Hybrid Diagnostics

I didn't see many posts on hybrid diagnostics on here and I thought a discussion might be a good idea. I'm not the most educated on these systems and have not recieved much in the way of formal hybrid training but I'll share what I have seen and learned so fart. In 2016 when I took the L3 ASE test there were only 1500 technicians in the US with that certification. I thought that was surprising but as I have gotten my hands on more and more of these vehicles I am starting to realize why. 

First, vehicles like the most common Prius don't seem to hold a high resale value. When the batteries fail it is often not worth replacing. These vehicles do have other failures but they are typically 12v related and diagnosed in the same manner as a gasoline engines electronics system.

Second, the hybrid systems themselves are somewhat self diagnosing. For example the P0A80 code essentially tells the technician to replace the battery. Even diagnostics on the inverters is typically simple circuit checks before replacing the component as a whole. If you're like me trouble tree diagnostics is a last resort. I want to know what the function is of the circuit I'm testing, what I should expect to see, and why. Because most of the electronics in these systems is internal and the manufacturers don't want us opening these units up it is hard to for someone like me to get a complete understanding of these systems. You will often find the term "Toyota's intellectual property" in hybrid code descriptions and service info.

Third, new replacement batteries are only available from the dealership and are very pricey. You can find Chinese knockoff cells but I am told they are more risky than used cells at the same price. 

Fourth, safety. Nobody wants to get electrocuted. Orange wires, lineman's gloves, and shepard's hooks can make hybrid service a little intimidating.

On the southern east coast hybrid vehicles aren't as trendy as the rest of the country but we are seeing more and more of them every year. Last year I decided I wanted to get my hands on more of them and learn the systems a little better. I bought a non running GS450h with a dead HV battery, a missing 12v battery (which has an integrated temperature sensor and is dealer only), and broken direct injectors. (the story is the vehicle wouldn't start. The technician put a 12v battery in it sending the old one in as a core. Since it still would not start they decided it was the starter and decided to check it. When they couldn't find it they decided it was under the intake. When they removed the intake there was no starter, and the injectors somehow got the connector ends snapped off.) From there the vehicle sat torn apart for a few years. On a side note this is also one of those vehicles with direct and multiport injectors. At some point I would like to put something together on how those systems work together.

So I get this vehicle as a bucket of parts. We have a guy in town that "rebuilds hybrid batteries". I dropped the HV battery off with him while I reassembled the intake. I got the battery back a few days later with "2 new cells and a level charge" and installed it. It cranked and after a few tries the engine started up and ran. Immediately, the dash lit up with P0A80 Replace HV battery pack. I snapped some screenshots of the battery voltages which I thought didn't look good and now that I've been working on these a bit more I now recognize as awful. It is important to note that when the engine is running or is in regen mode power is being sent to the batteries which makes the data pids look fairly even. Battery block data is most useful when the batteries are at rest or under load.

Before I go any further I want to talk about the cells. Specifically the cells in used by Toyota and GM. These cells have a nominal voltage of 7.2 volts but that voltage can vary and be high or low even with a healthy cell. In this Lexus there are 40 cells. The module can only read the voltages of 2 cells together which is puts into a data pid that it calls a block. so on this Lexus there are 20 blocks for 40 cells. 

I did some research on what the cells are supposed to look like and some of the code set criteria and quickly understood that this was going to be quite a challenge. These cells not only have to be of good health but also balanced meaning a code will set if the difference in block voltages is beyond a predetermined voltage. I researched different chargers and ways to charge the individual cells but the best option seemed to be to purchase a rebuilt battery. I researched high voltage chargers and also found plenty of forum information on people taking RC battery chargers and "cooking" and draining these batteries repeatedly to get them back to health. I doubt that would have been successful given the condition of my battery and even if so I would have been more interested in coming up with a format that I could duplicate and make profitable down the road. No matter how I sliced it rebuilding batteries wasn't going to be a profitable option so I did some research on battery rebuilders and chose a smaller company with good reviews, mainly because they were one of the few that actually had a GS450h battery in stock. This pack has 40 cells and is fairly different from the typical Prius packs although it shares the same cells.

With a rebuilt pack installed the vehicle started and ran great. I have since had the P0A80 code return a few times. The battery I purchased came with a year warranty that does not cover shipping so the best option when this happens is to diagnose the specific cells themselves and ship them back for replacement. This is a fairly easy task until you have to remove the pack and disassemble it. The first step is to check freeze frame data on the code. Here is that data from a fault I had in May with battery block 1. I have since replaced those cells and here is another screenshot of a fault I had recently where battery block 15 is low.

