[hosedup]

Looking at the way the coolant level sensor has been bypassed over the years has always struck me as not great. Let me say this now, I am not trying to start a debate on safety, necessity of the sensor, warranty voiding, etc. I want to share what I've come up with and am curious if anyone else is as well.

I look at car parts from the stand point that if it did not have to be there for some reason, it would not be there. Sure a part may only be a couple $ but multiplied by all the cars made plus the labor, it costs to have it.

So here is how I see it:
-The sensor needs to be there, it serves a necessary function
-It is far too sensitive
-When it false triggers, its expensive to undo
-There are implications to voiding of warranty
-There are risks to starting a fire
-Solution needs to be functional, not false trigger and be stealth

After thinking about it, I decided to buy a sensor, cut it open and play with it. First plan was to bypass it internally and clean up the outside so it looks like it was never modified. While I liked this a great deal, there is the implication that if the coolant did leak out, there is still a small possibility of a fire. The nice thing is that it could be blamed on a faulty sensor.

I thought more about it and came up with a different solution. Since im already inside the sensor, what would happen if I put a timer circuit inside so that if the sensor did trigger, it would have to stay in the faulted state for a minute or so before actually giving the fault signal to the ECM. This way, if cornering hard to dodge a deer or panic stopping, the fault would only be active for a short time thus preventing nuisance tripping that requires dealer intervention. If the car ever goes in for maintenance, the mechanic would have no idea this was done and therefore warranty voiding isnt an issue. If there was a loss of coolant, this would only delay the actions taken by the ECM for only a minute and I do not see that causing any extra damage.

I have not yet built the timer version but I am pretty sure it would work. I'm curious about what would be a good delay. Long enough for nuisance fault prevention but short enough to prevent damage.

 

[Steverino]

You don't want a debate, but then list a number of "here's how I see it" assertions open to debate. As a result, don't be surprised if you get a debate.

You should have kept the thread focused on the time question.

 

[hosedup]

You don't want a debate, but then list a number of "here's how I see it" assertions open to debate. As a result, don't be surprised if you get a debate.

You should have kept the thread focused on the time question.

Steve, you are correct. When I posted it I thought it would be best to say here is what I think and why. Oh well, if it stirs up a debate then it stirs up a debate. My thoughts were a way to stealthily solve the problem and the dealer tech would be none the wiser.


All that said, I am curious what others think of my version of WOTs defeat plug.

 

[Cord]

We need to FORCE GM to re-program for free each time the sensor does a glitch or hit a bump with low fluid and not have to pay $200 plus. This was an add in workaround and needs to be addresses again.

There is a program to upgrade the sensor and cable - may be hard to get dealer to do it.

AND the sensor is NOT a low level sensor it is a things are almost CRITICAL level sensor with no user reset.

I would love to see an actual liquid hight sensor for that battery coolent tank.

These days laser may be better that the Ultrasonic Sensor (used in my day )

https://www.protocentral.com/motion...-laser-tof-sensor-breakout-0642078949623.html
remember that from playing with a Arduino.

 

[JuneBug]

All that said, I am curious what others think of my version of WOTs defeat plug.

I thought about doing something stealthy like that, but in a much simpler way. My idea was to dig into the sensor as you did, and simply install the resistor and then re-pot it with epoxy. It would be WOT's defeat plug, but in a factory package.

 

[Barry]

Your solution seems to be based on the assumption that the only problem with the stock sensor is that it is too sensitive. That may be true, but I think WOT's assessment of the problem was that the sensor could malfunction, and worse yet, do so intermittently. I take that to mean it is not just a sensitivity issue, but the sensor could send a purely false signal even if the reservoir is full and there is no g-force disturbance of the liquid in it. I don't think your sensor design would solve that issue.

 

[Fred_B]

Assuming the sensor is the problem and not the wiring, a timer might work. Make it so that the low coolant level needs to be present for 30 seconds or something. But this could be fixed a lot easier by putting a timer in the software that is looking at the sensor, such that the logic would be "if battery coolant is low FOR 30 CONTIGUOUS SECONDS, set fault to true."

Of course the software would have to be retested and re-certified (to be sure you didn't introduce some other problem when you changed the software), and then we would all have to have the update done at a dealer. The only chance of this happening would be if they could include it in a software revision that was happening for some other reason. When I used to work with software we would collect problem reports until we had enough of them to justify putting out another release, unless a real urgent problem came along that demanded a release right now.

 

[Steverino]

Your solution seems to be based on the assumption that the only problem with the stock sensor is that it is too sensitive. That may be true, but I think WOT's assessment of the problem was that the sensor could malfunction, and worse yet, do so intermittently. I take that to mean it is not just a sensitivity issue, but the sensor could send a purely false signal even if the reservoir is full and there is no g-force disturbance of the liquid in it. I don't think your sensor design would solve that issue.

The sensor itself seems to be the core issue. It can flake out even with the coolant tank full.

Yes, others have triggered the fault by not keeping their coolant tank filled to the correct level, but it appears the sensor itself is more often the issue, and sometimes the spliced wiring used on 2011/2012 retrofits.