I was wondering if anyone here has retrofitted their broncos with a knock sensor, especially those running a few psi of boost, if so I was wondering what the best mounting location is on a 302 block? Not too many threaded holes in that block.

 

Depending upon the vintage, the 302 HAS a knock sensor mounted in the block between the back of the lower intake manifold and the transmission bell housing. ALL of the EEC-IV controlled 302's have one.

 

Where exactly is it? And can you tap it electrically, and is it electrically equivilant to a GM sensor? I have never seen knock sensor data in any EECIV based diagnostics. I am trying to add the J&S Knock Retard box to the truck, and can't find a suitable mounting location for the supplied (GM) knock sensor. This is an EEC-IV (94 302 engine). Also how high up is it, above the exhaust headers or below, in relation to heads, etc? Driver/passenger side? How far back (approx cylinder # location)? Color wiring going to it?
Thanks
brian

 

Its dead center on TOP but behind the lower intake manifold nestled in the small expanse of steel between the back of the manifold and where the tranny bellhousing bolts to the engine. If you follow the Thermactor (air injection) plumbing back through the diverter valve (pass. side) and behind the engine to the crossover tube between the cylinder heads, you will find it BELOW the crossover tube and just slightly behind it. Its function is based on seismic (vibrations) causing a fluctuation in reference voltage (5VDC) back to the EEC-IV computer. A change in reference voltage "tells" the EC-IV computer to retard the timing. When the vibrations subside, the flutuation in the reference voltage stops and the timing is returned to normal. All knock sensors work this way, the reference voltage could be different but won't matter to the sensor.

 

Ok, now if I have a 1994 bronco XLT EEC-IV (NOT OBDII), can I be SURE it has a knock sensor. The add on knock detector/retarder I am using looks for a signal from the GM sensor, do we know who makes the ford knock sensor, I am led to believe there are two voltage output ranges for knock sensors, one is based on bosch (25mV/G), the other is based on GM (500mV/G). Without removing the upper manifold or going into the ECU wiring, where is the best place to tap the wire, I believe it is pin 23 of the ECU (red/yellow)? Has anyone here ever used one of these on a bronco (www.jandssafeguard.com)? Most of the info on these out there pertains to mustang engines. The main reason I am using this unit is that it offers a wide retard range, and most importantly will only retard the cylinder that is knocking, as opposed to all cylinders by syncing the knock detector to the incoming ignition pulse, to avoid retarding the entire engine if only one cylinder is on the threshold of knocking, and losing lots of power in the process. Also I am led to believe there are two versions of this EFI system where one uses a knock sensor and one does not, the Sequential injection system supposedly uses a Mass air sensor and a knock sensor, where the MPI multiport injection uses a MAP sensor, and no knock sensor. Are there any broncos in 94 that used MPI injection or are all of them SPI?
thanks
brian

 

Also is pin 46 of the ECU GY/R (signal return) connected to common electrical ground???
Thanks
Brian

 

Well, unless Ford somehow managed to figure a way around the knock sensor signal between 93 and 94 the knock sensor is there. My 92 and my 93 both have them and OBD-II didn't see initial application until sometime in 95 to the best of my knowledge.

The ECA utilizes pin 46 as a common "grounding" location however, it "reads" the VREF at pin 46 to determine the variance, if any, in the signal voltage. If the wiring to pin 46 is physically "grounded" BEFORE signal gets to the ECA, the ECA will assume the sensor(s) has gone wide open or is not present. My assumption would be that once inside the ECA, the voltage from pin 46 is optically isolated so that the ECA can determine variances in VREF without the concern that a physical ground may be weak. I don't have any thing to prove this theory but from the standpoint of reliability and electrical theory it makes the most sense.

The knock sensor is merely a semiconductive seismic response device. It is incapable of doing anything but reducing the voltage that passes through it. So the voltage supplied to it is what will come through it. The VREF from the Ford ECA is approximately 5VDC. The voltage that comes out of the KS will be the same until such time as it gets "knocked" so to speak. Then the voltage drops and the ECA "reads" the voltage change using the information to retard the timing as necessary. The difference between the GM and Bosch models would be the actual change in resistance value when the sensor is activated. In this case, either 25mV or 500mV which is certainly significant enough to make a difference. The device you are talking about will do essentially the same thing as the ECA would do and "read" the changes in the VREF signal created by activation of the sensor.

I would be interested in knowing just HOW the device manages to sync the KS to the cylinder in question since the KS is a reactionary device that responds to conditions created during engine operation. In other words it cannot initiate change until a knock is present and therefore the ONLY determining factor in WHEN the timing is retarded is the combination of the response time of the sensor and the time it takes for the ECA to process and compensate for the information received. Even the BEST electronics in the world cannot PREDICT engine knock. At best it would be the NEXT time that cylinder fired that the timing was retarded and thats really all the better its ever gonna get.

With the EEC-IV system about the ONLY margin for increasing the response time from the KS would be to eliminate or reduce the time it takes for the ECA to process the information. Any device that did this would almost need to send a piggyback signal directly to the TFI module to inform the module when to retard the timing rather than sending the information back through the ECA. Otherwise any lag in response created by the ECA processing the information from the KS would still be there. Maybe I'm overthinking this but we are talking about a ten-thousandth of a second here at the most. What exactly are you hoping to gain with this device?

For the record, the EEC-IV MPI used in the Bronco utilizes BOTH a knock sensor AND a MAP sensor to determine timing.

Mass air is the ONLY true SFI ever used in the Bronco and that didn't happen until 95 in CA and 96 everywhere else.

The EEC_IV "bank fires" the fuel injectors. In other words the left bank of injectors fire then the right and back and forth. Mass air fires each injector individually hence a true SFI (Sequential Fuel Injection). Now if the device you are talking about can override SFI triggering to the injectors as well as ignition timing the response time of the KS signal would be reduced even further but its STILL a reactionary change and cannot be predicted and the truth is the best case scenario is still going to be a correction AFTER the inital knock is detected. All that said, EEC-IV MPI (Multiport Fuel Injection) cannot individually trigger the injectors anyway so its a moot point.

 

Okie, first off this unit interfaces directly with the coil (or in my case an msd box) to control timing, second off you are correct it does not know that cylinder #7 is knocking, all it knows is that every 8 ignition pulses it hears a knock, so after it sees number 7, it will retard the next pulse in anticipation of the knock, and it will then advance timing back closer to normal until knock returns then it will retard to correct it until optimal retard values for each cylinder are determined, you are correct in saying that the corrections are relative and not absolute. The unit basically looks at the knock sensor as a "microphone" which essentially generates its own signal, and does not rely on a reference, by converting sound/vibration energy into electrical energy, it feeds this signal through an AGC, and conditions it to determine knock. I am almost positive my truck is mass air, as there is one fat duct going into the air box instead to two smaller ducts, and it has a sensor in line with the duct right near the air box. The reason I am adding this box is that I want an intelligent knock retard as opposed to running just a straight boost retard such as the MSD.

 

SFI-mass air broncos started in mid 94. I own one and have had nothing but problems with it lately nickle and diming me to death with only 80,000 miles.

 

Welp, mine has 160k on it and aside from a tranny at 95k, I havent had any other major problems with it, sure I put my fair share of auto locking hubs on it, and lots of body work mainly to the tailgate, but it sure still runs good, and has all the "problematic" hardware, including electric transfer case, auto hubs, mass air, etc.