Na1 starts up like normal then immediately dies.

[JDMB20TDA]

I had the car sitting out in the cold a few weeks ago as I worked on another car in the garage. After a few hours went to pull it back in and it started right up as normal then immediately died. It seems to run fine as it start but dies about 2 seconds after running. I can feel the pgmfi relay clicking and it's getting air/spark/fuel.
I cracked the relay open and it appears visually to be fine (no broken traces/bad solder joints).
Ive done some research and it seems like most of the issues point to the pgmfi relay or ignition switch Is there a way to "test" out the relay. I can reflow the solder and hope that fixes the problem. As for the ignition switch ill be taking it apart and cleaning it up with some contact cleaner. Any ideas or known issues that have similar symptoms?

 

[Old Guy]

There are likely other things that can cause your problem; but, you have identified the likely candidates.

One of the modes of failure of the ignition switch is that it will start the engine in the 'start' position; but, if the main contact is failing or has failed, when you release the key back to the 'run' position it fails to maintain power to the main FI relay and the engine stops. This type of failure can be intermittent and you may be able to get it to operate again by switching the key back and forth; but, when it happens the writing is usually on the wall.

If you can repeat this start / die sequence a number of times I would put money on the problem being the ignition switch. If the start and then die sequence happened once and then the problem went away, then it could be either the ignition switch or the main FI relay or other things. If the start and die sequence happened once and then it was not possible to restart the car at all, it is more likely that it is the main FI relay; but, can be other things.

The 'tell' between the ignition switch and the main Fi relay is to watch the instrument cluster lights. When you first turn the ignition switch to the run position, the lights for the TCS, ABS, MIL all go through a self test sequence. If the ignition switch contact has failed, likely non of the lights will go through that self test sequence. If the main FI relay has failed, the TCS, ABS lights will probably do there thing; but, the MIL light will not light up. There is always an exception to this rule which is caused by the fact that the main FI relay is a two in one special. It has two relays in the case. One that powers up the ECU and the other that is controlled by the ECU and powers up the fuel pump relay. If the ECU relay fails, the MIL will not go through its self test. If the fuel pump relay fails the ECU may power up and the MIL self test may complete; but, the fuel pump will not operate and you will not get the 2 second prime pulse in the run position (that does not sound like your problem).

As a suggestion, clean up the ignition switch if you want; however, given the general hassle of working under the dash I would be inclined to put a new one back in. RockAuto sells them for about $45 US. RockAuto also sells replacements for the main FI relay for about the same price. Clean up the original and keep it as a spare.

 

[JDMB20TDA]

Wow, thanks for the quick detailed reply. I'm going to grab some contact cleaner and give that a go first. I'm thinking the ignition switch is the likely culprit. Even in the heated garage it does the same thing 100% of the time.
Is there any benefit to reworking the pgmfi relay? I can bring it into work and resolder/conformal coat it to OEM specs.
I will toss it back in there for now and give the ignition switch a go.
Thanks again

 

[Old Guy]

Na1, so its at least 21 years old. If the relay is already out of the car and the relay is original, reflowing the solder joints would be a good idea.

If you want to go with refurbishing the existing ignition switch, consider heading over to Home Depot or Lowes (or who ever) and picking up a small tube of Oxgard grease. Smear a little bit of the grease on the contacts when you have the switch apart because Oxgard enhances the conductivity of the contacts. One of the risks with refurbishing the ignition switch is that the operation of the contacts relies on some spring pressure applied to the contacts. That pressure commonly comes from simple cantilever type springs and after 20+ years of operation the spring action suffers from fatigue. Sometimes a little bending can restore the contact pressure. sometimes it doesn't work so well so it can be a crap shoot.

I am an 'old guy' and my neck muscles kind of knot up just at the thought of doing the deep dive under the dash. As such, I like to plan on the basis of minimizing the risk of having to go back under there, hence my preference for new.

 

[drew]

There is also a plating on the contact pads that wears away. Cleaning the switch is usually a stopgap while awaiting new parts.

 

[JDMB20TDA]

<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190505_101345_zpsmercc2ar.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190505_101345_zpsmercc2ar.jpg" border="0" alt=" photo IMG_20190505_101345_zpsmercc2ar.jpg"></a>
<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190505_101332_zpslts1va1h.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190505_101332_zpslts1va1h.jpg" border="0" alt=" photo IMG_20190505_101332_zpslts1va1h.jpg"></a>

Sorry for the wait, car is ready to leave hibernation. Still having the same issue, ive pulled and inspected the pgmfi relay, ive tried holding the key forward to make better contact on the ignition switch. No luck, it looks like the car has an aftermarket alarm, there's a horn, momentary switch, and a led on the dash. Any idea of there is a way to override it or where the brain maybe? Ive tried holding the switch without any luck. Car starts runs, then dies. No matter if I give it gas or not. I can hear the relay click on and off as it dies. Any help would be appreciated.

