Do you think I fried something?

Sam Powell · Post subject: Do you think I fried something? (Mon Sep 24, 2007 7:03 pm)

I inadvertantly put 12 volts to the intake air temp sensor, which would then have gone to the ECU. The car started and ran, but was very rich. I just turned everything off, and decided to call it a night. Do you think I damaged something? I can hook up the laptop and see if there is an error code, in which case it will go into a default value, which may be the cause of the rich mixture. I will look into this tomorrow evening.

The problem was caused by trying to use the bulkhead connector to route the sensor voltage to a dash gauge. I got mixed up reading the FSM since they turn the digram various ways from page to page. I find their diagrams very confusing, obviously. If I did damage something, Accel will repair it for $75,which is a pretty good deal, but I feel pretty silly and somewhat annoyed right now. What cha think?

Sam

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

turbofish · Post subject: (Mon Sep 24, 2007 7:11 pm)

might just be the IAT sensor. if it reads(or doesn't at all) really cold, it would add extra fuel to compensate for the much denser air.

_________________
'99 T/A
plannin out a 64-66 turbo SL6 barracuda

GTS225 · Post subject: (Mon Sep 24, 2007 7:27 pm)

With what I know about EFI systems, (which is mostly 80's vintage knowledge), I'd have to agree with Turbofish. If the IAS isn't giving a correct signal, or no signal at all, the processor is going to fall back on a default of cold, or very cold intake temp. The result will be a longer-timed shot at the injectors, because it'll believe it needs to richen the mixture for a cold start condition.
If you take a look at the Bob Debias article on EFI from an 80's Buick 3.8 engine, you might be able to fool the processor with a variable resistor, (potentiometer), and thus be able to richen the mixture at will.

Roger

Pierre · Board Sponsor Joined: 08 Feb 2003 Posts: 3657 Location: CA

Only way is to wire it right and find out, unles you have a time machine whats done is done you can't undo it.

I don't think you blew the sensor itself. Theres only a small amount of current going through those things and they are just resistive. If anything you blew the corresponding input on the computer, assuming it wasn't protected properly. But for the price they cost I sure hope they added a couple cents worth of components to protect the input/outputs.

turbofish · Post subject: (Mon Sep 24, 2007 10:28 pm)

a potentiometer in place of the resistor would be a way to check the computer. set it low then turn it up, see if it stays rich.

_________________
'99 T/A
plannin out a 64-66 turbo SL6 barracuda

Pierre · Board Sponsor Joined: 08 Feb 2003 Posts: 3657 Location: CA

Not a valid test unles you know the specific temp that the computer should read at x resistance. Just because it changes temp reading (or how much pulse width changes) with varying resistance doesn't mean its doing so according to spec. Silicon can do funny things when its blown. Also not a good idea because if you go too low it may pass too much current through the comp. Unles you know more about the computer don't experiment. Maybe something in their doc's about how to test sensors and/or that input?

Judging by the pictures the accel sensors look like standard gm parts. Measure yours Sam at what temp, and you can either find the sensor curves online or I can compare it to mine. Replacment may be a parts store away....

mpgmike · Post subject: (Tue Sep 25, 2007 2:00 am)

I agree that you didn't damage the temp sensor. I assume you went back and corrected your wiring problem and still have a rich mixture. The coolant temp sensor and oxygen sensor has more to do with mixture than the air temp sensor does. If anything, the air temp sensor affects ignition timing more (assuming you programmed that in).

The way the air and coolant temp sensors work is the ECU sends a 5 volt reference signal to the sensor through one wire and the other sensor wire is sent to ground. The sensor being a variable resistor affected by temperature will bleed off that voltage to a degree. The computer then monitors that voltage through a high impedence circuit. By sending 12 volts to the sensor (assuming the high side and not the ground side) the high impedence circuit in the ECU saw a voltage ranging between battery and ground; factored by the resistance of the sensor.

3 possibilities:
1- the 5 volt regulator fried. Check the voltage to the sensor with it unplugged, should be 5 volts. This same regulator circuit may be employed to drive the coolant temp circuit causing your rich mixture.
2- you fried the high impedence sampler
3- your gauge is bleeding off voltage and skewing the readings to the ECU.

Then of course you could have hooked up something else that you deemed irrelevent to the problem and didn't tell us about it and that's the problem.

Mike

_________________
Recognized by US Federal Courts as a Fuel Economy Expert!

Sam Powell · Post subject: (Tue Sep 25, 2007 4:42 am)

Thanks for all the thoughtfull replys. I disconnected the gauge.(I plan to put a second sender in for the gauge next time. Everything is now wired correctly, but the ecu keeps throwing a code for IAT temp over 4.95 volts, which sends it into a default that translates to 50 degrees. I can clear the code, and it goes right back after 2 seconds. All the other inputs seem to read ust fine, which would suggest that the reference voltage regulator is working OK.

It is a standard GM open element sender. I will go back out and check the voltage from the sender. If it is higher than 4.95, then the ECU is probably OK, and I will replace the sensor. I actually have a spare, new in the box. If the sender is putting out less than 4.95, then I think I can assume I will need to send the ECU back to Accel and they will repair it for $75. Not a high price to pay for stupidity, all things considered. Wink

I will let you know. Thanks again.

Is it correct to check this voltage by putting the red probe on the output of the sender, and the black one to ground? In the Accel system, all of the ground wires are the same grey with red tracer wire, so signal wire is easy to tell.

Sam

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

Sam Powell · Post subject: It was the sender! (Tue Sep 25, 2007 4:59 am)

I went out and checked the voltage coming from the sender, and it bounced around slightly above and below 5 volts. Obviously it was not doing anything to the reference signal any more. I put in a new one, and it now reads 2.5 volts. The price of supidity was not quite so high as I thought it might be.

My understanding of all things electronic is very sketchy at best. I can sort of understand the flow of electricty in simple circuits, but beyond that, I am kind of lost, and just put my faith in the things others tell me. It is not my thing. Thanks for all the replies again.

I have to take this dashboard out one more time, at least. The routing of my AC vent tubing is interfering with the windshield wipers. I am getting pretty good at getting this thing in and out. I certainly could drive it in good weather for awhile, and then fix the wipers this winter. Lou's '64 didn;t even have wipers last time I saw it.

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

	Slant Six Forum, Forum Index -> Fuel Injection	All times are GMT - 7 Hours
Page 1 of 1