z-ya

If you don't want to experiment with the PWM settings, you can turn PWM off and put series resistors. I either use a 280Z resistors pack (mounted next to clutch master cylinder) or 5W 5 Ohm power resistors. I put them in all installs because eliminates the possiblity of damaging the injector drivers if you use a V2.2 board.

Yes, how can the injectors be grounded to the chassis? The injectors should have +12V on one wire (with a series resistor if you want to turn off PWM), and the other wire goes to the MS which provides gorund.

What other grounds do you have? The MS needs at least 2 16 gauge ground wires to the chassis. The EDIS module should have a 14 gauge wire to ground. Do you have a 4 gauge ground wire from the engine to the chassis? Attach the MS ground wires where the engine block is grounded. I have the battery in the back of my race car, so that isn't the problem.

As Paul and myself have done, I would run a Chrysler coil pack. Paul can give you a better idea of what year and model. The nice part about the Chrysler coil packs is that they use standard HEI connectors. This gives you a lot more choices for spark plug wires, especially custom sets.

You don't need a new alternator, you need to filter the noise that the alternator is creating. Chances are, a new GM alternator may exhibit the same problem. The problem is in the circuit the alternator is connected to. For some reason there is a resonant frequency. Try a capacity at the alternator. Let us know what configuration fixes the problem.

I don't think I got your email. resend them to: psanders240@adelphia.net

When it exhibits the problem, the waveform going from the MS to the MSD looks the same as when the engine is running OK? Why don't you use and HEI module or the VB921 to eliminate the MSD as a possible source of the problem? I have a feeling it is not the MSD, since you have tried two now. The MSD just fires the coil (yea, multiple times during that firing cycle) when the MS tells it to. It has no idistributor shaft rotation information. I think there is something with the firmware, or how the distributor pump shaft is timed.

You need to run EasyTherm to calibrate any OEM sensor that is not GM. You can find out how on the MSEFI.com website. A 10uF 20V capacitor should work fine. The EDIS module will fire any plug with the proper gap and heat range for the application.

Yes, the capacitor should connect from teh +12v at the coil pack, to a short wire to chassis ground. This acts as a low frequency filter. It's really there to surpress radio noise, but it can't hurt. When you disconnect to alternator (remove belt) to test to see if it is the source of interference, make sure you put a battery charger on the battery so you have at least 13V in the system. For plugs, I would use an NGK "6" heat range, gapped to 0.035". This should work find with the EDIS coil pack. For the track, you may want to go to a "7" heat range.

Pin 7 at the EDIS module should connect to shielding. See below:

Th VR sensor needs to be shielded, and the shield should only tie to the shield wire on the EDIS module. It should not be connected to the engine block or you could have ground loop problems. Same goes for the TPS. Both the +5V and TPS signal wires should be shielded with the shield tied to ground only at the Megasquirt. Do not tie it to ground at the engine block. Sometimes I will twist the 5V, TPS signal, and TPS GND together with an electric drill when I am making the harness. Sometimes this is good enough. Sometimes there is a lot of noise running through the harness or a round the engine compartment that requires a real shield.

200sxBoY: I just took a TB off a 240SX yesterday, and the signal wire from the TPS WAS sheilded from the factory by Nissan. This is not the first sheilded TPS harness I've seen BTW. I've been an EE for 20+ years, and I know if you run a sensitive analog signal near 40,000V high tension leads, you will have interference problems. I've also had this same problem with installations of Megasquirts and other engine management systems where sheilding the TPS signal fixed the problem. I don't want to make this into an argument, I'm just try to help drews240 get his problem fixed. Are you involved with this project?

OK, if it is sheilded, then that is not the problem. You are running EDIS? I would mount the trigger wheel to the solid part of the crank pully (not rubber isolated). Just in case at high rpms the damped pulley distorts, it could effect the gap between the sensor and the trigger wheel. I dought this is the problem as BRAAP has his mounted to the damped pulley and hasn't had problems. You have no more than 1mm between the VR sensor and the trigger wheel, right? Yes, I can check your settings, send me a PM.

Did you put a scope on the TPS signal wire? Again, is this wire sheilded from interference? I don't think it has anything to do with ignition except for the fact that the ignition energy is the agressor, and the TPS signal is the victim. When you have 40,000V+ noise running around sensitive 5V analog signals, the smaller of the two signals is bound to be interfered with.

