dj55b

iTrader: (2)

This is a problem that I've had over the winter time, can't remember exactly when it happened but it did. This is what was happening;

As i turned on things like my lights or especially the blower, i can see an increase of upto 3% in TPS value on idle and would drop my temps to under 66*C all the time so it was always under cold enrichments until i dropped the temperature in the chart, but the idle at night time plus on the cold winter days having the blower on, I had to go in there and change my TPS values all the time.

First thing I tried was beefing up (4 gauge) the ground to the engine. I originally had some 14gauge wire going to it, but after checking the voltage drop between the engine, engine bay strut towers, and a couple of other areas I noted that I had about .15V difference. After the 4 gauge the engine was pretty much 0 all around then but all the stuff in the engine bay was still around the 0.15V difference.

Second thing I did was from the 4gauge point on the engine I ran a 8 gauge wire to the strut tower where the igniters were since that area seemed to have the highest voltage difference, but apperently that didn't do too much neither. That when I went a little excessive probably and added direct 8 gauge ground to each of the igniters and presto the whole engine bay is now 0.01V difference at the highest.

Lastly I decided to take some inside reading. Checking random bolts here and there seemed like the center of the dash area had about 0.1V difference also, so i decided to run another 8 gauge wire tapping from the engine area to there and once again brought that voltage down to 0.01V.

Turned on the car (which by the way now seems to turn over a bit quicker too), and check the reading on Megatune ... we're all back to normal and readings seem to be even better than ever.

On a side note I've always had a 0 gauge from the battery to the chassis, but it seems that even that monster isn't enought to take care of everything.

muythaibxr

Often having several small ground wires is better than having one large ground wire.

I can take a picture of an example on my rx8 (in factory condition).

They used several small wires (8 or 10 GA I think).

The way this was explained to me is that a large wire with periodic amounts of current going over it (like injectors squirting) can breifly act like a resistor, which shows up as a slight voltage drop due to the cross-sectional area of the wire. Having several smaller wires still allows the same amount of current to pass in total, but without that slight resistance during on-off loads.

That can usually help a great deal with jitter and voltage differences.

dj55b

iTrader: (2)

Good to know that I was stepping in the right direction then!

dj55b

iTrader: (2)

Alright, I guess I haven't won the battle just yet again. Well about a week or so later after doing all the grounding and though things we're going perfectly, it started acting up again. I check all my ground cables and everything is still in order as I had left it.

So getting pretty frustrated, I remembered that my power cable from alternator to the fuse block had quite a drop and thought that maybe it wasn't a ground increase that is causing, but too high of a voltage drop rather. So as I change out the wire, I disconnect the wire loom and low an behold, my TPS ground wire was not soldered on for some stupid reason. And thought to my self it makes kinda sense now since my main ground was right in that area i must of touched the loom and got them closer and had a better connection going through them. Well wrap everything back up, start the car up and presto back to normal and even better actually than normal ... ah the things that soldering can do ....

Well atleast I thought that was the end of it all trouble will be freed up now and living large until today again (once again about a week or so after being fixed up) the problem re-occurs. Once again start back at ground one ground wires, check, soldering on the ground, check, but still problem presist. If you have any suggestions let me know.

The 2 things effected are always coolant temps and TPS. Temps drop, and TPS increase. What could make both of these do so simultaneously?

pmrobert

Are you bringing the TPS and CLT sensor grounds back to the MS ground or grounding them to chassis, block, etc.? The effect you're seeing is probably a difference in ground potential between what the MS is seeing and what the sensor is seeing, simplistically.

-Mike

dj55b

iTrader: (2)

I'm 98% sure that I used a MS ground. I'll probably have to hack away at that harness to double check things. I don't understand why it goes on and off though even if thats the case.

pmrobert

Does your IAT change as well when this happens? If it doesn't, emulate what you did with it. By MS ground, I mean bringing the sensor grounds back to the same place the the MS is grounded. I just bring the sensor grounds back to one of the many ground pins in the DB37 ensuring that the processor and sensor/voltage divider circuit in the MS all see the exact same ground potential. Always works just fine. It sure sounds like you're seeing a ground offset effect but I'm sure you know that.

-Mike

dj55b

iTrader: (2)

Well its having one of its moments again, and it decided that after going to work and starting up the car it will be fine again. Only difference I turned off the car and turned it back on lol. This time though I took voltage readouts of everything and wrote everything down. So that if it decides to come back, which I believe it will, i'll check all those points and see if I can spot a difference.

All my sensor ground go right back to the grounds on the squirt. I have a ground from the battery to the case of the squirt (8 gauge), then I have one of the grounds from the squirt go to the engine itself. I was going to try out disconnecting the ground from the squirt to the engine next though till it decided to work.

Other grounds on car include:
0 ga from battery to solid chassis point (seat belt holder)
4 ga from batter to engine (which branches off to)
- 4 ga to driver strut tower
- Another 4 ga "T"ed off to the engine (stock ground engine wire)
- 8 ga to leading coil pack
- 8 ga to trailing coil pack
- 8 ga to inside dash

I think thats all the grounds in the car. Like I said previously most of areas in the engine bay do not see more than 0.01V, the highest point that I actually found today was from the engine ground to the alternator casing at 0.04V.

