When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Ground is cleaned off, but the wire itself seems corroded. I might try to replace the whole line from the battery to the PPF (ah...another adventure). It never had a problem starting the old motor with the same ground and sitting in a garage in san diego so likely no moisture or exposure to affect it. Of note, I removed both the antenna and CD player head unit which other than the MS3 install are the only major changes to the electrical system that I can think of. I think the next thing is a new ignition switch? I've already jumper'd the clutch switch so it isn't the problem.
I took the starter out today and did a quick bench test. With the battery reading 13.4v across, the starter bench tested seemingly okay with both the solenoid engaging and spinning up but hard to measure how much actual torque it was putting out. One note, when reading the cross voltage from the solenoid, it only showed 9v, maybe a bad starter? I took it to autozone and their version of the bench test said it passed. New one ordered anyway to rule it out but I'm getting a little more afraid that its either battery wiring (solveable) or something actually wrong with my new motor...
Quick way to check for inadequate remote battery wiring is to use some jumper cables as supplemental cables, if they're long enough its easy to ground the battery directly to the block, and depending on your hand and cable clamp size one to the supply side of the starter solenoid.
It was enough to get my heat-soaked solenoid to trip on my RX7.
Installed a brand new 4AWG ground from the battery all the way to the PPF with a freshly sanded and cleaned up mounting point. Good stuff. Wire was corroded a tiny bit so certainly possibly some increased resistance there. 2nd, I changed out the ignition switch. a cool $30 later and fresh contacts underneath the steering wheel so no possible intermittent ground/contact there. Last, new starter. The one I had tested fine at autozone but again, not a huge deal with only two bolts and two contact wires. Additionally I cleaned up the cable post from the battery to the primary lead.
Where did this take me?
115 rpm. Still dog nuts slow and a lot of head scratching. I might just go buy a new optima battery and try it? The Bosch one again, tested fine at local parts store X and claims to have all the volts/amps that it should require to be successful.
A couple other things to note that I may have screwed up. I'm basically going by the notion that the diypnp ms3 auto syncronizes my CAM and CRK sensors through the original CAM harness without me changing settings. I've wired them up according to the VVT swap megathread. Second is that I wired my AEM X-series wideband to the 5v ref on the options plug pin "H". I still don't think though that any of this would lend towards a slow cranking RPM. I keep going back to the idea that it has to be volts/amps and the primary relay since the circut only has two or three total stops before hitting the juice on the alternator!
So, going around the car with a multi-meter attempting to locate the source of this stray voltage drop is rather frustrating indeed. I have a couple ideas though for the crowd that maybe someone else with more experience can shed light on. The way I understand our starters, a 1kW starter is going to pull somewhere in the neighborhood of ~60-85a @ 12v. Longer cables suck and most online voltage drop tests state that anything greater than 0.5v drop across the battery primaries to the starter can cause slow starting. This means that the Miata starting system with a 15' cable from the trunk inherently drops more voltage just due to distance. It is also slightly undersized for an initial amperage draw from a 400cca battery but most OEMs do this and it doesn't necessarily mean much.
Where does that leave me? My tests indicate that voltage drop is above 1.0v at the starter and that there is some parasitic voltage in the system which actually increases the amperage draw to the starter. High amp load and lower voltage (i.e cranking rpm below 10v) which is indicated both by my voltmeter and by the composite log. In my mind however, this would pop the 80a main fuse and it hasn't (more head scratching). I've tried pulling various fuses to isolate parts of the wiring harness to see if anything (wipers/windows/o2 sensor/fans) might have been shorted causing this but so far, nothing seems to change the drop and the rpm.
My bird brain knowledge of the starting system following the mellen's diagram is that the only stops the battery cable makes prior to the 80a main fuse is the starter and the alternator. I've tried disconnecting the alternator as well entirely just to see if it was sapping volts. I think my next replace might be the ppf cable itself from the battery. Debating running a full 4AWG lead directly to the starter then a "Y" to the junction box and alternator. I hate messing with factory wiring and it has to be something else though.
Don't eliminate the easy stuff first. I've seen some incredibly rusty, corroded, shitty looking Miatas crank just fine. Either put a new battery in or try your battery in another car.
Agreed, I keep coming back to the "simple solution" has to be the answer idea. I dropped by the local auto store and had them check it again, it passed. I told them it was the root cause of failure in my car and they agreed to let me credit it towards something else. I picked up an Optima yellow top 51R which I assume is a higher quality unit than the Bosch U1RT and only for twice the price! It didn't solve my problem. A few more cca's and probably less charge/discharge cycles than the previous based on how many times i've tried to isolate this problem ended up spinning it a few rpms faster but not an order of magnitude change. So, its not the battery, and its not the starter or main relay unless both new and old units each are failing.
I'm starting to dive deep into the wiring diagrams and think that the only other changes I have made besides the obvious ignition related stuff (when disconnected, same slow rpm) is removing the radio head unit, rear retractable antenna, and cruise control. I'm chasing down what feels like something has to be shorting, but only when the key is in "start" and most of the connectors in the engine bay are disconnected (so they can't be drawing current).
Yeah, I have a brand new unit on there. Exactly the same effect as the previous NB unit. Unless I am 0 for 2.
Same with batteries. "New" battery that tested fine, sat on my shelf while this rebuild was happening and now replaced by an actual new yellow top. Now 0 for 2.
Tried unplugging the alternator and various fuses to see if any of them were leaking volts. No change so far.
I had a similar problem, and it was a problem with a replacement main cap... Apparently clearance was tighter and wouldn't let the engine rotate much at all.
