When an ignored automotive cooling system acts like a battery

Oct. 30, 2017
I'm bringing up the typical car battery because I want to talk about the relationship of batteries and cooling systems. What's that? You say you're not aware of any relationship between the two? Maybe that’s because you weren’t told why it can exist.

I've said it for years that if my body was a car, its value wouldn’t be so adversely affected by its age than it would be by its mileage! High mileage and vehicle age are things we must take into consideration when performing maintenance. It's much the same as when a doctor sees a patient. Doctors take into consideration the age and other factors of the patient before making recommendations.

This article is a follow-up to one I wrote last year (“Juggling routine maintenance and repair challenges,” November 2016) about some maintenance I was performing on my own car. In the time since, it has occurred to me that all of us know what maintenance should be performed and at what mileage. If we don't, we can always look it up in our service information or in the owner's handbook. What a lot of us don't know though, is WHY we perform tasks in certain ways.

Batteries and cooling systems

Take for instance, in my previous article I mentioned when refilling a cooling system, I mix distilled water with pure antifreeze instead of using tap water to produce a 50/50 mix. I received numerous letters and emails regarding that particular maintenance procedure. Here, I will expound on why it's so critically important to maintain a clean cooling system.

(Image courtesy Mitchell 1)  This is an OBD I vehicle, using a single wire oxygen sensor. Notice the double digit code instead of the “P” style codes you may be more used to.

Now, I don't claim to know a lot about chemistry, physics or science. What I've learned came from people much more educated than myself and has been proven repeatedly, therefore I believe it. I try to keep things simple in my mind (I am getting older, you know).

Take for instance the ways that I think of a typical car battery. We start with an insulated case that houses plates that are made of lead and other metals. We fill this case with a non-conductive fluid, which happens to have a very low pH. In other words, the fluid is acidic. So basically we've submerged two or more types of metals in an acid solution. Doing so creates voltage. Is that a basic enough description? 

Yes that's a very basic description, and most of the time I really don't need to know any more details. I mean, do I? Do I need to know exactly what pH? Do I need to know what metals? Do I need to know how many plates? Even when performing battery maintenance or electrical system diagnostics, I don't need to know all of the scientific details. No, I’ll perform the tests as instructed and determine whether they pass or fail to know where a fault is. I said I like to keep things simple.

I'm bringing up the typical car battery because I want to talk about the relationship of batteries and cooling systems. What's that? You say you're not aware of any relationship between the two? Maybe that’s because you weren’t told why it can exist.

A story of explanation

Randy is a friend of mine who lives even further away from the city than I.  He tries to save himself time and money when he has a vehicle problem by doing what he can himself. Randy’s 1988 Chevrolet pick-up truck had been illuminating the check engine light from the time he purchased it at the farmer’s auction several years before. I think this truck was the second production year that GM had used a Throttle Body Injection (TBI) system; and because of the newness to us at the time, we techs were generally nervous and excited at the same time when we got to diagnose one. I think back to those days when new systems are brought to market today, like the Gas Direct Injection (GDI) systems that have arrived on the scene in the past few years. There’s no difference in my emotions each time. Are you that way, too?

(Image courtesy Mitchell 1) ECMs were simpler in the days of OBD I. Notice the single wire oxygen sensor. There was no heater circuit back in those days and it was not uncommon for sensors to go offline during an extended idle.

Randy cared little about the light and drove the truck a couple of years like that. He was using it as a work truck around the farm. Then he had an interested buyer for it so Randy decided to take his truck to have the codes read by his local parts supplier. He felt, since they didn’t charge anything for the service, that he might not need to bother me (read: travel to and have to pay me) for the same service. He also figured that once he knew why the light was staying on, it might be a simple fix (and that he could do it himself, thereby saving even more time and money).  Randy may sound like one of your “cheap” customers, but in reality, he has very little time in his day to waste (and he lives about 45 minutes away from me). His primary motivator was the time factor in this case.

The scan of the truck’s Engine Control Module (ECM) revealed a code 13, described in the service information only as “Oxygen Sensor.” No, it was not a P0013. Those of you who aren’t familiar with anything built prior to the second generation of On-Board Diagnostics (OBDII) might not be familiar with any non-standardized Diagnostic Trouble Code (DTC) format that contained as few as one digit. I use the term “privileged” loosely with my tongue in cheek, but we older folks were privileged to have been around before the second generation was even conceived. We had to diagnose those primitive — by today’s standards — systems using what few tools we could get our hands on.

