Repair-It

Mother of god. I harvested my FSM for charging information.

There is so much gold in here. Here are the highlights:
General parasitic drain info. How to compute current draw vs battery amp-hour rating and how many days that draw can exist before causing a no-start. Temperature effects.

How to use the Tech 2 to monitor Class 2 and CAN communications to understand what modules may or may not be sleeping properly. Analog vs digital OnStar current draw profile. Apparently there is a sleep mode to force modules to sleep prior to some nominal 25 minute window where they may be active?!

This one is great. Basically GM visited different dealers and concluded that they suck at batteries. Not rotating stock so they'd end up with some old-ass batteries on the shelf, which a customer may end up getting stuck with. Not charging their stock batteries (yes, GM actually requires them to do this) and this doc even says the standard is 12.6V or higher resting voltage is required before delivering a battery to a customer.

All new batteries should be tested for a minimum of 12.6 volts before they are delivered to the
customer. Those that don't meet minimum voltage requirements should be charged using a
reputable battery charger / tender and re-tested before delivery to the customer . You may wish to
task all employees who sell a battery that they jot down the voltage as tested on the customer
work order or receipt. This will not only serve as a quality assurance step for your customer, but
help create the habit of testing all batteries delivered. Your goal should be to always provide the
customer with a premium quality, fully-charged battery and ready for service.

All about load shed. Looks like it has 3 levels of load shed and 3 levels of idle boost, depending on circumstances. Introduces the idea of a battery temperature calculation as well as amp-hour calculation. It also says voltage calculation but it is a little odd that it would need to calculate that...but yeah it'll even duty cycle the rear defroster and heated mirrors down from 100% to 80%, 50% and finally off when shit really hits the fan. The driver is not presented with any warnings until load shed mode 2 or higher. The car will quietly duty cycle the defroster at 80% for mode 1. It does other stuff not documented here I guess.

All about how the Instrument Panel Module and Rear Integration Module deal with body control.

Electrical Power Management (EPM)
Load Management
The power management function is designed to monitor the vehicle electrical load and determine
when the battery is potentially in a high discharge condition. This is accomplished by using a high
accuracy battery voltage and current reading by a current sensor as an indicator of battery
discharge rate. The following 6 levels of load management will execute in the load management
control algorithm when there is a high discharge condition:

Loads subject to reduction include the following:
• The A/C clutch
• The heated mirrors
• The heated seats
• The rear defog
• The HVAC blowers
• The interior lighting

Regulator Voltage Control (RVC)
Regulator voltage control (RVC) will result in the battery being charged at its optimum voltage for
improved battery life and state of charge (SOC), fuel economy and lamp life by lowering the
system voltage when the SOC is high. The electrical power management (EPM) algorithm in the
instrument panel module (IPM) will determine the optimum charging voltage, based on estimates
of its SOC and battery electrolyte temperature. Optimum charge voltage is defined as the battery
charge voltage that results in maximum battery life, while maintaining energy storage for engine
starting, discharge at idle, and parasitic loads. The optimum battery charge voltage will be
converted to a percent duty cycle command that will be sent to the ECM via serial data link. The
ECM will then place the 128 Hz pulse width modulation (PWM) duty cycle on the L line.

So there it is. Fuel economy, lamp life, and battery life.

This actually has to do with charging the battery with an external charger.

Important: The table is accurate to 10 percent only after the battery has been at rest for
12 hours.
4. Measure the battery voltage at the battery terminals. Refer to the following table to
determine the SOC according to the estimated battery temperature:

Then it has a table of battery voltage vs state of charge at 32F and 75F.

Battery Voltage % Charge at 0°C (32°F) % Charge at 25°C (75°F)
12.75 V 100% 100%
12.7 V 100% 90%
12.6 V 90% 75%
12.45 V 75% 65%
12.2 V 65% 45%
12.0 V 40% 20%
• A battery with a SOC that is below 65 percent must always be recharged before returning it
to service or continuing storage.
• A battery with a SOC that is 65 percent or greater is generally considered to be charged
enough in order to be returned to normal service or in order to continue storage. However, if
the battery is being used in slow traffic or with short drive times, or if the temperature is
very hot or very cold, the battery should be fully charged, to at least 90 percent, before
returning it to service or continuing storage.

