Aftermarket ecu installation questions. (7mgte)

[camrygt?]

The only grounds that are on the engine should be the one for the starter (the huge main ground to the car frame) and coil packs/ignitor etc can ground on the motor as well.

All of the sensors especially TPS water and air temp ALWAYS go only to the sensor ground wire directly from the ecu. Often TPS has its own.

ECU sensors never ground to the engine block

Also never join the ecu sensor ground wires to the engine block it must go only to the sensors

[chris davey]

I have my air temp and water temp sensors grounded through their mounting points and the negative wires which would normally go to the sensors go the microtech ground point. This works perfect for me.

The tps ground goes to the tps negative wire on the ecu.

CAS and crank sensors go straight to ecu through shielded cable like yours.

So from what camrygt says my sensors shouldn’t be working correctly?

[camrygt?]

in the ecu all of the grounds are basically joined together..the main ground for the ecu the sensor ground etc..

If you join the sensor grounds to the engine block you create another path for high current devices like the starter motor to draw current through, back through the ecu to its main ground.

As the sensor ground wires are rather thin any significant current flow will create some voltage drop on the sensor ground wire. This means that at the sensor ground is not quite ground anymore its a voltage.

This changes the reading the ecu gets from that sensor. By how much depends how good the main engine block ground in the car is. Its only going to be significant for large current flows ie starter motor.

That is why factory EFI temp sensors always have 2 pins but the temp gauge sender usually only 1, the temp gauge isn't sensitve or fast enough reacting to the starter motor stuffing up its output.
The amount of fuel the ecu needs send in to start a cold motor is however.

Also in a car with a Variable Reluctor CAS the signal output from the CAS during cranking is very small often less than 0.5 volts peak. This can be totally destroyed by any induced voltage within the ground circuit of the ecu, it really depends on how the ecu is built. To avoid any possiblity of these problems all sensor grounds should be isolated from "power type grounds"

[thechuckster]

Toyota ECU's usually bring the ground out from the ECU to a big bolt the intake plenum, there should be also an earth strap from engine to chassis and obviously then from chassis to battery.

Any flying 'earths' in the ECU engine loom will also connect to the earth going to the intake plenum.

Seeing as kemicalx is still experiencing symptons like a dodgy-earth, i'd still put money on flakey earths OR shielding earthed at both ends ofa cable run.

[kemicalx]

The wiring loom it came with had the grounds out of the ECU joined together i grounded these initially to a bolt under the dash, should I reconnect all the sensor grounds directly to the EMS? I would have though having long runs of ground wires next to each other would have induced earth loops or something??

[kemicalx]

With the air/water temp sensor grounds connected to the chassis and not the engine they are now stable during crank Thanks camrygt?! I'm yet to re-ground the TPS to the chassis aswell

Though now i'm getting an RPM spike on cranking which is a little wired, heres a log file , you can see the first time it goes up to 8850 then second crank attempt to 2950..



With regaurds to the sheilded ground wire from the CPS, on the EMS end it connects to the sheilding, at the CPS end it goes into pin G(-) on the CPS plug, as far as i can tell there it isn't grounding at both ends.

edit - i might also try wiring the CPS lead directly to the wires comming out of the CPS (bypassing the standard plug) perhaps the length of unshielded wires is too long?

[camrygt?]

usually up to 20cm unshielded is Ok as long as it doesn't sit on an ignition lead or something.

You really need to get all of the earths for all of the sensors (water Air tps and CAS) to be earthed on there own only to the ecu sensor earth wires.

the power type grounds don't really matter but sensor ones do.

I can explain how having the tps, water and air temp sensors grounded to the chassis rather than just the ecu can stuff up the signal for the CAS and hence the rpm reading but it will be rather long so I won't unless someone really wants to hear it.

Try moving all of the sensor grounds to the sensor ground wires on the ems if you are still having trouble its the most likely thing to be giving you trouble at the moment.

[kemicalx]

Cheers mate, I’ve done that now,

Having inspected the ems plug you could see out of 3 of the 4 earth pins there were two wires, with the 4th sharing the shielding and ground wire for the CPS, so i have now connected the water, air and TPS each to a ground pin of there own and the readings they put out during logging look even more accurate than they did before.

Still troubles with the injector pulse, but I’m becoming confident it's a grounding/wiring issue.

It's very hard to get consistent working/not working results, but I noticed if the coil packs were plugged in the injectors wouldn't pulse, unplugged they would.. It also appears that changing the sensor grounds was affecting the erratic injector pulse. When i get a pulse on the test light, and then test an injector by plugging it in and listening via a screw driver, they are working.. So I'm pretty confident the ecu is fine, the injector drivers are working, it just seems the signal to run the injector drivers is flakey.

[kemicalx]

update time,

I replaced the chassis to battery, battery to engine, head to firewall and added an engine to chassis strap with 4 gauge cable,

I emailed EMS about my troubles and he said all sensor grounds connect together at the ecu and the ecu is grounded (way I have it now) and that they usually ground the igniter to the engine, which i have now done. He also said to keep the trigger and sync shielded cable away from the coils/ plugs and leads which is logical, and i have done.

The bloke i bought it off seems confident that the way the loom was wired up when i got it was correct, this had sensors grounding to the head, ignitor ground likely conencted to the chassis (judging by length of cable) and the ems ground to the chassis presumably inside the car.

The problem with this setup was the wildly flucuating sensor readings, but i could depending on a few varibles get the injectors to pulse. When i had it wired up this way i did not have the engine to chassis strap installed.

The way it is now wired up i can not get any injector pulsing reguardless of spark plugs in/out. However the sensors are rock solid.

One other possibilty, the bloke i bought it off said that his Wolf had trouble reading the CPS as it puts out very little voltage and the ecu couldn't read the signal.
He said to make sure that the pickups are adjusted as close as possible to the trigger
wheel to get the most output. He thinks that if i'm getting injectors switching with the plugs out (faster engine speed hence more output from cam sensor) but not with the plugs in it could liekly be the problem.

What do you guys think? I had a look at my spare CPS and the pickups are pretty close, but I'll give it a shot anyway.

[CrUZida]

Its a pretty typical Wolf problem to not see the CPS signals properly.
Almost every one I've heard of with Wolf has spent a couple of weeks getting the signals strong enough.

Whether the EMS suffers the same problem I don't know.

[kemicalx]

one thing, do you think that if the signal is strong enough for the igniton to work fine, that the signal should be strong enough for the injectors to be working too? If so I'll think i'll send it off to EMS to have a look at

[JustCallMeOrlando]

If the ECU is igniting, then it should be firing the injectors.

[camrygt?]

true
Also the rpm reading on the ecu is the most important indication should be pretty consistance 150-300rpm when cranking

If it goes higher than about 500 or lower than about 100 or fluctuates a lot you have a problem with the CPS probably a sensitivity issue but could also be interference

A steading RPM reading means the ecu should send out consistance injector and ignition pulses check this before you send it to EMS

[kemicalx]

on the last few logs i took, RPM stayed smack on 150 throughout the intire crank, air and water temp stayed steady too,

[20ValveAE92]

hey mate i wonder if its possible that you have got an ECU with faulty injector drivers, may well be worth while to take it out and get EMS to benchtest it, atleast then you can rule out the ECU as a possible issue.