Weird hesitation after header installation

Hi all.

Prior to my header and test pipe install the car ran without issues. Here is what I changed/installed:

1. Aftermarket headers and test pipes, bolted straight in without a hitch. I had to extended the cables on the 02 sensors, it s a 91 so only had 2. I soldered the same gauge of silver coated, heat resistant wire and am proficient with a soldering iron and did a nice clean job, checked the resistance of the cable and it matched an original cable.

2. Painted the brake callipers
3. Removed the AC pump, left the cable unplugged and capped to prevent dust.
4. Installed a 12 volt relay connected to the reverse switch to operate the reverse lights as this was an auto to manual conversion and I never got around to doing this. Nice and clean, running straight off the 12 volt live connection on the fuse box.

The hesitation only happens once the car has warmed up to operating temperature. No CEL codes are reported when it hesitates. The hesitation only happens below 3000rpm, above that it is fine. Happens in all gears. Engine isn't missing, rather just feels boggy. Prior to operating temp the engine will rev to redline fine, vtec kicks in aswell.

Coolant is topped up
Replaced spark plugs
replaced coils
replaced ignitor

I have waited for the car to fail, then pulled over and disconnected both o2 sensors. The CEL light comes on immediately but it continues to hesitate so this tells me that the o2 sensors are not the issue nor is the cable extension.

Fuel pressure is fine

When it is hesitating, if I pull over and let it idle, the idle will be normal, a bit below 1000 and then it suddenly raises to 1500 rpm on its own for a split second like a quick rev. The car does this itself. then its fine again for a few mins and then starts to hesitate again

I have a spare ECU, installed that, same issue

I have drained the tank and refilled with 98 octane, same issue.

I have gone over all the connectors that could have been disturbed when I installed the headers, they seem fine, connect properly etc

Question. When I remove the cats and installed the test pipes I just disconnected the sensors from the cats inside the boot and rubber bunged the holes. I assumed these are just blocked cat sensors and wont have any effect, is this correct?

Any help is appreciated

When you checked the resistance of the cable on the O2 sensor, by any chance did you end up trying to measure the resistance of the sensing element? If your sensor uses a zirconia element, the voltage in the ohmmeter battery may have damaged the sensor. If the sensor is Titania, then you may be OK. The only way to test the operation of the O2 sensor is with an oscilloscope connected to the sensor while the car is running. NGK suggests that you will see the voltage swing back and forth from .1 to 1 volt in about 0.3 seconds . With a .3 seconds period you will never see that swing using a conventional meter which is why you need the scope. I have never done this on my NSX; but, I did do this test on an old turbo 745 Volvo with a Bosch sensor and Bosch LH fuel injection. In that case the voltage swings were much slower than 0.3 seconds. If you don't have access to a scope, I think you are in to the test by replacement mode, which is not very satisfying.

My recollection is that as the O2 sensor ages, it may fail in a mode where it outputs a voltage of 1 v which says that all is good (AFR wise). The ECU may not be able to detect this failure mode and as such may not generate a code. In such case the 02 sensor does not generate a changing signal for mixture correction.

You noted that if you disconnect the 02 sensors you generate a code. This does not surprise me. It should generate a heater failure code for sure and would surely generate a sensor failure code due a complete absence of signal. However, as noted previously, that may not necessarily mean that the 02 sensor is operational.

The header should require a mixture change if it is at all effective at improving exhaust scavenging. If your O2 sensors are 'stuck' they may not be providing the required feedback to change mixture strength which is why this problem manifested itself after adding the header (or you damaged the sensors as previously noted). Just speculating at this point.

My knowledge of O2 sensors and their operating chemistry is pretty minimal and a such I would defer to anybody with more direct knowledge or experience.

thanks for the detailed reply. Good information.

I finally figured it out. The rear banks o2 sensor was just slightly touching the heat shield below the boot compartment and had worn the insulator enough to earth out once the engine was at operating temp assuming the expansion was enough force to make contact. I removed the o2 sensor and applied heat shrink to the worn wire. Then I bent the shield slightly to allow room for the o2 sensor.

running great again. Finally get to appreciate the added power, response and sound the headers and test pipes give.

cheers