• The battery charger capacity--The higher the charger amperage, the less time it will take to
charge the battery.
• The SOC of the battery--A completely discharged battery requires more than twice as much
charging time as a half charged battery. In a discharged battery with a voltage below
11 volts, the battery has a very high internal resistance and may only accept a very low
current at first. Later, as the charging current causes the acid content to increase in the
electrolyte, the charging current will increase. Extremely discharged batteries may not
activate the reversed voltage protection in some chargers. Refer to the manufacturer's
instructions for operating this circuitry.
• The temperature of the battery--The colder the battery is, the more time it takes to recharge
the battery. The charging current accepted by a cold battery is very low at first. As the
battery warms, the charging current will increase.

What's interesting about this though is that the car is trying to calculate all of this stuff.

Not too exciting but talks about amp-hours, reserve capacity and cold cranking amps.

Amp Hour
The amp hour rating of a battery is the amount of time it takes a fully charged battery, being
discharged at a constant rate of 1 amperes and a constant temperature of 27°C (80°F), to reach a
terminal voltage of 10.5 volts. Refer to Battery Usage for the amp hour rating of the original equipment battery.

Reserve Capacity
Reserve capacity is the amount of time in minutes it takes a fully charged battery, being
discharged at a constant rate of 25 amperes and a constant temperature of 27°C (80°F), to reach
a terminal voltage of 10.5 volts. Refer to Battery Usage for the reserve capacity rating of the
original equipment battery.

Cold Cranking Amperage
The cold cranking amperage is an indication of the ability of the battery to crank the engine at cold
temperatures. The cold cranking amperage rating is the minimum amperage the battery must
maintain for 30 seconds at -18°C (0°F) while maintaining at least 7.2 volts. Refer to Battery Usage
for the cold cranking amperage rating for this vehicle.

This one is interesting.

Diagnostic Aids
• The HVAC control module produces a parasitic draw of 33 mA for 3 hours then turns off.
• Be sure to rule out any possible obvious influences, such as customer error or aftermarket
equipment.
• Customer driving habits, such as regular short trips. This does not allow enough time to
properly charge the battery. Refer to Battery Description and Operation .
• Verify that the battery and charging system are in proper working order. Refer to Battery
Charging and Charging System Test .
• A battery discharging for no apparent reason while the vehicle is parked can be caused by an
intermittent draw, such as a module waking up, or a continuous draw, such as a dome light
or stuck relay.
• Some systems and modules such as OnStar®, and regulated voltage control (RVC), if
equipped, are designed to wake-up, perform a task, and go back asleep at regular intervals.
Refer to Body Control System Description and Operation for the system or modules
description and operation.
• Remote keyless entry (RKE) will wake up due to an outside input. Refer to Keyless Entry
System Description and Operation .
Important: The battery specification listed below is a generic specification. Refer to Battery
Usage when testing the battery.
• The battery run down time will vary depending on cold cranking amperage (CCA) and reserve
capacity (RC). If the CCA and RC are higher, then the battery run down time would be
longer. If the CCA and RC are lower, then the battery run down time would be shorter. The
graph below indicates roughly how many days a 690 CCA battery with at 110 min. RC
(60.5 AH) starting at 80 percent state of charge will last with a constant current draw until it
reaches 50 percent state of charge. Differences in battery rating and temperature will affect
the results.

Current Drain Days
25 mA 30.5
50 mA 16.5
75 mA 11
100 mA 8.25
250 mA 3.3
500 mA 1.65
750 mA 1
1 A 0.8
2 A 0.4

In short, it tells you how to compute what kind of draw would be acceptable based on number of days the vehicle could sit with that draw and still start. And it also basically tells you to pull fuses to isolate the circuit that could be drawing (sort of an old school sister process to go with the fancy Tech 2 approach at the module level).

But what's this?

• Some systems and modules such as OnStar®, and regulated voltage control (RVC), if
equipped, are designed to wake-up, perform a task, and go back asleep at regular intervals.

So that answers one question I had. I guess the system will periodically monitor voltage and maybe current during sleep? You'd think this car should really be aware of its own parasitic draws and charge the battery accordingly. Yet I'm sort of fighting the system it seems like - if it's broken I'd like to fix it. That's the whole point of this adventure.

This is what I was waiting for really.

Electrical Power Management (EPM) Overview
The electrical power management (EPM) system is designed to monitor and control the charging
system and send diagnostic messages to alert the driver of possible problems with the battery and
generator. This EPM system primarily utilizes existing on-board computer capability to maximize
the effectiveness of the generator, to manage the load, improve battery state-of-charge and life,
and minimize the system's impact on fuel economy. The EPM system performs 3 functions:

• It monitors the battery voltage and estimates the battery condition.
• It takes corrective actions by boosting idle speeds, managing the loads, and adjusting the
regulated voltage.
• It performs diagnostics and driver notification.