The parts supplier was kind enough to provide Randy with a print-out of the DTC, a brief description of what the code meant and offered a list of parts prioritized by what they felt most often solved their customer’s Code 13 problem. Now, I don’t know from where the data was obtained that was used to prioritize that list, but imagine the tape register receipt on which it was printed — it was three feet long! I’m confident a majority of the customers who received that list probably resolved their Code 13 before purchasing and installing every possible option listed on the print-out. I also believe those customers most likely were not driving a high-mileage vehicle. If I recall, it was in the late-1990s that I saw the vehicle and at that time, the truck’s odometer showed more than 200,000 miles.

My readings were normal with the ground lead of my scope referenced to the exhaust manifold and my positive lead on the oxygen sensor wire.

In Randy’s case though, he went through the list methodically. He installed the oxygen sensor first. Notice I didn’t call it a heated oxygen sensor (HO2S)? Sorry, those hadn’t been in use yet on a production-line Chevy. Notice too, that I didn’t mention the location (like B1S1 or B2S2).  Well, that’s because they only had one O2 sensor on this vehicle! I said it was primitive, but if you believe it would be simple to diagnose because of its comparative simplicity, you’d be mistaken.

In the past, we couldn’t clear codes using a scan tool. The procedure to immediately clear the stored DTC, instead of waiting for the 50th key cycle to occur without the problem being recognized again, was to disconnect one of the battery terminals for at least 10 seconds. Randy followed the oxygen sensor replacement with the clear DTC procedure and excitedly started the truck. It was within moments the check engine light illuminated and completely eliminated any confidence that he had resolved the problem. Back to the auto parts supplier Randy went.

It was an indication of a lack of confidence in the quality of parts they carried when the store representative jumped to the conclusion the part that was sold to Randy was a “defect” and offered to replace it. Randy installed it right in their parking lot and got the same results that he got with the first one. Then he started working down the list of parts that could cause a Code 13 to set. He bought a few at a time, brought them home and after the replacement of each component, the clear DTC procedure was followed. He grew increasingly frustrated when every time he replaced a part and cleared codes, the problem would not go away.  Eventually, you guessed it; the threads in the battery gave way, requiring a battery replacement!

With my meter referenced to battery ground, I was reading a negative output from the same sensor that was reading normally a moment ago. It’s all about the meter lead placement and referencing the entire circuit path.

Now it’s mine

It was at this time that Randy admitted he had neither saved any money, nor any time by choosing the methods he did to address his check engine light. After calling me to verify I could (and would) take on the job, he trailered the truck to my shop. Upon arrival, he shared with me each of the events chronologically and reviewed with me the list he was given at the auto parts store. He was past the ECM, and was asking if I “had a PROM (Programmable Read Only Memory) ‘chip’ that would work in this vehicle.” For the benefit of those who don’t know what that is, we used to calibrate a module to a particular vehicle using replaceable components that were snapped into the module. Most were called PROMs. Today’s modules are calibrated quite differently.

It is my first step in every diagnostic challenge — visual inspection. I include a check of all fluid levels if the complaint requires the hood to be opened for any reason, as was the case with Randy’s pick-up. Nothing was noted to be low, but I did see a lot of evidence of rusty water stains on the right inner fender area that had come from the radiator at some point in time. Since it was safe for me to open it, I looked in the radiator and saw what I expected from an old work truck. I saw no evidence of any antifreeze and returned the cap from where it came.

I scanned the ECM and found the Code 13 (and no others) and at the same time, I reviewed the data list. The oxygen sensor voltage parameter was stuck at zero volts, even after running the truck a few minutes at around 2,000 RPM. I reviewed the DTC “Diagnostic Trouble Tree,” the chart of instructions written to direct technicians in a logical manner (hopefully) to where the problem exists that made the light turn on. It will usually include “branches” like an inverted tree, where if you answer “yes” you are directed down one branch, and if you answer “no” you’re directed down a different branch.

