Help understanding the gt4dc citcuit for code 54

CMS-GT4

Active member
That may be my problem. I am using my INT as a power source (B) before the diode. I thought it provided the 12v source.

Could it be that it makes enough power from the INT to prevent code 54 at low rpms or idle but when there is a larger draw that it can not ive enough power to the source?

Where should INT be hooked up in this equation?
 

Gary

Moderator
In the original setup, it is hooked to a relay coil and from a relay coil to ignition key.
So, I would guess, any 12V source should be OK.
But when ever I'm not sure, I put a fuse in between.
(so that the fuse will blow before the ECU is fried)

Are you using ST185 harness?
On ST185 harness, the 12V ignition key wire is Black-Orange.
 

CMS-GT4

Active member
Yeah, I am using the 185 harness.

So should I just run a power wire from my fuse block under the dash to act as B and leave INT connec to the B wire?
 

Gary

Moderator
That what Scott did.
He uses a 10W bulb instead of resistors.
It doesn't make much different though.
But I'd still put a temporary fuse on the power wire.
 

CMS-GT4

Active member
Here is the one I am about to try. My changes are in red.

1om2kw1.jpg
 

CMS-GT4

Active member
So I unhooked the grounded LEV and then started the car for a test. Imediatly the code 54 started. I think my LEV is fine just being grounded. My pump circuit must be the problem since the IC pump is not designed to run all the time, maybe it does not check for it till the car is under load.
So I am going to run a fused power source to where E and INT are then see if that solves the problem.

I also hooked it up like the image I made. That caused a code 54 shortly after startup.

Then I hooked it up the same order as Scott's bulb circuit.

The light stays off for about 4 seconds then comes on. Maybe the way I had it hooked up the first time is right, but the resistance value needs to be increased or something.

Any suggestions?
 

Gary

Moderator
Hmmm.... I wonder the same thing.
Scott and Greg are happy with their bulb circuit.
And 1k resistors also work according to Kris Carter.
:shrug:
I'm kinda stump now.
Noone is really sure how the ECU react to the missing pump.
May be you should try higher (or lower) resistance.
Too much resistance, the ECU will think that the pump is not there.
 

CMS-GT4

Active member
Yeah, I am lost. Its weird. Any other combo makes the code 54 kickin from startup. Using the first was keeps the light off till I push the gas.

I am going to see if scott might have any ideas. If not, I might try the bulb idea.
 

CMS-GT4

Active member
So in the st205 wiring diagram, the other wire on the LEV sensor is just a ground. So grounding it to the body is its solution.

Here is the full bgb page and an enlarged veiw of the pump layout.
After looking at this, I wonder if a relay needs to be involved somehow.

I am going to try a few more things and if all else fails, I will try the bulb method.

rm399e315lr7.jpg


explodedrc9.jpg
 

CMS-GT4

Active member
Here is an image and reply from Kris (Who did the writeup).

Hi JC,

glad you find the article fairly useful...so far! Not so good is that you are having problems getting it to work for your particular installation.

I fear that the control of the pump is a bit tricky and on the 205 this is not a straight forward solution, unlike the 165 and 185.

For the stock setup on a 205 the article I wrote works fine. The IC level sensor is bridged and the resistors are added to the wiring AT THE PUMP.

The monitoring of the circuit so that the ECU does not throw a code is open for debate, and I know that this has been tried before, without success

I have a diagram I sent to someone before and can send you that may help? Please drop me an e-mail to kris@gtfours.co.uk and I'll reply with the diagram.

The solution may work around when the ECU activates the pump relay, and using this signal/voltage may help.

For background info the pump operates like this on the stock 205:

Press throttle - pump starts and runs all the time the throttle is pressed
Throttle released - pump continues to run for 30s and then stops
Throttle pressed - pump runs again

Hope that helps?

The image looks like the circuit we made at the ecu.

ewd7colourcr7.jpg


Sofar what it looks like is I prevent the code at startup. So it much check for the LEV first, then it checks for the connected curcuit. Which is why at start up and idle the current setup at the ecu keeps the light off. Once i get ovre 1700 rpm, the relay activates the pump which looks for the pump working. So somehow for this to work at the ecu, I think the INT needs to trip a relay that might send a signal to MTT, M+ and M-

What I wonder is, if at idle the base value is fine, is just needs to change. So maybe a relay with a different resistor. Activated by INT then a resisted 12V passes through the relay to the rest of the cucuit.

If I can't get this to work, then I will just do the light bulb method.
 