The battery condition is estimated during ignition-off and during ignition-on. During ignition-off the
state-of-charge (SOC) of the battery is determined by measuring the open-circuit voltage. The SOC
is a function of the acid concentration and the internal resistance of the battery, and is estimated
by reading the battery open circuit voltage when the battery has been at rest for several hours.

The SOC can be used as a diagnostic tool to tell the customer or the dealer the condition of the
battery. Throughout ignition-on, the algorithm continuously estimates SOC based on adjusted net
amp hours, battery capacity, initial SOC, and temperature.

While running, the battery degree of discharge is primarily determined by a battery current sensor,
which is integrated to obtain net amp hours.

In addition, the EPM function is designed to perform regulated voltage control (RVC) to improve
battery SOC, battery life, and fuel economy. This is accomplished by using knowledge of the
battery SOC and temperature to set the charging voltage to an optimum battery voltage level for
recharging without detriment to battery life.

The Charging System Description and Operation is divided into 3 sections. The first section
describes the charging system components and their integration into the EPM. The second section
describes charging system operation. The third section describes the instrument panel cluster (IPC)
operation of the charge indicator, driver information center (DIC) messages, and voltmeter
operation.

Want to highlight this:

The battery condition is estimated during ignition-off and during ignition-on. During ignition-off the
state-of-charge (SOC) of the battery is determined by measuring the open-circuit voltage. The SOC
is a function of the acid concentration and the internal resistance of the battery, and is estimated
by reading the battery open circuit voltage when the battery has been at rest for several hours.

The SOC can be used as a diagnostic tool to tell the customer or the dealer the condition of the
battery. Throughout ignition-on, the algorithm continuously estimates SOC based on adjusted net
amp hours, battery capacity, initial SOC, and temperature.

While running, the battery degree of discharge is primarily determined by a battery current sensor,
which is integrated to obtain net amp hours.

So it definitely has an ignition off monitoring mode, so that when the ignition is switched on, the IPM knows the SoC and uses that as a starting point for the charging cycle during the next ignition cycle. Really seems like the car has all the information and algorithms for this not to be an issue....unless something is broken. Definitely would like to verify that current sensor is working properly (or just get a new one). Also, it would be interesting if there was an updated firmware for that IPM that addresses this. I may look into that.

Battery Current Sensor
The battery current sensor is a serviceable component that is connected to the negative battery
cable at the battery. The battery current sensor is a 3-wire hall effect current sensor. The battery
current sensor monitors the battery current. It directly inputs to the IPM. It creates a 10-volt pulse
width modulation (PWM) signal of 128 Hz with a duty cycle of 0-100 percent. Normal duty cycle is
between 5-95 percent. Between 0-5 percent and 95-100 percent are for diagnostic purposes.

Engine Control Module (ECM)
The ECM directly controls the generator field control circuit input to the generator. The ECM
receives control decisions based on messages from the IPM. It monitors the generators generator
field duty cycle signal circuit and sends the information to the IPM.

The two modes are introduced. Charge mode and voltage reduction mode. When I describe what I see in system voltage, I'm basically describing the system going between these two modes.

Charging System Operation
The purpose of the charging system is to maintain the battery charge and vehicle loads. There are
2 modes of operation and they include:

The engine control module (ECM) controls the generator through the generator field control circuit.
It monitors the generator performance though the generator field duty cycle signal circuit. The ECM
controls the generator through the generator field control circuit. The signal is a 10-volt pulse
width modulation (PWM) signal of 128 Hz with a duty cycle of 0-100 percent. Normal duty cycle is
between 5-95 percent. Between 0-5 percent and 95-100 percent are for diagnostic purposes. The
following table shows the commanded duty cycle and output voltage of the generator:
• Charge Mode
• Voltage Reduction Mode

Commanded Duty Cycle Generator Output Voltage
10% 11 V
20% 11.56 V
30% 12.12 V
40% 12.68 V
50% 13.25 V
60% 13.81 V
70% 14.37 V
80% 14.94 V
90% 15.5 V

The generator provides a feedback signal of the generator voltage output through the generator
field duty cycle signal circuit to the ECM. This information is sent to the instrument panel module
(IPM). The signal is a 5-volt PWM signal of 128 Hz with a duty cycle of 0-100 percent. Normal duty
cycle is between 5-99 percent. Between 0-5 percent and 100 percent are for diagnostic purposes.

