You can get your cable from these guys. I got mine about 4 years ago for the same price before COVID
Tbi rough cold starts
- Thread starter JPVortex
- Start date
Disclaimer: Links on this page pointing to Amazon, eBay and other sites may include affiliate code. If you click them and make a purchase, we may earn a small commission.
Greetings JPVortex,
If you give yourself the troubleshooting gift of seeing the live data, then I'll bet that
you will see actionable info leading to a fix on the 1st try. For example, assuming that
your vehicle runs 100% when fully warmed up (eliminating lots of troubleshooting
variables elsewhere) ...then an ECT sensor which has lost it's calibrated response to
temperature changes would explain the symptoms you describe. (Especially the no fast
idle when cold, but idle OK at normal temps.)
In the '94 GM Service Manual** the guidance is that when the engine is cold, a good
ECT sensor will measure within 5° F of actual temps. This is pretty tight, so obviously
a relatively accurate temp sensing must have been considered a key component in the
strategy of meeting cold engine customer driveability (+ emissions) goals. (!)
So if you have access to the right scan tool, comparing what the computer thinks is
the ambient temp vs reality would be a quick & easy go/no-go check. (ie: may not
prove that the sensor is completely accurate throughout the range, but will find a
gross failure.)
OK, definitely a solid ongoing maintenance item. And even if this didn't clear up the
cold running issue, it wasn't a parts cannon misfire at all. Why? Because it does help
the troubleshooting process to know both what to consider, and also we know what
it's not.
Given this, the rebuilt TBI unit is off the troubleshooting table for now. (ie: the problem
definition is more of a incorrect fueling calculation delivered accurately vs a correct fueling
calculation delivered incorrectly due to a malfunctioning TBI assembly. Does this make sense?)
Alright, at this point I'm going to offer up something that is a bit controversial in this forum.
Ideally, in order to find/fix the issue on the very first try, you would start & finish your
troubleshooting session with a proven-good scan tool & monitoring the live data.
(ie: looking over the computer's shoulder and seeing what it's perception of reality is.)
I have no argument with this. Actually it would be my personal preference as well.
But at the same time I realize that when there's no scan tool on hand (or available anytime soon)
you need the truck to go work, and instead of just throwing parts at it out of desperation,
you'd like to at least follow some sort of basic troubleshooting approach out of the GM service manual.
If you find yourself in this pickle, then if I was your neighbor we'd scare up an ohmmeter and measure your
ECT sensor resistance (still on the engine) against this Temperature vs Resistance chart:
You must be registered for see images attach
Cold unstarted engine:
68° F garage? 3,520 ohms
5° F driveway? 21,450 ohms
Engine fully warmed up:
(195° F thermostat) ~241 ohms
Notice we're not trying to see only a couple of ohms difference between a
stone cold unstarted engine vs a happy engine fully up to temp. (I
hate those 'angels dancing on the head of a pin' type measurements
that can't be made reliably outside of a calibrated lab. :0)
The readings are in the middle of the meter range, clear, and unambiguous.
This is a go/no-go check that can be done anywhere that there's a
reliable multimeter available. (Bonus points for verifying/proving these
readings to yourself when your GMT400 is operating perfectly *before*
needing to troubleshoot a genuine cold-start or cold-running issue.)
One more thing. One of the major strengths/advantages of the TBI
setup is it's relative simplicity. When it's all said and done, the 4
primary inputs that the TBI relies upon to work properly are:
* How hot or cold is the engine? ECT
* What is the load? (inferred from the manifold 'vacuum') MAP
* What is the RPM? (Ignition)
* What is the driver demanding? TPS
Pulling this all together, if the engine runs poorly 100% of the time,
we have to figure out pretty much everything, including the mechanical
health of the engine.
But IF the engine runs well at normal operating temps, but only gets it
wrong when it's cold, then that allows us to remove lots of stuff from
the troubleshooting table. Especially when the TBI unit has been recently rebuilt.
Hopefully you get to use live data to prove/disprove that the ECT sensor is no longer
reporting accurately/misleading the computer.
But while waiting for the live data stuff to show up, if you feel like it you could grab
an ohmmeter off the shelf & do the quick temp vs resistance check that GM brewed
up for their techs.
