This is for my Right Hand Drive 1982 GSL equivalent RX-7. Here is a map of my earthing circuit.



From Megasquirt DB37 there are 4x 10A earth wires with are bolted to the front of the passenger side strut tower. There are two bolts, and each wire goes to a different lug which has been crimped. The two lugs/bolt are spaced appart by a nut which has had the thread drilled out.

This is a list of the devices which use the Megasquirt ground pins as earths

Coolant Temp
Air Temp
Innovate LC-1 system ground

The thermo fans are earthed to a radiator support bolt using a crimp lug and heavy gauge wire.

The LC-1 heater and system earths are seperate because I didn't want noise from the SMPS heater circuit putting noise on my other sensors signals. The heater for the LC-1 is earthed on the back side of the strut tower. Beside this lug two relay earths are stacked using a drilled bolt.


I am having sporadic problems with my LC-1 and also with the connection between the laptop and the megasquirt using megatune.

Sporadic as in it seems to work when it wants to, and nothing will change and it will stop working


I am wondering if there is a more robust way of grounding, and this will fix my problems.

 

10 Amp or 10 gauge wire? (confused). If this helps, what I did to my car was run a grounding "chain" starting from the (-) side of the battery to the driver's side strut tower, the alternator body, 2 on the rotor housings, the firewall, and then into the ECU area underneath the metal cover by the passenger's feet. I tied my ground wires into that chain, and I used the negative wire (or shielding - I can't remember), to the exact spot where all my MS ground wires are connected. The wiring diagram will show you exactly what needs to be grounded.
You need to have ALL your MS and LC-1 grounds in the same "chain", or better yet, in the same spot. I personally keep my chain contacting the (-) side of the battery no matter what.
Hope this helps.

-Brian

 

The wire is good for 10 Amps. Not sure what gauge it is.

 

Try measuring the voltage on the wire you have connected to the O2 input on the MS... If the voltage is as expected, your MS board might have a fault on that input circult.

Ken

 

I have performed 2 different tests today.

#1 - I can never connect to the LC-1 using logworks, when it isn't reporting to the MS.

#2 - I just put 12V on the power wire for the LC-1 with the ignition on (but the car not running), and the LC-1 started working as normal after ~30secs (for the warmup period). I was also able to connect to the LC-1 using logworks. I did notice a slight difference in the logworks display compared to the megatune display. Can I just tell the megasquirt the voltage vs. lambda relationship for the LC-1 by choosing the 10-20AFR setting in megatune configurator? How does that get programmed onto the Megasquirt cpu?

I will take the car out tomorrow morning, and If it doesn't work I don't know what to expect.

From what I can tell the unit has ground and power, whether there is only a specific grounding problem caused by this ground being seperate from the megasquirt only when the car is running :S?

 

Forgive me if I don't remember your setup, but are you running ms2 or ms1?

 

megasquirt 1 with 029y4 extra

 

I'll have to see if I can duplicate your behavior on the bench. I seem to remember my friend's lc1 doing weird things like yours when he first wired the AFR gauge to it, but I don't remember how we fixed it.

 

Thank you Ken,

Does the data logged 02 make anyone else think that SOMEHOW the Programmable output I'm using on the WB02 has reset itself to be a narrow band output.

I just rewired the LC-1 to ms so that the default output is the wide band port. I will see if it at least gives me reasonable results.

I must have had 2 problems also, as the unit is reliably turning on but not accurately working now, just giving me garbage outputs. The odd thing is, it was working once wired this way, not so long ago. So I am a little dubious that switching between output1 and output 2 will do anything at all.

As far as I knew I had both Output 1 and Output 2 programmed to 0v to 5v 10afr to 20afr.

 

Heres the datalog

 

did this one work

 

Provided the engine is grounded (earthed) properly, the engine itself is the best place to ground other components. The LC-1 should be grounded at the same location as the MS grounds, preferably on the engine, such as UIM or rotor housing. However it sounds like a wideband controller issue more than a grounding issue. How far down along your exhaust is your w.b. sensor located? It is becoming common practice to locate the sensor at least a couple feet behind the manifold flange (or turbo flange, if yours is so equipped). This eases the heat load on the controller in the cable. Threaded o2 bungs are available that can be welded onto the pipe as needed. This was more of a problem on earlier releases of LC-1, the newer controllers appear to be a bit more robust.

 

I generally ground the heater ground somewhere not near the MS, and the system ground to the MS itself. That has always worked for me. The worst offset I've seen doing that between what the controller thinks the AFR is, and what the MS thinks the AFR is ends up being about .2 AFR just from ground offset, and the accuracy that's lost from going from digital to analog and then back to digital again.

Ken

 

One assumption I can make is that the sensor is located properly it was working 100% before the rewire!

Have a look at the datalog the wb02 readings were going mental!

I did have it wired as you suggest Ken, its not causing any interference on any other devices atm, but just it isn't working itself

 

In your first post, you said you were also having intermittent problems between MS and Megatune. If I interpreted it correctly, you might also have a solder joint issue on the board. Try gently wiggling the db37 and db9 connectors while in Megatune to check for bad joints at those connectors. The solder pads are quite small on the board. You also mentioned the 4 MS grounds, I'm assuming you were referring to pins 8 thru 11 on the db37. Pin 19 is also a ground, where do you have it grounded?

 

I have discovered an exhaust leak right before the sensor

I have to go for a drive tomorrow while I'm out I will get stuff to fix the leak then I will try the things you suggested scrip

I may very well have a dry joint between the db37 and the board. I've found other dry joints recently.

 

the sensor was still working but going spastic. In the end I fixed the exhaust leak, and at the same time I installed the diagnostic led. Straight away I found error 4, which was only apparent with the engine running. Warmed up fine without the engine running.

Well I just wiggled the plug from the sensor to the controller and it started working normally again :p