Gary

Moderator
Another throry, every relay has internal resistance.
Thus, if you connect INT to 12V directly, the ECU (if it's that smart) could realize that there is somthing wrong.
Try put a small resister on the INT line and see if it works.
 

CMS-GT4

Active member
I m just going to use the light bulb route. Only problem now is that they no longer print the wattage on the package or the bulb.

So as soon as I find a 10 W bulb, I will wire this up and call it a day.

Went to the website to get a part number for a 12v 10w bulb. So I will go pick it up in the morning.
 

CMS-GT4

Active member
:x :x :x :x :x :x :x :x :x :x :x :x :x

So.

I hooked it up exactly how scott did and I still get code 54. Its off for a few seconds then turns on.

Does it matter the kind of relay I use? I pullede a 4 wire relay out of my st205 fuse box.

The light comeson when I turn the key to acc, but shortly after starting I get code 54....

I also treid a variation of things to make this work. Even mixing resistors with bulb.
 

Gary

Moderator
:shrug:
That ECU has a mind of its own. :shoots:
Sell the damn ECU and get the MR2 one.

Someone with ST205 and FMIC setup please check the voltage at each terminal for him....
 

CMS-GT4

Active member
Does the amperage of the bulb matter? They took forever to find a 10W 12V bulb. They did say the amerage was sorta weird on it. Its a jetta bulb.

If I knew the amps required, I could get a resister thats the same value as a 10W bulb.
 

hacker_720

New member
P= VA
10W = 12(A)
A = 10/12
A = .8

I can't see the amp mattering for the ECU, it's just sinals, it will be like 2-40mA (0.02 - 0.40 amps) so that bulb will probably work for you.
 

hacker_720

New member
oh.. sorry. resitance...

V=IR
12 = (.8)R
R = 12/(.8)
R = 15

yup thats about right, I aggree a bulb is probably 15 ohms

what you really want to know is what is the resitance of the sensor in the IC then just put that resistance in, don't get carried away with the old bulb trick.
 

hacker_720

New member
CMS-GT4":2xxfgzow said:
Here is an image and reply from Kris (Who did the writeup).

Hi JC,

glad you find the article fairly useful...so far! Not so good is that you are having problems getting it to work for your particular installation.

I fear that the control of the pump is a bit tricky and on the 205 this is not a straight forward solution, unlike the 165 and 185.

For the stock setup on a 205 the article I wrote works fine. The IC level sensor is bridged and the resistors are added to the wiring AT THE PUMP.

The monitoring of the circuit so that the ECU does not throw a code is open for debate, and I know that this has been tried before, without success

I have a diagram I sent to someone before and can send you that may help? Please drop me an e-mail to kris@gtfours.co.uk and I'll reply with the diagram.

The solution may work around when the ECU activates the pump relay, and using this signal/voltage may help.

For background info the pump operates like this on the stock 205:

Press throttle - pump starts and runs all the time the throttle is pressed
Throttle released - pump continues to run for 30s and then stops
Throttle pressed - pump runs again

Hope that helps?

The image looks like the circuit we made at the ecu.

ewd7colourcr7.jpg


Sofar what it looks like is I prevent the code at startup. So it much check for the LEV first, then it checks for the connected curcuit. Which is why at start up and idle the current setup at the ecu keeps the light off. Once i get ovre 1700 rpm, the relay activates the pump which looks for the pump working. So somehow for this to work at the ecu, I think the INT needs to trip a relay that might send a signal to MTT, M+ and M-

What I wonder is, if at idle the base value is fine, is just needs to change. So maybe a relay with a different resistor. Activated by INT then a resisted 12V passes through the relay to the rest of the cucuit.

If I can't get this to work, then I will just do the light bulb method.

this is wrong, that 1k resistor that is going from ground to 12v is not doing anything.
 

CMS-GT4

Active member
Update.

I was working on it again. I started all over again from scratch to do the lightbulb method.

So here is the thing.

INT acts to trigger the relay. INT is not supposed to trigger the intercooler relay till you press the gas. Yet the light turns on as soon as the key is set to acc.

I though something did not seem right. So I started messing around.

I decided to switch the INT trigger wire with the lightbulb. I turned the key on and the light did not come on. Either I was reading this relay wrong or its printed backwards.
However, the cel stayed off till I hit the gas.

So anyway. I think the relay is broken. I just used an old relay out of the st205 fuse box.

I am going to get a new relay in a few and see if that works.
 
Top