibrating it, i noticed my wire was loose. once it was connected the coolant gauge started working as well.

Joe Perez

iTrader: (8)

Let's start with the very basics. Disconnect the ignition if you're worried about burning a coil, and disconnect the injectors too so you don't flood it.

Ok. When you are cranking the engine on the starter, do you see any movement of the RPM gauge in MegaTune? It should show you something in the 100-300 RPM range (depending on the condition of the battery) when the engine is turning over. If you don't have this, then you need to start focusing on the CKP/CMP inputs.

If you do, then I'd start focusing on the ignition. It's surprisingly easy to get the two ignition channels reversed, so try swapping them. Any effect?

What about fuel? Is the ECU turning the fuel pump on? Are you getting any fuel injection action? You can usually hear the injectors click when the operate- if you need to, stick a screwdriver between your ear and the injector body- pointy side down.

Let's have a log of you cranking, and an MSQ.

Cosmic_Eraser

iTrader: (1)

Thanks to everyone for the quick and thorough replies. I ran home at lunch and checked for spark. I also got some pictures of the MS board. I'll run a data log tonight while trying to crank. That is a great idea, Joe, I never even thought about running a log while the car isn't moving.

Here's what I did learn at lunch: I'm not getting spark. I disconnected the injectors and tried jobamo's screwdriver method. I couldn't get it all to stay balanced while the engine was turning over. So, I hooked up a timing light to the #1 plug wire. The instant that the key engages the run position and before the starter even gets going, I get a flash from the timing light. Then, I don't get any more regardless of how long I hold the key in the start position. I get the same result (including the initial flash) on the #2 plug wire. Even if I had the channels reversed, I should get something from the timing light, so I have some digging to do.

The engine turns over noticeably faster with the injectors disconnected. It also doesn't start to slow like it does with the injectors connected. Last night, Megatune showed about 90 RPM while cranking. I didn't have the laptop outside this afternoon, but it was probably 150 to 200 RPM.

I'll run the data log tonight and then post the pictures, log, and msq. Thanks again for all the help.

Chris Dobbins

paul

iTrader: (21)

What is the timing light getting power from? If its the blue plug behind the driver side headlight that thing is not live during cranking.(i think)

miatamania

iTrader: (9)

Where is the best place to get power from for a timing light?

Braineack

iTrader: (62)

a jumper cable connected to the battery...


that or the alternator.

Cosmic_Eraser

iTrader: (1)

--EDIT: I just noticed that the data log didn't post. I'll try again at lunch--

Paul, great catch. You probably saved me hours of misdirected troubleshooting. I confirmed that the blue plug does not receive power while cranking—at least not on a 1997. I ended up pulling my other car up beside the Miata and using its battery. I am getting spark while cranking according to the timing light. I can’t hear the injectors, even with the stock ecu. However, with the injectors disconnected the car cranks at about 250 RPMs instead of 90 RPMs. So, I’m pretty sure I’m getting spark and fuel, but it doesn’t even try to catch.

I took several pictures of my board:
cidobbins - Photobucket - Video and Image Hosting
There is also one picture in there showing the gauges working on the stim. With the RMPs low, the lights on the stim blink slow enough to see the individual sparks and injector pulses.

I also got a datalog while the car was cranking. This log should have two attempts with a short pause in between. I’ve never run a log before and people seem to have nice graphs. My options didn’t have that. I chose an Excel file over the text files. I can run another log if there is a way to get the nice graphs. Yes, I know the system clock on my laptop is off and that caused an odd date to appear in the file name.

I am using an unmodified version of the MSPNP MM9697 without MAF msq, except that I set the INNOVATE_LC1_DEFAULT option.
I have not been able to calibrate my TPS yet, but would that keep the engine from catching?

Joe recommended swapping the ignition channels and I am just out of time tonight. I’ll try that as one of the next steps. Based on the slower cranking with the injectors plugged in, I assume I am getting fuel. Is there an easy way to assure that the fuel pump is coming on? Again, the car is otherwise stock and the fuel pump is nearly silent.

I’ll do more testing tomorrow. Thanks again to everyone who responded with ideas.
Chris Dobbins

Joe Perez

iTrader: (8)

That is really, really odd. I cannot think of any reason why the fuel injectors would cause the cranking RPM to decrease. Can you tell a difference in the sound of the engine, or is this only displayed in MegaTune?

MegaTune produces the XLS file, and then to get the pretty graph you load that file into MegaLogViewer, a separate application.