In both of these screenshots it is very easy to see the bad block with low voltage. I usually like to drive the vehicle and monitor the blocks to see if there are any other blocks with issues and verify that the block identified in the freeze frame does in fact have lower than the rest voltage. After all tearing into some of these packs is a little labor intensive so its worthwhile to spend a little extra time in the scan tool. From there the next step is to open up and disassemble the battery. HV gloves are important. With the HV service plug removed handling the battery is safe. I encourage anyone going inside of the battery to take voltage readings along the bus bars and at the relays to help gain a better understanding of where the circuit is interrupted by the service plug. Once inside the battery and comfortable individual readings can be taken on each cell with the bus bars removed. Typically, a bad cell is easily identified within the block by checking voltage. In the case of the fault I had in May the vehicle has sat for a while before I had a chance to get into the cells which, in my opinion gives the cells a chance to balance out. At this point the bad cell showed 7.2v while all of the others showed 7.7 -7.8v. Block one on this battery is the passenger side but if I wasn't sure I could measure all four outside packs and determine which cell has low voltage. From there the cell can be replaced. In my instance the battery rebuilder that I purchased the battery from takes the ID number from the battery and tries to match it with a comparable cell so that it is closely balanced to the rest of the pack. There are resistance and other measurements that they use... or at least claim to use as I am replacing cells in this rebuilt pack fairly often, to determine the health of the cells that I am not completely familiar with. I mention this because you can find used cells on Ebay and Amazon. Because these cells must be balanced it is possible you could replace a specific cell that was failing in your customers pack only to have the light come back on because the cell you bought was in better shape than the rest of the pack. For this reason I wouldn't encourage individual cell replacement unless you're like me, married to a 1 year warranty battery! In the future I think I will go with a better known rebuilder purchased through a parts supplier that pays labor claims.

Some other things I wanted to touch on:

I want to point out the screenshots that show 17 battery blocks. In the Snap On software only 17 battery block pids are available. Using the Techstream software you can see all 20. This is one of the shortcomings of aftermarket tooling.

There are a ton of data pids in battery modules. There are resistance pids however I have never seen them differ from .2 (if I remember correctly). One of the batter rebuilders explained to me why this measurement in the scan tool was not accurate but I'm not even going to pretend like I understood it. I have not found SOC (state of charge) to be useful in diagnostics unless the battery is very dead. Different than a 12v battery it is normal to see the SOC swing high and low drastically. In a 40 cell battery where each cell may drop down or up a volt from 7.2 you will see an 80v difference which might be alarming to a technician. The important thing is for the battery to not get low enough to not be able to crank the ICE. Many of these vehicles have ways to charge or jump the HV battery, like this Tahoe with a weak HV pack that would not start the vehicle until after the cells were brought up in voltage. There are also some that will require the battery to be removed and bench charged in this instance. A HV pack with one or two bad cells will not typically drain the entire pack. A vehicles HV pack usually gets to this point because it either sat for an extended period of time or because there is an issue with the ICE. Many times the HV battery in this condition will need replaced so it is important to try to determine what caused the failure before getting the customer to commit to and expensive battery replacement. 

In my case I don't drive that hybrid very often which is not good for the battery, the same as it is for a 12v battery. I have heard of taxi companies are getting 300k + miles out of packs that are almost constantly in service which further supports the idea that the battery must be charged and discharged to maintain longevity. This is important to consider when selling a hybrid battery to a customer who's hybrid is not a daily driver. We should be questioning our hybrid vehicle customers a little differently than the rest.

On the other hand, a vehicle that comes in with a battery pack too weak to crank the engine may also have an engine fault that caused the battery to become drained. This is where it is VERY important to check for any engine related codes and not just clearing them. I will say that the majority of hybrid vehicles I've worked on seem to set plenty of erroneous codes that can make diagnostics difficult. Sometimes a code clear is necessary to try and determine what the actual fault may be. We also see codes set in multiple modules which can be intimidating for technicians. A HV fault will often set codes in multiple modules for example in the power steering controller you may find a code for a 42v system fault. These set in the same way a traction light will set a code telling the technician that there is a fault in an engine control module that may affect performance of the traction system. When I'm presented with 20+ codes in a code scan through all modules the first thing I want to do is clear them. In this instance if the ICE will not start I may get the customer to commit to replacing a $2500 HV battery only to find out that there is a dead cylinder that may have been the cause of the battery failing. Documenting codes is critical.

One last thing I would like to mention is that many of these battery modules have updated software. I can only assume that this software opens up the fault setting criteria but I really am not sure. One very interesting thing I noticed in a TSB for my Lexus is a procedure to reprogram the HV ECU and go through the HVAC unit and make changes to the automatic intake control and smog sensor sensitivity adjustment. I can only assume that is either because this vehicle is not equipped with a battery air filter and it wants to keep recirculating air through the cabin filter or because the cabin is too tightly sealed and there is a need for increased airflow. I would assume the latter as the smog sensor is turned to a lower sensitivity which I would assume keeps the system off of recirculate more often.