 

[goldNSX]

Does the aftermarket 'alarm' also has a function to cut the fuel or are you able to start the car without any other things to do?

For peace of mind: the main relay is pretty cheap (well, in Japan). Or as [MENTION=26435]Old Guy[/MENTION] supposes to check the ignition switch. Both are prone to fail in any NSX. It's not worth investing hours and hours on these two item unless you know that they already have been changed. On the main relay, the cracks are thin as a hair, hard to detect. Sometimes the relay itself goes bad, no visible at all.

 

[Old Guy]

Are you still experiencing the same start and die sequence or have you moved on to a total no start sequence?

Non OEM parts installed by somebody else = lost in the forest without a map. Sorry, can't provide much advice there except that from my one brief fling with aftermarket alarms, the pushbutton is typically part of the sequence required to put the alarm system into programming mode. If you managed to get the alarm system into program mode and then did some subsequent actions, you might have programmed something interesting into the controller. The LED typically flashes a few times indicating that the system has entered programming mode.

Assuming that the alarm is armed and that it has a starter interlock, you should not be able to start the engine at all. If you can get the engine to start and then it dies, the aftermarket alarm is not likely causing your engine to quit (unless somebody got creatively weird with the wiring). There are a number of places that someone could stick the module; but, a logical place to look would be in the area under the glove box. That is where the OEM keyless entry goes when it is installed. The location is logical because it is close to the power door lock module and security module which it has to interface with. It also has the great advantage that it is easy to check because I think there are only 3 screws holding the access panel on.

If your engine still suffering the start and then die problem, does the engine die almost immediately when you release the key from the start position to the run position or does it die a couple of seconds after you release from the start to the run position? If it dies almost immediately after releasing to the run position then that is almost certainly a problem with the ignition switch. If the engine dies a couple of seconds later then that is a different and more confounding problem. As a related note, holding the key forward or pushing on it will likely have no effect on switch operation. In fact, pushing forward on the key might makes things worse by releasing pressure on the contacts. Rotating the key back and forth slightly can sometimes help because it wipes the contacts back and forth which sometimes restores the connection. If it does, it is a temporary fix at best.

 

[Old Guy]

On the main relay, the cracks are thin as a hair, hard to detect. Sometimes the relay itself goes bad, no visible at all.

Indeed. You will typically need a very strong magnifying glass or industrial inspection microscope to see the fractures.

The main FI relay is available from Rock Auto for less than $50 for the 1994+ cars. If you have an earlier car the later relay works just fine and fits just fine if you swap the old case and bracket onto the new relay (i.e. transplant the guts). Or you can spend $100+ to get the pre 1994 relay! However, given the description of the start and quit failure I would be inclined to put money on the ignition switch which is also available from Rock Auto for around $50. At this point, $50 for the ignition switch versus $50 for the main relay seems like the better choice, or 'go big' and do both.

 

[drew]

To kill the aftermarket alarm, trace back the switch or the led to the alarm control unit. Either unplug the main harness or pull the fuse. Nearly all alarms are NC and everything should work as if the alarm is not present.

Usually the alarm unit is tucked away under the driver side dash or kick panel....so it should be be easy to find: just look for the zip ties and mess of wiring that doesn't belong.

 

[JDMB20TDA]

Awesome thanks guys, yea this has gone on too long. I will do a quick search of the dash to see if there is indeed a brain anywhere I just figured it was related. I will however order a new relay. I'd like to have one handy in case its not the issue should I need one down the road.
As for the ignition switch I should have time to tear into it next week and inspect it for wear/clean it. Either way if that's the cause a new one is in order.
As for the updated relay, is it common for people to swap out the internals with the earlier case? Never heard of this but it takes less time to open it up than the ~45$ I'll save not at work. Thanks again and I'll update as I have time to look into the issue.

 

[Old Guy]

As for the updated relay, is it common for people to swap out the internals with the earlier case? Never heard of this but it takes less time to open it up than the ~45$ I'll save not at work. Thanks again and I'll update as I have time to look into the issue.