Well, on most NAs I've seen (or rather heard) you can actually hear the fuel running through the regulator on the back of the fuel rail if you listen very closely.

Cosmic_Eraser

iTrader: (1)

I have attached my data log. I had not looked at the MegaLogViewer program much before now. That is a nice application and I'm going to register it tonight. Can anyone see anything obvious?

If anyone has an opportunity, could you let me know if my board (pictures in the photobucket link above) and the design of my harness https://www.miataturbo.net/forum/att...rt-harness.pdf show the spark outputs crossed? I can swap them to test, but I don't want to cause more problems.

I'll try to get some help to listen for fuel before this weekend. Right now, I'm a one man show and my arms won't reach the ignition when my ear is under the hood.

Once again, I truly appreciate the quick and thorough responses I have been receiving. One of the reasons I had put this aside so long is that I was sure I would be flamed by all the experienced users on this board. I was completely mistaken.

Chris Dobbins

Cosmic_Eraser

iTrader: (1)

The car does crank in a noticeably different manner with the large black plug above the injectors disconnected (this does disconnect all the injectors at once, right?).

When connected, the car cranks more slowly and every fourth beat is a little lower. I get a noise like ra - ra - ra - ruuh - ra - ra - ra - ruuh - ra - ra - ra - ruuh.

Disconnected, I get the faster, consistant ra - ra - ra - ra - ra - ra - ra - ra - ra.... noise.

This may be a moot point, though. It looks like I need to make sure that my fuel pump is coming on before I try to assure that the injectors are firing.

Chris Dobbins

miatamania

iTrader: (9)

Large black plug above the injectors should be the CAS right? So you are disconnecting fuel and spark.


Edit:
Wait, this is a 97. Maybe not. I'm not familiar with those.

Saml01

iTrader: (9)

Large black plug above the injectors is coolant sensor, injectors, and the coils. if i remember correctly.

Joe Perez

iTrader: (8)

Hmmm...

The RPM behavior in that log is a bit odd, but at least it's reading.

What I find exceedingly unusual is that your MAP reading is in the 120s-130s when cranking, and 154 when at rest. How many miles below sea level do you live?

Cpt_Slow

iTrader: (2)

Sorry to dirty up the thread... I have < 15 posts, so I can't PM you. I am interested in some of your harness components. Let me know how much to ship to 84095. Thanks, and good luck getting her to start.

-SC

Cosmic_Eraser

iTrader: (1)

Cpt_Slow, I sent you a message.

Would an odd MAP reading keep the car from cranking? I am only 40' above sea level and it was raining when I took this log. The MAP sensor was idiot proof when building the MS--there's really no way to install it wrong. I checked my vacuum line and had no kinks. When both ends are disconnected, I can blow through the hose. I tapped the vacuum line as instructed in the MSPNP instructions. I'm not running any forced induction, so that mid spot on the manifold should be just as good as the rear of the manifold, right?

I'm trying to chase down some help to check for a running fuel pump and working injectors this weekend. I think the odd cranking sounds were due to my unplugging more than just the injectors with that large connector. In the future, I'll unplug the individual injectors...assuming I can get my fingers in there.

I plugged the stock ECU back in and the car still drives great on that, so I haven't broken anything.

Chris Dobbins

turbobluemiata

iTrader: (8)

check in the megatune configurator that you have the right map sensor chosen. Mine was setup with the wrong one once and messed up the map reading when I loaded a new map

Joe Perez

iTrader: (8)

*Ding* I think we have a winner.

If you compare the scaling of the 2.5 bar sensor to that of the 4 bar sensor, the former puts out 2v at 100kPa, and the latter at ~160 kPa. Thus, if you have accidentally told the software (in MegaTune Configurator) that you have a 4 bar sensor, it'll interpret 2v as meaning ~160 kPa, which is just about what you're getting at rest.


You can eliminate the fuel pump as a suspect very easily. Pop the hood and find the diagnostic box, which I think is still on the hotside shock tower in '97. In it, is a terminal labeled "FP" and a terminal labeled "GND". Using a paperclip, scrap of wire, etc., short these two terminals together. This will turn on the fuel pump whenever they key is in the "run" or "start" positions. Try tapping on this terminal with the engine not running but the key on, and you should be able to hear the stock fuel pump. If not, open the fuel filler and stick your ear to it. If you can't hear the pump there, you need to consult an audiologist to have your ears examined.

Saml01

iTrader: (9)

That doctor is called an Otologist.