Battery Diagnostics

Anawa · December 2023

I’m sharing this post I made on the Outback Forum. I think you guys may be more knowledgeable in this area.

This post is an update from an earlier one. I’m needing advice on the replacement status of a 1,100 Ah FLA battery bank.

The system is: 9.2kw array, Radian 8048A, 3ea. MN Classics, 1,100 ah Forklift battery bank. System was brought on-line in Nov 2017.

After having “burned” out 3 Kohler generators, a new one was installed on 12/19/23 (14RES). During the night of 12/26/23, the generator started at the 55% SOC rate and cut-off the charging at set-point of 78%. All of this was fine. The low cut-off set point was to minimize generator run time. The uncertainty as to the cause of the prior generator failures was the primary factor in the 78% cut-off.

During the installation of the new generator, the installer set the generator for an exercise-run using the internal Kohler programming system and not with the Outback Mate3. During this unanticipated exercise run on 12/27/22, this generator started running erratically and had to be eventually shutdown by manually switching off at the generator power panel. The repair technician found evidence of “windings” failure, but opined that Kohler did not consider this an equipment “failure” because the generator started and was providing power to the system. This is kinda odd and maybe will be a source for another post.

Here’s the really interesting part where I’m really needing advice. Let me begin by saying the batteries have had a rough life. For some reason and early on, it took us a long time to properly program the Classics to stop a random equalization charge. Needless to say, over the life of these batteries, they’ve been both undercharged and overcharged.

Here’s the battery symptoms observed yesterday as the technician was checking out the generator. The battery input from the generator was a steady 90 amps the entire time. The battery voltage reading was rising steady and the “battery High Voltage” alert on the Classics came on at about 59 volts. The absorb voltage is set at 58.4 V. The Mate3 was showing 83% SOC. A specific gravity reading was taken and reflected 65%. This great disparity was unsettling. The battery voltage kept rising steady and reached 60+V. We decided to shutdown the generator for fear that this ain’t a good situation and we would rather have the Classic’s do the charging the following anticipated sunny day (today).

Within minutes, the “high voltage” alert was off and the Classic readings on the battery were down to 56 or so voltage. This morning the OpticsRE reflected an average high during the generator run (14:00) to be 59.7V. At the 15:00 time mark, the reading was 51.5V. A drop of 8.2 V in one hour. The lowest voltage reading was 48.8V this morning at 06:00.

I’m thinking not much of this is good. Y’all let me know what you think.

Paul in Georgia

Vic · December 2023

Paul,

Briefly, it appears that your batteries were not fully charged, even at the end of that 90 Amp generator run, as evidenced by the battery charge current (assume that that 90 Amp number was 90 A DC battery current) did not diminish, as this charge continued.

When you took the SG reading, what was the SG of the electrolyte?
What device was used for this measurment?
What is the target SG, from the battery manufacturer, for a full charge?

I know nothing of a HV "alert" from Classic CCs. What is that exact message?
What Firmware rev are the Classic running? Is each Classic running the same firmware rev?

Thanks for adding this detail. AND, thank you for using Paragraphs in your Post, it really makes a huge difference in the readability of the info, and really governs the feeling for the reader of weather or not, one wants to try to decode info, from a SEA of text !! Vic

Anawa · December 2023

Vic,

When you took the SG reading, what was the SG of the electrolyte? Yes
What device was used for this measurment? HydroVolt
What is the target SG, from the battery manufacturer, for a full charge? 1.285

I know nothing of a HV "alert" from Classic CCs. What is that exact message? See Pic

Image: https://us.v-cdn.net/6024911/uploads/editor/nd/n8o8v5xk28vp.jpeg

Not sure on the firmware.

I’m sending some current data taken a few minutes ago.

Thanks
Paul in Georgia

Anawa · December 2023

Vic,

Sorry for the stupid Yes. The SG was about 1.240.

As of 45 minutes ago, both the Classics and the Mate3 were reading 100% SOC. The SG reading was 1.255. The Classics had been in float for about 65 minutes, I wasn’t sure how to re-start the Classics to continue a charge (sun is shinning), I turn on the Generator.

The pics below show the Mate3 readings:

I reset the Inverter charging parameters to 60.5 volts.

The current battery amp input is now 22.7 amps.

Paul in Georgia

Vic · December 2023

Hi Paul,

Thanks fo the detailed reply.

I will need to get to a better computer, to do a real Post. Pls gimme an hour, or so. Thanks, Vic

Vic · December 2023

OK,

First, the Hydrovolt is a good Hydrormeter.

