Intermittent loss of power/engine shut down - Help requested

[NSaneXr]

Hey Primers. I'm looking for some advice/help on a problem I'm having with my car. The problem is sometimes the car cuts out when I hit a bump. Sometimes I get a ABS/TCS warning (both lights on), sometimes that is associated with a loss of electrical power (no radio) and sometimes a full loss of electrical power and engine cut off. When the ABS/TCS lights turn on it is just a blip; I see the lights and they go out. Sometimes when I see the ABS/TCS lights the radio goes dead and that lasts for 1-2 seconds. One time I did get a full loss of electrical and the engine shut down for 2 seconds and then everything came back. It seems to happen when I hit a bump however, there must be some unknown specific circumstance as it doesn't happen all the time.

I have tried to pull the codes but there are none (using code reader as well as using the service connector)

For some back ground information the car is a 97, 6spd with just under 100K Kms. The only mod to the electronics is a USA Spec adapter. No engine mods just comptech headers and AP-X exhaust. The only other major change is the EPS rack was swapped for a manual rack. Prior to this issue, last year I replaced the starter (due to a click no start issue) and the battery (due to age). After those replacements the car ran fine until the beginning of this year.

It was doing this a few months ago and after researching on Prime I concluded it was either the ignition switch or the main relay. I replaced the main relay and ignition switch. After that the problem went away and the car worked great for the last two months (drove it almost every day to work).

The only "change" that happened right before the problem started again was I got on the throttle a little hard and then the alternator started to make a whirring noise and a relay on the driver's side was clicking rapidly when I pressed on the throttle (at any speed/rpm). I removed the alternator belt to confirm that was the problem and put the belt back on and the whirring went away and the relay stopped clicking. I tested the alternator output and it is still 14V or so.

At this point I have checked the fuse boxes for loose or damaged fuses/wiring and specifically those mentioned to be possible culprits for this issue (fuse 4 and 12). The battery connections are good and solid. I ran the engine and wiggled the wires leading to the ignition switch, main relay, other relays, fuse boxes and anything else that I could think of and wasn't able to reproduce the issue.

I'm not sure if there is something else that I haven't considered. Any suggestions on what to look at next would be greatly appreciated.

Thanks,

-e

 

[shawn110975]

you can remove the 10mm bolts on all the grounds sand them and re-install.

there are multiple grounds thru out the car the ones that I would look at are the one in the trunk bolted to the driverside wall behind the carpet,

then the 2 in the engine bay one from the tranny to chassis and the one from the valve cover to the chassis.

then the 2 in the front of the car they are both mounted on each side of the front rails.

sand them till they are bright and shiny and re-install.

better grounds are always a good thing.

 

[NSaneXr]

Thanks for the suggestion. I cleaned the ones that I hadn't done yet just in case. They didn't look bad but since it didn't cost anything I cleaned as many as I could find - even the smaller ones. I also checked the fuel pump resistor for bad connections but it seemed fine as well.

-e

 

[jwmelvin]

It seems significant that the power loss is broader than just the engine or one isolated system. That suggests an issue like the ignition switch. I know you replaced that but have you studied the wiring diagram to look for other key pathways? If the radio goes out then it would seem to be something other than the supply to the engine compartment fuse box.

 

[Old Guy]

Do the ABS and TCS come on after you have had a temporary loss of power and power is restored? If so, it might be that the ABS / TCS is reinitializing just like it does when doing a first start-up. If that is the case, at the risk of stating the obvious, you are likely getting a momentary loss of 12 VDC from a bad connection which is causing the engine shut down and all the other related issues. If that is the case, it sounds like you have already addressed the obvious causes. Unfortunately, you may have to wait for failure to become permanent.

If the ABS / TCS lights come on before you have the loss of power, then something else is at play.

 

[NSaneXr]

Do the ABS and TCS come on after you have had a temporary loss of power and power is restored? If so, it might be that the ABS / TCS is reinitializing just like it does when doing a first start-up.

I believe this might be the case. I'm thinking that in most cases the interruption is so quick that I'm not seeing the effects. Just that one time as it was at night I could see that all the electrical components went out. I'll look at the ignition switch again. With the old part it had a very similar effect on the car when it failed. The slightest tap on the bottom of the key would cause a momentary break in the contacts. The new part doesn't do that and I did re-check the connections and wiring to it.

It seems logical that the issue is somewhere on the key pathway. If you had to guess, would you say I was looking for a short in the system or a break.

Thanks for the replies.

 

[jwmelvin]

I would think that a short of any significance would cause a fuse to blow so I suspect an intermittent open/break.

 

[Old Guy]

I agree with jwmelvin. If the short circuit was good enough to cause the engine to momentarily stop and the lights to go out, it would normally blow a fuse. I think you are looking for a connection that is opening up under vibration.

You already tested the condition of the wiring on the back of the ignition switch which is one of the 'next' things on my list. Assuming the ign switch wiring is OK, there are a lot of the NSX systems that get power directly from the engine compartment fuse box and the big under hood fuse / relay box. These boxes are powered up continuously off of the battery. A lot of the devices connected to the fuse boxes are connected though relays which get closed when you turn the ignition switch on. Honda probably does this to reduce the amount of current flowing through the ignition switch.