Charge Mode
The IPM will enter Charge Mode when the headlamps are ON, low or high beams. The voltage is
controlled to 13.4 volts if the system is below 13.3 volts or to 14.5 volts if the system voltage is
above 14.6 volts.

Voltage Reduction Mode
The IPM will enter Voltage Reduction Mode when the calculated ambient air temperature is above
0°C (32°F). The calculated battery current is less than 2 amperes and greater than -7 amperes,
and the generator field duty cycle is less than 99 percent. Its targeted generator output voltage is
12.9 volts. The IPM will exit this mode once the criteria are met for Charge Mode.

So that's interesting. The duty cycle is a direct voltage request? I find that odd. I would think the ECM would just have a PID loop on the duty cycle until it gets the voltage it wants at any given time. Instead, it seems like there is a duty cycle vs voltage mapping. This would suggest that an alternator malfunction could result in the wrong voltage being sent even if the ECM is commanding the "right" duty cycle. I don't think this is my issue because system voltages are basically exactly what is intended and out of spec probably throws a code anyway. The charge mode paragraph is not coherent though. But generally, there seem to be two main target voltages: 14.5V which I have seen very commonly, and 12.9V which I saw just today. I think the alternator is working fine.

Finally there is a schematic PDF that shows the current sensor is connected to the IPM. They communicate on, I assume, the low speed LAN, and the ECM controls the alternator. The ECM also may ask the IPC to illuminate the battery indicator or flash the battery saver active message on the DIC.

There, is that enough information?

So, massive brain fart. Well kind of. I don't think this is an AGM battery. I remember distinctly thinking I had done AGMs in all three Cadillacs. I don't think Advance even sells a Group 101 in AGM. That was the whole thing. The SRX got the first AGM but is a newer GM style, Group H6 in AGM. And in the CTS-V, which specifies the same older (and very uncommon) Group 101 of my STS, I converted to a Group 78 AGM (which is a very common size). The STS is still running a stock size Group 101 and I'm not sure ANYONE makes an AGM in that size. Page 1 of this thread clearly states I will probably switch to AGM next time. Whoops. Too many things going on...good thing I document things on the forum!

So this is a regular battery. I don't know if that totally changes any of the calculus here but considering I was about to go outside and try Adam's battery tester (hoping for accurate results while connected to the car), I need to know the chemistry and specs. Will report back. Hopefully everything is asleep.

It seems to be very slowly falling back to the behavior before. It seems like the starter is very slightly dropping in rpm, and I also noticed the first real charging behavior today. Not the full like 14.7 but 14.3V. In my mind it feels like the car wants to maintain something like 70% SoC and not really get it to 100%. However, I ran the connected test with Adam's tool which is not a full load tester but does some magic with microprocessors and stuff. I put in 650CCA, standard lead acid and it came up with this result: Engine bay was still warm. I think I will get a better result after it cools off, maybe tomorrow morning.

I also feel like this tool is more accurate when the battery is not connected to the car. I may try that this weekend (as part of my reset routine that I still want to do). But, preliminary results suggest the battery is in great shape and that it is fully charged. I wish I tested it before throwing the Battery Tender on it because I think it would have had a lower state of charge, despite driving the car all the time. It is not my imagination. If the battery was fine, and the charging system was fine, and my ONLY problem was a draw, these would not be the symptoms. Charging it with the battery tender would not have resulted in better cranking performance (can't get higher than 100% charged), and even the car "admitted" the battery seemed quite charged. So it made a difference. Need more data yet. But with another week I bet it would be back to what it was doing before.

Then I went out again and tested it after I opened and closed the door to wake stuff up. Very different result but also an invalid one. I let the loads settle down and tested again. Voltage back up so it was happy again. Waited even longer and the voltage rose a little more and gave this result. Managed to clamp my meter around the positive cable. I saw something in the 0.8A range. I opened a door which turns on interior lights, wakes stuff, and engages the air leveling compressor exhaust solenoid. Current spiked to 10.x amps! But then I watched as stuff slowly shut off. The interior lights alone seem to be worth nearly 5A. Then little by little it worked its way back down to 0.8A. Of course I need to catch it after it's really been sitting. A 0.8A draw would pull the battery down pretty quickly, like 1 day. To replace the current meter, I'd have to slide it off the end of the ground cable which conveniently goes right to the shock tower on the body. Looks pretty easy.