The relay mechanism didn't change, although Honda re numbered the pins in the service manual just to cause some confusion. Honda changed the location of the main relay which required a different bracket with a slightly different moulded clip on the case. The later case might fit on the original bracket; but, the guaranteed solution is to transplant the old case on to the new guts for a guaranteed exact fit. This thread discusses it.

http://www.nsxprime.com/forum/showthread.php/182760-Main-Relay-differences?highlight=main+relay

Larry B confirmed that the internals are identical and Regaz reaffirmed that the swap works. The more interesting question might be why the '94+ relay is more than twice the price of the pre '94 relay.

 

[Kaz-kzukNA1]

Don't know how your aftermarket security devices are wired so will leave them.
Though not nice if you can crank or even start the engine for a short period.

Re-solder the Main relay.
It's tiny hairline crack so just re-solder all of the joints.
Once new one has arrived, keep the re-solder one on the car as an emergency spare for yourself as well as for another owner.

IGSW loom as per above recommendation from others.

Most likely to be Main relay or IGSW related but check the state of the fuel pump resistor cable exit.
If wires are damaged, you can still crank and start the Eng but immediately afterwards, the Eng will either stall or run erratically depending on how far the wires are cut/damaged.

DO NOT straighten it 'yet' if already sharply bent.
If straightening, first, either release the tiny screw attaching the cable bracket to the resistor heatsink or loosen the cramp section slightly to prevent the wire from being pulled.

http://www.nsxcb.co.uk/entry.php?2057-Fuel-Pump-Resistor-01

http://www.nsxcb.co.uk/entry.php?2058-Fuel-Pump-Resistor-02

http://www.nsxcb.co.uk/entry.php?2361-Eng-Refresh-Cooling-Sys-etc-01


As a quick check, bypass the resistor as per above photo.
You can use paper clip as a quick check but pay extra attention not to damage the mating female terminals.
Ideally, you should use 090 terminals.
Do not let it touch other metal objects.
It's live circuit.


Kaz

 

[JDMB20TDA]

Alright, relays in the mail. I'll re sweat the solder joints as a back up like suggested. I'm out of town this weekend when the relay is scheduled to arrive but, I'll report back with the status.
I will also look into that fuel resistor. I hadn't found any info on that while searching I appreciate any help or direction you can offer.
Thanks again

 

[JDMB20TDA]

Back in town, threw the new relay in and the exact same issue, starts revs up like a normal cold start then, shuts right back down. Its definitely getting fuel and spark as it runs like normal at first.
<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190515_210340_zpskarsljqr.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190515_210340_zpskarsljqr.jpg" border="0" alt=" photo IMG_20190515_210340_zpskarsljqr.jpg"></a>
The resistor and harness all look to be in good shape. Doesnt seem like its kinked or moved at all.
I'm currently uploading a short video of the start/die condition.

 

[JDMB20TDA]

<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190515_212504_zpslvy6ad0k.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190515_212504_zpslvy6ad0k.jpg" border="0" alt=" photo IMG_20190515_212504_zpslvy6ad0k.jpg"></a>
Well it runs! So the resistor is indeed the culprit. I bypassed it and she runs. I didn't keep it on out of fear it'd cause harm to the pump at full voltage/current. Is there any way to rebuild the unit? Actually seems to be indecent shape compared to others you had posted Kaz. I can resolder/crimp a new set of wires to the relay if that's all the issue is.
Thanks again, you guys are super helpful. I was discourage after finding old alarm garbage and the relay not being the fix. A spare is not gunna hurt. She's been running for about a half hour now, no hickups. She does smell pretty rich, not too worried as long as it won't cause harm until the new unit arrives... If I can find one. Looks like Acura Part No.: 16717-PR7-A01
Resistor Assy., Fuel Pump (196171-008) (Denso) is discontinued. Can't wait to hear back just happy she's running!

a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190515_215731_zps8vcgh4lv.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190515_215731_zps8vcgh4lv.jpg" border="0" alt=" photo IMG_20190515_215731_zps8vcgh4lv.jpg"></a>
<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190515_215211_zps5dtjzhns.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190515_215211_zps5dtjzhns.jpg" border="0" alt=" photo IMG_20190515_215211_zps5dtjzhns.jpg"></a>

 

[Old Guy]

Measure the resistance of the resistor and compare it to the value in the service manual. If its out of spec then you need to replace the resistor. I am surprised that it is delisted as another Prime member sourced a new one last year. Try checking for later year cars as the resistor was used right through to 2005; but, the value of the resistance might have changed which may have resulted in a new part number. A slight change in the value of the resistance will not have a material effect on operation.

If the resistor is in spec, then check the plug terminals and the wiring. The problem may not actually be in the resistor; but, rather heat damaged wiring leading to a bad connection.