What you want to be doing is checking the condition of the connections to these fuse boxes. The first ones to check would be the main connections back to the battery. After that it gets more complicated. Pulling the fuse boxes to check the individual fuse connections and the plug connections is messy and you do risk doing damage to the wiring. I would get the service manual which lists what system is supplied from each fuse. Make a list of all the things that went dead in the car when the problem occurred, then see if you can group those items around one or two connectors coming out of the fuse boxes and then focus on checking the wiring associated with those connectors. If you can't group the failure's, I might be inclined to leave the fuse boxes alone after checking the main battery supply just because there is an opportunity for screwing things up. Come back to the fuse boxes after exhausting other (unknown) possibilities. Any 'probing' of these two fuse boxes should be done with the battery disconnected. Accidently shorting the main connection back to the battery will have lots of potential for damage.

Finally, you checked the battery connections for integrity; but, to pick up on Shawn's point, did you check the negative cable where it connects to the body?

My next suggestion get more complicated because it involves using an Arduino with a data logging piggy back to record voltages. I had to use this on another car to track down a voltage spike which was causing the ECU to reset on me.

 

[NSaneXr]

Thanks again. I was talking through the issue with somebody here at work and realized that the events seem to happen only after the car has been run for a longer period of time (about an hour or more). I was wondering if any of those relays or switches that were mentioned might be more sensitive to heat than others. Mind you the car worked perfectly fine throughout July and August and we had some pretty warm weather here.

I did check and clean the negative cable that runs from the battery to the body. I also cleaned the ones that Shawn mentioned as well last night but unfortunately I experienced the problem again today just before getting to work. I will look more closely at the fuse boxes and fuses in those two areas focusing on the main connections at first.

 

[Old Guy]

I am not aware of any particular relay that is more sensitive to heat; that said, heat will cause expansion which can lead to the permanent or intermittent failure of a compromised electrical connection. If you think that the failure is always preceded by high operating temperatures, the engine compartment fuse box might be the preferred place to start looking.

If you get desperate:
If you know somebody with some electronics and code skills, it can be useful to set up an arduino so that it monitors the voltage and can 'trap' an event where the voltage drops below 2 volts (or whatever you want to set the threshold to). You connect the arduino to various points in the car's 12 volt system, turn the arduino on and go for a drive and wait for the lights to go out. You then check the arduino to see if it trapped a voltage drop. If it did trap a voltage drop, then you know the problem is upstream of your first test point and you relocate the arduino connection to a logical upstream point and start the process again. If the arduino did not trap a voltage excursion when the lights went out, then you know that the problem is downstream of your test point and you move your arduino to some logical downstream point and again start the process. The connections at the ignition switch would be a logical point to start this process. The arduino analog inputs are limited to 5 v so you need a simple resistive voltage divider on the input to the arduino to scale the 14.3 car voltage to something acceptable to the arduino. I also put a 5 v zener across the input of the arduino to make sure that the arduino does not get hit by nasty voltage excursions. If you electronics whiz is good, he can set the arduino up so that a LED would light up when it trapped a voltage drop. Save you having to connect up a laptop to query the arduino to find out what is going on.

The business with the arduino is a royal pain because it involves connecting the arduino to various test points which are not always accessible or easy to make connection to. However, when you have an intermittent fault, it is ultimately the only systematic way of tracking down a power supply problem unless you go for the preemptive complete replacement route. There are probably other devices that you could use to do this, it just happens that I am familiar with the arduino because my engineering school son had one sitting around.

You might want to consider a drive down some really rough roads to see if you can make the power failure permanent :smile: . Much easier to find and fix a permanent failure.

 

[NSaneXr]

The issue didn't occur on the way home even though there were some bumpy stretches during the drive. I'm not sure I'm at the arduino stage yet but I will look into it. Oddly, the issue did occur today - radio cut out, relays on the rear wall clicked, dash lights came on but the engine didn't seem to die this time. It was like the car cycled the ignition sequence again (minus the starter) as Old Guy suggested earlier. Unfortunately, it only did it once even after subsequent bumps in the road. I will check the manual when I get home to verify, but is there a relay or something that temporarily cuts the power to the accessories during the "start" stage of turning over the engine or is that just done by the ignition switch?

 

[NSaneXr]

I am still experiencing the issue. Its not heat related as it happened once or twice now at the beginning of a trip.

After a whole day of checking connectors I found one rear wheel sensor that disconnected very easily and one bad wire on an O2 extension (wire just fell out of the connector). I was hoping by some miracle that was the problem but it wasn't. I drove the car around every bumpy road around my house and right before returning home it happened.

I don't have access to an arduino so I thought perhaps I could use the min-max record option on my multimeter. I did setup my multimeter to record the event when the issue happened it recorded a voltage drop. Unfortunately, I wasn't really thinking clearly as I had connected the meter to the rear fuse box positive jumper location and the negative to the ground. I'm not sure what that tells me, but this could take a long long time to test all the various systems.