That HV alert on the Classic/s just means that there is another charge source, that has raised the battery voltage, above the setpoint of the Classic, for the charge stage that that Classic had been in.

SOC readings are usually, guesses that CCs and bat monitors try to make. For Classics, with a WbJr battery current monitoring accessory, the SOC is set to 100% SOC, when the Classic that has the WbJr attached, ends Absorb (on its own, not from someone forcing Float), and goes to Float. These guesses are only as good as the settings entered by the user, and will usually drift away from actual SOC, when the battery has not gotten a "full charge", for some number of days, as you know.

Are your three Classics set up for, "Follow Me"?

It does look like your battery needs more charging. It is possible that there could be some hard Sulfation of the battery, but the battery might benefit from a reasonably-long EQ, at about 63-ish volts (probably temp compensated), perhaps a bit higher V if your system hardware can tolerate that Veq. You should really see the battery current, from the FNDC, and/or the WbJr diminish, as the EQ progresses. It is best to only EQ a FLA battery, when it is as fully-charged, as you can get it.

More later, probably have left out too many things, Vic

Anawa · December 2023

Hey Vic, got the batteries to full charge at 1.285.

Another question is: When the generator is on and the inverter charger is being used, can the controllers be operational also? If so, does the Outback inverter still control how much energy goes to the load and the batteries?

Thanks
Paul in Georgia

Vic · December 2023

Paul,

Great news, that you got that battery fully charged !

When the battery needs a Bulk charge, all charge sources will deliver their maximum, based on the input power available to them.

When the battery voltage setpoint has been reached, all charger will do their best, but it becomes difficult, due to Charge Controller (CC) voltage tolerances, voltage drop on the CC to battery cables, etc. But, generally some CCs may stop charging if their voltage setpoint is exceeded, etc.

I know nothing of the behavior of all of the OB hardware (FNDC, Optics RE, or whatever is, SkyBox, etc), but, often there is nothing that manages just where all available energy is distributed. However, the battery accepts its charge current based on the voltages from various charge sources, as does any AC loads on the inverters ... if this is not too ambiguous.

As you know, with off-grid systems with FLAs, we generally try to use a genset sparingly, by using it more for Bulk charge, and try to let solar, to finish charging the batts.

All, IMO, Take care, HNY, Vic

Anawa · December 2023

Vic,

Thanks for all your info and guidance. The Classics are all linked in “Follow Me” and seems to be operating as designed. As soon as we have another long sunny day, we’ll Equalize per your advice. I’m thinking to sample SG’s during the process to determine when to stop Equalizing. Amirite?

I’m very concerned about the extremely low readings I’m seeing on the battery during nighttime usage. Typically, the nighttime discharge is about 12-15%. That’s not what the concern, it’s the voltage reading going all the way down to 50.4V by morning.

Also, as soon as any charging source (controllers or generator) stops, the voltage readings drop precipitously. For instance, yesterday at 17:00 the voltage was 55.01V, at 18:00 its was 50.95V. The voltage declined steady to 50.4V just before dipping down to 48.6V (06:00) during the morning microwave and other wake-up activities.

A couple of questions: What do these low voltage readings on the batteries indicate as to life, viability, reliability, etc? Your recommendations to validate these readings with multi-meter or other device?

Our current thinking regarding the battery bank is to get as much life out of them that’s reasonable before replacing replacing them with LFP’s. Whether it’s 30 more days or 2 years, that’s what we’re thinking. Your thoughts?

Paul in Georgia

Vic · December 2023

Hi Paul,

Briefly, as we have a storm coming, and am still running around, getting more ready for it, etc.

First regarding EQ, normally for our FLAs, the primary reason to EQ, is when the measured SGs of any cells varies more than 20-25-ish points, it is that time. As the batts age, then the variation twix cells might never be as close as those noted above.

It might be fine to dp an EQ of the batt, soon, after a full charge ... and SG readings for all cells taken, for reference. For some FLAs the manufacturer might recommend periodic EQ, regardless of SG readings ...

Your voltage variations look fine to me, even for the early AM readings with heavier current draw.

Will try to add a comment or two, in a few hours, but, it looks like your batts are behaving fairly normally, now.

Take care, Vic

WaterWheel · December 2023

There are much more knowledgeable people on this site than me but as someone who has run FLAs the last 10 years I suspect the forklift battery is nearing the end of it's useful lifetime.

Where in GA is this system? I'm about an hour above Atlanta.