 

[goldNSX]

^^
plus check the operation temperature (right after it stalled). I recall one case where it got extremly hot and I'd be interested in the F or C range it reaches.

16717-PR7-A01 is available in Japan. $50 something.

 

[Kaz-kzukNA1]

From what I can see in your photo, I think your issue is more likely to be the eng loom side and not the resistor.
Also, unless someone wiped off the area, there is distinctive colour/dirt difference between the male/female connector so there is a chance that someone already replaced the resistor block in the past.

Another owner had both issues (broken wire exit at the fuel pump resistor + damaged WPC 090 female terminal) at the same time so it is still possible that you may have issue on the resistor side as well but I doubt.

Before placing the order for the replacement resistor block, measure the resistance of the existing one.
It depends on the temperature, quality/spec of the test lead/multimeter, etc so can vary easily in the order of 0.1ohm.
As long as it's not short circuit or well over the written figure on the metal case body, can be re-used.

Still, unless you already have all the materials and tools, it would be cheaper to get new one as a replacement for potential future issue so may be still good idea placing the order any way after replacing the connector.
As goldNSX mentioned, you can still get hold of -A01 in Japan.
Contact Marc at theNSXshop.com.


You must replace that 2pin connector on the eng loom.
It's already damaged.

You can see the white retainer being chewed off at the red arrow.
This will happen if the male terminal was not aligned properly before locking the two connectors.
Most likely, black seal inside the connector got swelled due to chemical reaction and someone forced the connection.

Also, it could be reflection or camera angle but at the yellow arrow, looked as if the presence of patina corrosion.
Proof of moisture.
Remove white retainer and look inside the female terminal.
You may find that one of them lost the part of the metal spring loaded rail/body.

For the replacement, just get in touch with Joe at cycleterminal.com.
He is on Prime and NSX owner.
Sumitomo HW WPC 090 2pin Female.

Hope you have the proper crimping tool. If not, just ask someone with electronics skill.
There is a black cable clamp attached to the side frame of the eng bay.
By replacing the terminal/connector, the free play of the cable will be reduced but you can easily re-adjust it by re-positionling the clamp with the tape.


As a side note, be careful with the fuel hose/pipe and filter alignment.
Your photo shows that the filter is misaligned and most likely, kinked the hose/pipe alignment


Kaz

 

[Jinks]

I think I might have a used fuel pump resistor which was removed as working after going full stand alone. Need to check my used parts bucket.

 

[JDMB20TDA]

Well the resistor seems to be in good shape.
I'm getting ~.55 ohms or resistance.
<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190516_182332_zpsdavrrmro.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190516_182332_zpsdavrrmro.jpg" border="0" alt=" photo IMG_20190516_182332_zpsdavrrmro.jpg"></a>
7.9x.55= 4.345v
I loaded it up and the resistance seemed steady and produced heat like it should.
I'm going to take a better look at the female connector tonight, maybe try and clean it out a bit.
<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190516_182143_zpspzipdhlo.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190516_182143_zpspzipdhlo.jpg" border="0" alt=" photo IMG_20190516_182143_zpspzipdhlo.jpg"></a>
The wires do seem a little stiff but, it seems to check out and pass the tests.
Is there any test I could perform on the otherwise of the resistor?
And the fuel pump had been replaced a while back. I know this doesn't mean much but there could be a reason for that.

 

[JDMB20TDA]

Hopefully more good news, this may have been the cause as long as I didnt push it out with the jumper.
<a href="http://s31.photobucket.com/user/JDMB20TDA/media/IMG_20190516_220640_zpssc4tejgb.jpg.html" target="_blank"><img src="http://i31.photobucket.com/albums/c387/JDMB20TDA/IMG_20190516_220640_zpssc4tejgb.jpg" border="0" alt=" photo IMG_20190516_220640_zpssc4tejgb.jpg"></a>
So this pin is just completely loose and won't lock.
Is there anyway to fix this without ghetto rigging it? I figure some glue while I wait for the new connector may work.
While it hasn't been too much fun, its been a learning process.
When I order the new connector, should I solder it in with some strain relief to prevent this from happening in the future?

 

[drew]

That connector shell is used all over in Honda engine compartments. Easy to source at a pick a part junk yard.

 

[goldNSX]

And the fuel pump had been replaced a while back. I know this doesn't mean much but there could be a reason for that.

Here might be one of the keys to the problem. OEM or aftermarket pump? A special feature of the OEM fuel pump is it high resistance or low current draw and the resistor has been designed for that.

 

[JDMB20TDA]

It was an OEM unit done by acura.