Again, the car continues to run longer in charging mode each morning. At the beginning, it stayed in low voltage mode for the entire drive. After a few days, it starts some charging. Now it is doing ~14.3 for a good 15 minutes (about half of my commute). As this progresses, the starter seems to go slower and slower. And slower is relative. I don't mean it barely cranks but it doesn't sound 100% to me.

So I let it sit all night, then this morning popped the hood and waited a few more hours for systems to go back to sleep. Here's the result of that: So first of all I disagree with the charger that 12.57V is 95%. I also ran it again: The table that I pasted before says 90% if it is 32F or 75% at 75F. It is 50 out now but was colder earlier so let's put it in the middle. Either way, not 95%.

I also ran the cranking test. It passed but no idea how good this is at....anything. That said if 10.47V is what the battery dropped to, I can't say I'm overly impressed by that result. Then I ran the charging test. Next thing to do is charge it again and run the test and see if any of this improves. If it does that means the car isn't charging the battery as good as it could be.

My instincts on this seem to be good so far. I disconnected the battery, and retested it. It's been sitting since yesterday afternoon but I drove to Boston on Friday and back on Saturday so the battery should have been fully charged. In fact the car was in voltage reduced mode for the entire return trip if I recall.

Anyway here's what Adam's battery tester reports now: Well that's a different result, eh? I thought I learned this lesson with the CTS-V, that the results are worthless while connected to the car, despite the fact that it has a setting for that condition.

So 12.49V. GM says at 32F, 12.45V is 75% SoC. Voltage is slightly higher than 12.45 and also it was cold last night so even though it is 32F now the battery is probably a little colder, so we'll say the tester is probably pretty close here. It does, however, still report a good battery. But look what happened to the capacity measurement. 74% battery health. It's like 6 months old, and the date code was pretty recent unlike the first battery I bought and only got 2 years out of. Now you can see why this isn't all in my head.

Anyway, it is charging now. Voltage was already over 13V and it was connected for under 10 minutes, so obviously it wasn't deeply discharged (nor should it be).

An hour later the charger is already onto the next mode which means it wasn't all that low.

According to Deltran, the bulk mode (the first mode) is for 0-85% SoC. So that means the battery probably wasn't too far below 85%. This checks out based on the battery tester result (when disconnected).

2 additional hours later, still in absorption mode. Absorption mode is kind of the opposite of bulk mode. Bulk mode (solid red light) just maxes out the charger at 1.25A I think. But as the battery SoC rises, eventually the voltage hits a threshold (temp dependent but in the high 14s). Once that happens, it switches to absorption mode (red + blinking green) which is constant voltage instead of constant current. It holds that and lets the battery take additional charge, from 85 to 100%. As the battery nears 100%, current draw decreases. Once it gets to a minimum threshold, it lowers the voltage to maintenance mode (solid green, no red) which is good for long term storage (low 13s typically but I think also temperature compensated).

Another 2 hours, still in absorption mode.

I think that battery should have been topped off after such a long trip.

Been scouring my FSM. Found some more gold.

Beginning with the 2007 SRX, I found some new data elements: And look at what it says about the 2008 SRX BCM (highlighted things that are related to the charging system): Holy crap. Including parasitic draw and sulfation (?!). Looks like the SRX actually uses a BCM. Well, so that's the SRX. Let's look at the STS.

The STS architecture is a little different. It has an IPM, an Instrument Panel Module. Beginning in 2008 I found this list: Not nearly as much stuff but does have some of the stuff I'd like, like battery current or other stuff. Problem is, I don't see these items on previous model years in the FSM. The 08 SRX has a ridiculous amount of stuff that would all be helpful for this exact situation. Of course, our 08 SRX has none of these issues, always cranks well...figures.

So the plan is wait for the battery to finish charging, figure out a way to get rid of the surface charge (or not), test it and record the results, and then reconnect. I will get out the Tech 2 and see if I can get it to tell me all the stuff the 2008 does by bringing up the 2008 STS and connecting to the IPM. Maybe, just maybe, it will give me all the data parameters. Either that or they changed stuff and I am SOL on that. It's dumb though because the 2005 clearly does all that stuff, they just didn't add the data parameters to the stream. I hope that's not the case.

2 more hours....still absorbing....been on the charger for 8 hours at this point. A topped off battery would not have needed this much charging.

Scratch that, finally finished. 8 hours.