Seems like a lot of work to end up with something that is not easily tunable. Plus, it's 10+ year old wiring, connectors, and sensors. Doing a Megasquirt conversion seems like it might be less work, especially when you get to the debug phase. As far as making it work, I don't see why a Maxima ECU wouldn't work on an L28, as people have used Z31 ECUs on them for years.

Two ways to add ignition: 1) Use the 82' or 83' distributor and oil pump shaft. This is probably the easiest route. You don't need a new coil or an MSD box. The MS V3 mainboard has a VB921 coil driver built in, which produces a great spark with the stock coil. 2) Add Ford EDIS ignition. Raid an Aerostar or Exploder in the junkyard for the module, trigger wheel, and VR sensor (get the connector pigtails too). Some machining and fabrication is required, but this option eliminates the distributor for more accurate timing. It also allows much more timing adjustment range than the distributor setup. I'm using this setup on my supercharged L28 240Z with great sucess. The easiest is numebr 1, the best is number 2.

The only SU turbo setup I've seen is a single SU. I had one, and ran two float bowls to prevent starvation during hard cornering. It was a draw through setup:

If you look at the bottom most plot, the blue trace (TPSacc), you will see three glitches in the higher RPMs. Am I reading this wrong? It looks to me like glitches on the TPS signal. This is very common if the TPS signal is routed near spark plug wires and or other EMI sources.

trippintl0, I'm not using an HEI module, I am using the VB921 in the Megasquirt to drive the coil directly. You should also have the spark output set to inverted even if you are using the HEI module. Tony, I did have a spark plug connected when I was testing the VB921. I adjusted the dwell until the spark stopped getting brighter with no misfires, and then backed it off just a hair. The plug actually get's hot if you let it run long enough. I let it run at 4000RPM for about 2 hours, and the plug was too hot too touch!

Why do you have the dwell dettings to a fixed duty cycle of 50%? Also, why is your trim angle ony 50, shouldn'tit be closer to 70?

OK, found the problem when I put a scope on the signal (LED14) that is driving the VB921. The signal is inverted from what it should be, so the VB921 is sourcing current at a very high duty cycle through the coil. Yep, that would cause it to get hot. You MUST invert the spark output in Megatune! Go to "Spark => Spark Settings", then change the "Spark Output Inverted" to "yes" Now the VB921 only gets hot if I run the stim at 7000RPM for an extended period. For race applications where the RPMs will be higher for longer periods, an additional heat sink should be fitted to the VB921. For street and drag racers, the standard V3 heat sink is fine. For the dwell settings, I changed the crank dwell to 4mS, and the running dwell to 3mS. Any more than 3mS does not appear to make the spark any hotter (makes the VB921 hot though!). I used a 280ZX coil for this test.

Anyone resolve the VB921 overheating problem? I'm testing a customers unit using a stimulator, and a ZX coil. The VB921 will overheat in a matter of 30s or so. I tried adding a ballast resistor, but that will just make it spark a little bit longer. I'm also having reaet problems when the VB921 is connected to the coil. I post any fixes I figure out or find.

So, looking at the second plot from the top, the RPM is yellow, MAP is blue. What is red, injector pulse width? If so, the PW follows the MAP. This looks OK. On the bottom plot, why is the TPSacc glitching? This would absolutely cause the engine to buck. From what I can see, the problem is not the MAP sensor, but the TPS or the wiring to the TPS. Are the TPS wires shielded? At 5000RPM, there may be noise coupling into the TPS circuit. Get some two wire sheilded cable, and run the +5 and TPS signal on the inner two wires, and connect the sheild to ground back at the MS. Do not ground the sheild at the TPS. Another thing to check is the TPS. With an ohm meter connected between the middle terminal and either outer terminal move the throttle plate. You should see a very smooth change in resistance. If the resistance jumps at all, this is your problem. Try cleaning it with some TV tuner spray, or just buy a new one. Let us know how you make out.

I would be concerned about the oil pickup under hard cornering (oil starvation), and the transmission being properly lubricated a the new angle. Making the down pipe smaller is a lot easier than changing the engine angle. Anything larger than 3", even on a high horsepower turbo motor is very close the point of diminishing returns anyway. IOW, you probably won't gain much power with a 4" downpipe over a 3" one.

I have mounted ECUs on the tunnel between the seats, and have had no problems with MAP responsivness. If you are running a radical cam, you need to switch to Alpha-N mode. With a radical cam, you have verry little pressure range, so you don't have much control over fuel using the VE table and MAP sensor.