Also while the car is running, the main engine ground to the ground on the TPS was 0.08V. Would be nice to see what you guys see for voltage drops across the engine bay for reference perhaps and TPS maybe?

pmrobert

While interesting and always good to identify and minimize, your small ground offsets bewteen all those components shouldn't have anything to do with your problem. The MS has a 5 volt regulator insode it; it uses that accurately regulated 5 volt potential to it's local ground from the DB37 to generate voltage based on the voltage divider principle. You are somehow developing a ground offset beteen the processor's idea of 0v and the sensor's view of 0v. Even though you are bringing sensor ground back to the DB37, if the sensor ground is contacting say block ground tyou have a ground loop. I wonder if the metallic shell of the CLT sensor or the mounting bolts for the TPS are electrically connected to the sensor ground wire. That would be plenty enough of a ground loop to explain the problem. I just checked 2 GM CLT sensors and both keep the sensor ground isolated from the sensor ground. Oddly, I checked a 2nd gen CLT sensor and it does not isolate the sensor ground from the sensor body. Grabbin' at straws here, I know! Does the IAT exhibit this behavior as well?

muythaibxr

In all your listed grounds in your last post, you don't say where the MS is grounded...

Ken

dj55b

iTrader: (2)

The paragraph above it says it a bit:
- Ground from battery to squirt
- ground from Squirt to engine
- ground to the case of squirt

dj55b

iTrader: (2)

I don't usually monitor IAT's but I'll switch out one of the guage next time for the IAT and see what that does. What are the possibilities that the 5 volt regulator is going? I know they don't fail easily, but possibility? Or could it be overheating (ie. not touching the aluminum bar in the squirt properly to dissapate the heat?)

pmrobert

If you could create a logfile of good and bad behavior that would be extremely helpful. I'm not sure that providing 2 different ground locations to the squirt is a great idea. I'll have to think about that one. It shouldn't matter as far as the sensors are concerned - but - the less variables the better. The regulator going partially ****-up would be unusual. In my experience they either work or not and still wouldn't explain the apparent differences you see. Log would be great!

-Mike

dj55b

iTrader: (2)

I'll get a log going, but I did try to take out the ground wire going to the engine and the car runs really really bad and just incase I did turn on the blower fan to see if it was still effecting things and it was, so that can't be the problem then. The casing ground I did after the fact of all this started happening but didn't do anything. I just figured maybe the casing wasn't shielding the component properly.

Also the car is back to doing it this morning, and all the stuff that I had checked ou yesterday are still all the same values as today. All that I changed on the car today was a starter wire. I had 2 wires in there and didn't always have the greatest connection so changed it out to one single long one that is better quality. Another thing I did also was put on the cover for the passenger foot area to cover up the ecu again.

Another thing that I can maybe think about is that maybe the carputer is sending back a dirty ground or something through the DB-9 connector, but wether the computer is on or off, it still makes then engine go richer which tells me its still raising the TPS value. I can try to disconnect the DB9 wire and try it out fully without it ... but I still have my doubts.

dj55b

iTrader: (2)

Oh and I did check if the IAT drops and it does (~2 degrees). Not as much as the CLT does (~5 degrees with lights on and blower to max), also TPS is about the same value as that 4-5% increase.

muythaibxr

This is a bit bad. That's definitely going to cause ground loops.

Ground the squirt to 1 place and 1 place only. I generally ground it to the engine.

Ken

dj55b

iTrader: (2)

So disconnect the battery ground and put it on the engine bay also? and can I keep the case ground ?

I just find it odd that it worked for about 1/2 a year all fine.

muythaibxr

That is a bit odd.

Generally, I do the following:

Ground the battery to the stock ground points (strut tower, and starter).

Ground the MS to the stock ground point for the ECU (2 grounds to the engine) using as many wires as I can running all the way out to the ring terminal (instead of combining them).

Ground the engine to the chassis at the stock point (firewall to bell housing on a 2nd gen).

This seems to work well for every install I've done.

Ken

dj55b

iTrader: (2)

Well i removed the battery grounds and moved them on the engine with the other ones but that didn't do anything. Exactly the same as previous.

I did try a crazy thing which would make thing it should act worse but made things go back to normal. I added a ground wire from the engine squirt ground to a ground on the TPS (which I think I wired with all the other sensor grounds) and everything "works" right now like it should be. We'll see if that last a week or so though before going back to square one.

To me though that should be introducing more noise is anything to that signal no?

muythaibxr

The trick is finding where the voltage drops are, and fixing those. It sounds like you found them.

Generally, the more different places you ground things, the more likely that you're going to end up with ground loops, and the more likely issues like this will occur.

With the way I recommended to ground, electricity will flow out of the positive terminal to the various things that are hooked up to +12v. From there to the MS, it would go to the engine, then to the chassis, then back to the battery.

With multiple different ground points, you can also end up with current flow on the ground wires between the different points.

Ken