Well, the engine rotates by hand well enough so back to the drawing board. I've gone full circle on the wiring harness and rebuilt everything from a fully sequential ignition to a batch fire setup. Additionally, I rewired the o2 sensor to use the factory plug vice the options plug in the MS3 and I'm thinking that may have done more to get good signals to the right places than tearing apart the alternator wiring harness. Again, the simple solution....
So here we sit. It cranks at ~150-100rpm as battery voltage drops and holds pretty steady there without plugs. Laughably, I just realized after some serious headscratching that my composite log doesn't have a cam signal. FML. If this whole no start issue is due to no cam signal, well.....I would say I'm going to burn the car down but I'm already trying to do that as it is.
I took apart the pigtail again and confirmed with the 99 manual and VVT megathread that the wiring should be, yellow blue input direct center of the three inputs from the original CAS sensor, then power and ground to the opposing sides. I tried swapping them but still no signal so maybe its possible I fried the sensor? Is there any way to check and can someone confirm my wiring is good. The trackspeed engineering pigtail was helpful to get the connector, but had three black wires coming out of it so I easily could have miss-assembled it.
On the product page for the pigtail, there is an image which calls out which wire does which function. Just re-select "Cam" from the dropdown and the image shows up. FYI
Beautiful. If I haven't said it before, I will again. No blame other than my own noobishness for this type of stuff. I don't mind learning the hard way and I think all the information is on these forums (or trackspeeds website!). I'm supposing that the slow cranking rpm is just one symptom of high resistance wire and poor battery voltage but the cause of no start is at least CAM signal if not other problems as well.
Right now fuel pressure at the rail is static at 70 psi and unscrewing the radium DMR all the way doesn't change it. My guess is well. I have no clue on that one. I think a new fuel pump might be ok n order once this thing is running.
The good news is that the only way as I understand it to test a hall effect sensor (CAM) is a bench test oscilliscope which...surprisingly I don't have. Maybe its fried or not. Looking hard now at it the MS3PNP pro to see if a VR pull up resistor is required or not for the 94. Ordering another sensor anyway just to disprove it.
She lives! Or he. Possibly hard to say which one like all the kids these days who refuse to identify. It ran for a total of 10 seconds a couple times before I stopped to check for leaks. So thank you again for all the good suggestions. Now for the hard part. I don't know what fixed it in total .There were many small issues but no one single thing that brought it all together.
The big shout out goes to Toby at AED who thought that the fuel lines were swapped (they were) which allowed me to get fuel to the rail and set the pressure correctly. Genius. Also helps if you read directions and pay attention about where to install gauges in a fuel system. Second, I went through every single fuel related setting from ID725 in TS to battery and dwell settings for the LS2 coils. Not sure any one of the settings specifically was the issue, but it helped me start to rule things out. My going in fear was proved incorrect, that the battery or starter were bad the second time around and it was still those things. It wasn't, everything else was just shitty.
A couple things I have high confidence in. Timing was definitely an issue. I double checked mechanical timing and with the TDC mark lined up with the cover, a dowel in place of spark plug one, and both cams verified (and the ATI pulley checked as well for manufacturing error!) and it was correct. 36-2 tooth was installed correctly on the pulley and tooth gap was fine. Cam signal after verification was present in the high-speed logger. I had been getting lots of starter kick back and my guess was that either old fuel in the tank was causing pre-mature combustion or the coils were spontaneously discharging...or they weren't at all and they were just pulling high current. I re-did my coil harness, checked grounds to no avail, but the new loam looks better and I feel more confident in it. I also removed a dealer added interlock/kill switch that may have been causing issues as well.
Now for the hardest part. Tuning. It won't start until I give it 20-30% throttle and then it does not idle. It surges high (3000) and then dives to 0 unless I play with the throttle and it will stabilize near ~1500 with 4-7% throttle. I've backed out the idle screw on the Skunk2 a bit and I've pulled ignition timing back on my entire map since it likes to jump up and around in RPMs quite a bit. At fixed advanced (10*), its a bit better behaved but once I switch to use table, it goes nuts.
I set the trigger offset at 4.4 after verifying with a timing light it was damn close. 5 was right out, and 4ish was pretty rough. It likes falling edge, skip 1 pulse, 0% cranking advance, and only 3ms on cranking dwell. 4 with a battery correction is probably overdwelling it.
I moved the MAP line up to post throttle body. It was on the plenum but didn't really ever show any drop during cranking. I've been told this spot has non-laminar airflow and I respect that, but the flat top manifold with no EGR doesn't have any place at the back of the head to put it. Additionally, my AEM X-series is wired into the narrowband inputs into the MS3. No CAN inputs.
I've set the NB 2-wire IACV at 511Hz, which is the closest I can find in tunerstudio to what is on these forums for the ~500Hz resonant frequency.
Its dead jim. More on that topic in the build thread I'm about to start but needless to say, I've learned a lot of rough lessons in a very short period of time about machine work, engine shops, built engines, and aftermarket engine control units. For those curious, The engine didn't sieze, but it refused to develop oil pressure during cranking (suspected air bubble) and after start which significantly damaged the thrust bearings, all mains and rod bearings, and crank journals. Block is trashed. Crank is trashed. Every bearing and seal in the bottom end of it is trash. All of this was new, to include the crankshaft after the previous engine tear down revealed un-repairable damage previously. The engine was pre-oiled but based on discussion with the builder, it seems that I used the wrong oiling port (drivers side of the block) and too low of a psi setting on the compressor. furthermore. The oil pump seems to have survived this event in its entirety unscathed and I'm not sure I want to try it in another engine anyway.
This is after not driving the car for nearly two years after putting it in storage for deployment then coming home to start this project. Fortunately, (or not for my bank account) this saga isn't over. But its going to take a new direction. I still have all the trackspeed turbo bits, MS3, and an entire parts catalog worth of suspension stuff for my next victim.
04-08-2018, 08:20 PM
0