In the case of a Code 13, the technician is instructed to check only a few things. I don’t always follow these charts step-by-step, but instead use them as guides and then test circuits and components logically. Experience plays a big part in my decision whether I’ll use a chart or not.

(Image courtesy of Pete Meier) As it turns out, I was reading voltage produced by the cooling system. When the coolant mix gets too acidic, the cooling system becomes one big galvanic cell – just like a battery.

In this case, having read the O2 sensor data list value is zero volts, I went directly to the sensor to read the voltage it was producing. I attached my oscilloscope’s negative lead to the exhaust manifold then back-probed the sensor’s connector using my red lead. To my surprise I was reading a varying voltage between 100mV and 900mV! Convinced the sensor was working as designed, I felt it was time to check the voltage at the ECM. I placed the scope on the open hood facing the windshield (the ECM is located inside, forward of the glove box) and connected my ground lead to the battery’s negative cable attaching bolt (very close to the ECM).  Then I back-probed the purple wire at the ECM connector when I got settled inside the vehicle. I was seeing no pattern displayed at all!

So, armed with this information, wouldn’t you think like I did, that there had to be an open circuit between the sensor and the ECM? Well, the Ohms test showed the circuit was intact! What do you think I did now? Re-test of course, and I got the same results (of course)! It was time to step back and think about what I was doing and what I was seeing. Do you already know what I found?

Lead placement matters

I started thinking about how I was performing my scope tests and concluded the movement of my negative lead was contributing to the different readings on the scope. I installed longer leads so I didn’t have to keep moving it (the O2 sensor was on one side, the battery on the other). I prefer to keep my negative lead on the battery negative terminal whenever possible and this truck is a perfect example WHY.

(Image courtesy of Pete Meier) — You can measure voltage in coolant by placing your positive meter lead in the coolant itself. Don’t touch anything else though to avoid skewing your result. Voltage can be caused by an acidic mixture or poor electrical grounds and warrants further testing to isolate the real cause.

Without moving my negative lead from the battery I went back to the sensor. This time I used the scope’s “Signal Finder” function. It displayed one of the most beautiful images, an almost textbook picture of an O2 sensor, only the pattern was displayed between the (-)2.8-volt mark and the (-)1.8 volt mark. That’s right, it was operating perfectly, but it was well below zero volts! Testing at the ECM connector gave me identical results. How in the world can this happen? So, what do you think an ECM sees when a signal is below zero VDC? The answer is “zero,, just like it was displaying in the data list!

Why did I get these results? Remember my simple explanation of what a typical car battery is made of? Let’s apply the same principals here. There are dissimilar metals that coolant flows through. The cooling system had been badly neglected thus causing a very low pH, or highly acidic, coolant mix. Basically, instead of submerging metal in a solution, we had an acidic solution surrounding many types of metals producing voltage in the cooling system! In this case it was a negative voltage. Dipping my Digital Multi-Meter red lead into the neck of the radiator gave me proof.

Now, where do you suppose the ECM’s grounds terminate on this vehicle? The bolts that mount the thermostat housing to the intake manifold (through which all the engine coolant eventually passes) hold the ECM’s ground eyelets. Imagine how this change in the (perceived) ground circuit would affect all the sensors’ signals!

It took nearly the whole afternoon and an unknown number of cooling system flushes to reduce the cooling system voltage to below half of a volt difference to the battery negative. I eventually used baking soda in the radiator, a last resort, to neutralize the acid that took years to get so bad.

Now you know — there’s a close relationship between batteries and cooling systems! Keep those services done in a timely manner and leave the voltage production to those folks who build batteries!

Sponsored Recommendations

Best Body Shop and the 360-Degree-Concept

Spanesi ‘360-Degree-Concept’ Enables Kansas Body Shop to Complete High-Quality Repairs

ADAS Applications: What They Are & What They Do

Learn how ADAS utilizes sensors such as radar, sonar, lidar and cameras to perceive the world around the vehicle, and either provide critical information to the driver or take...

Banking on Bigger Profits with a Heavy-Duty Truck Paint Booth

The addition of a heavy-duty paint booth for oversized trucks & vehicles can open the door to new or expanded service opportunities.

Boosting Your Shop's Bottom Line with an Extended Height Paint Booths

Discover how the investment in an extended-height paint booth is a game-changer for most collision shops with this Free Guide.