And in the perfect world? You would do BOTH, and show the forum how this
cross-check agreed or disagreed with each other.
Best of luck. And let us know what you discover.
Cheers --
**Highly recommend that you download your own copy of the '94 FSM from here. ('94 Factory Service Manuals)
Last edited:
Do you have EGR valve installed still? try making a block off plate and see if problem goes away. had a '91 that would intermittently run like crap on startup. EGR had been replaced but once I deleted the problem went away.
Hi?
When I did intake gaskets, my original CTS broke and that’s when I replaced it with a Duralast sensor from autozone. It seems to be around that time of when I replaced that sensor that the cold running issue really began.
But the last thing I want to do is throw parts at it, so I think if I can get a way to live read data and see a definitive answer or what info my ECM is getting, that would be great.
Hi, thanks for this amazing write up! I do agree I suspect something with the engine temperature monitoring.
When I did intake gaskets, my original CTS broke and that’s when I replaced it with a Duralast sensor from autozone. It seems to be around that time of when I replaced that sensor that the cold running issue really began.
But the last thing I want to do is throw parts at it, so I think if I can get a way to live read data and see a definitive answer or what info my ECM is getting, that would be great.
I do have an EGR valve still installed. The diaphragm seems to hold pressure as it should with the old finger holding test lol.
At one point I did disconnect the vac lines to it and plugged them off, that made no difference in cold engine operation.
Yeah, this is one of those issues that can be easily argued from either
side of the fence.
Those who consider their time billable would argue that throwing a new sensor
at the problem (empirical testing) is ultimately cheaper than all the time they
would spend attempting to put together a test setup capable of giving them a
reliable 'live data' feed.
On the other side of this philosophical fence, putting together a live data feed
setup is a *one time* cost, and taking the long term view, the next time you
need to troubleshoot you can now bring the unfair troubleshooting advantage
to the new problem for -0- additional dollars.
****
And in the days gone by, I would have easily justified the cost of pulling together
the live data test setup by contrasting it against what the local repair shops
charge per hour of diagnostic time. That is, you don't have to avoid
paying for all that many hours of diagnostic time before you more than pay
for your live data tools.
Now? At the risk of sounding a bit like a prepper, the more self-reliant you
are when it comes to troubleshooting a GMT400-era issue, the better off you
will be. And if anything, it's going to get even harder to find true TBI talent
as the old guys who grew up with it retire. (Or just get tired of typing. :0)
Not to mention the additional complication that solid troubleshooting can no
longer be guaranteed by the substitution of new parts. Increasingly, this is
simply the unvarnished truth. :-(
****
So that's the troubleshooting philosophical dichotomy we're faced with.
But for me there's an absolute tie-breaker that must be considered. I've
fixed *plenty* of machines over the years without being able to describe
exactly what fixed the failure? Always made me feel like I was still a kid
stuffing dirty dishes into a Hobart in a (early '70s) El Chico's kitchen, with
clean dishes magically coming out the other end. Sure, I got stuff done,
but at the end of each day what did I learn besides a good work ethic?
On the other hand, when I get a chance to study the failure, make a
few measurements, possibly observe real time data, use this info to
tighten the problem description...and then choose the right part and
fix it on the 1st try? Man, that's feels g o o d.
Not only does the problem get fixed, but as a side-benefit I get some
valuable real-world schooling that might even help me out down the road
on a new & different troubleshooting opportunity.
****
So, empirically testing with a *known-good* ECT sensor substitution is one
method. Using live data for mo' betta info (and also at the same time verifying
the associated wiring harness signal integrity too) ...is also a valid approach.
Good stuff. Looking forward to what you find/fix.
Cheers --
Last edited:
Thank you, hopefully I’ll find something! I’m going to throw a new temp sensor in aswell as purchase an ALDL to usb for tunerprort. Hopefully thatll give me the data I need! Going to throw one part at it because it’s a cheap sensor!
Copy that. Whatever you do, please don't throw the old one away,
especially if the new part clears the fault.
In a previous life there was a large banner over a doorway into the QA section
that simply said, "A Defect Is A Treasure." If we can measure the bad part
caught red-handed, then this is another valuable data point we can add to the
collection.
Good stuff. Let us know what you find.