Anawa · December 2023

Hey Wheel,
Thanks for the post. I’ve noticed you’ve been around the forum for a few years now. Nice to have “local” folks checking in as well as all them West Coast Hippies and Canucks!!

I’m southeast of ATL in Carroll County, real near Historic Banning Mills on Snake Creek. Ya-whooo.

I am concerned about the batteries longevity, but I respect ole Vic’s battery wisdom and the NAZ forum in guiding us to get the best from what cha-got.

Paul in Georgia
Living the Dream and Praising the Lord

Vic · December 2023

Paul,

When charging stops on an FLA battery, its terminal voltage will drop, fairly rapidly. Generally, the higher this charge voltage was, compared to the Float voltage setting, the longer this voltage will be elevated. This elevated voltage is said to be the result of Surface Charge. The greater the loads on the battery during this period, the more rapidly will this elevated voltage descend to this compensated Float voltage to the Float V setpoint. Of course, the CC/s will maintain the Float V, it there is enough input power to maintain it. This is normal, and does not mean that there was any meaningful discharge of the battery. If there were no loads on the inverter, the Surface Charge will dissipate, usually in several hours, and then the chargers will then try to maintain Vflt, if there is sufficient input power to them.

To me, your voltages look about like those with my battery -- early morning (with a full charge on the previous day), before any charging, Vbat is about 50.3-ish, with about 80 Ah, or so removed from the fully charged batt, over night. The coffee maker, plus normal background loads on the batt, the Vbat is usually about, 49.5 V, with the WbJr current reading, about -23 Amps.

About EQ, since, usually, lift truck batteries are quite tall, stratification can be an important consideration. Acid in the Electrolyte isE significantly heavier than water, so acid tends to work its way to the bottom of the cells, which can cause more plate erosion of the lower part of the plates. This erosion can reduce mixing of this strong acid, back into the entirety of the electrolyte in the cells. SO, longer, more frequent EQs, perhaps at a bit higher Veq may be needed to keep the electrolyte fully mixed.

The above, is the main reason to only add DIstilled Water to the battery, followed by a mini EQ, or additional Absorb, perhaps at a bit higher Absorb voltage.

More later, but it seems that your battery might well be in reasonable shape than one might expect ... Vic

Photowhit · December 2023

I just skimmed over some of the replies...
A 6 year old forklift battery is just in it's teen years and you should expect many more years.
I may have missed it but what are your charging settings on your outback inverter?
a 50.4 volt reading in the morning may indicate some even somewhat minor load on the system. Unless your batteries are removed from the system and at rest for a while, voltage is a terrible way to measure state of charge (SOC), and while connected to a system should always be considered 'system voltage'.

804ah rings a bell, If you have a GB 24-85-13 battery, please understand that they over rate their batteries in the 20hr rate!
24-85-13 is;
24 2-volt cells
85 amps (6hr rate) per positive plate
13 plates per cell of which 6 are positive for 6x85=510amps at 6 hr rate.

They use a multiplier of nearly 1.6x to come up with 804ah at a 20 hr rate!
Realistically, a multiplier of 1.3 or 1.35 is commonly used for tall cell batteries.
I use the 24 volt version of the GB 12-85-13

Anawa · December 2023

Thanks for chiming in Wit! It’s been a very long time.

I also bought (2) GB 12-85-13’s based upon your recommendations. I then advised my nephew to buy the GB 1,100 Ah this post is all about! As I recall, some 9 years ago, you likened buying the forklift batteries as “training wheels” for us folks just starting the off-grid lifestyle. That is sage advice and worthy to all off-griders to consider.

I just last month replaced my GB’s with a Pytes 600 Ah LFP bank. My GB’s were struggling and the best SG I could get was 1.272-1.265 range after a very long EQ at about 62V+. I was struggling too with constantly checking water levels every 2-3 days, watching more of the cell casings bungling with the over-heating, and wondering how my wife was going to be able handle this kinda crap if something happened to me. I’m old and have a lot of land and a garden to maintain. Putt-sing around and worrying about FLA’s isn’t how I can spend my days. Besides…. I love being FREE from SG bondage!

I also remember you saying that the 20hr GB rating was off. But, at that time, you didn’t have a multiplier. What would you calculate the 20hr rate for the GB 1,100 to be? Also, I’m not suggesting that the low voltage of 50.0V during nighttime is a any kind of measurement for the SOC, my concern was the that huge drop in Voltage was an indicator of catastrophic battery failure, Vic has helped cured that concern for me by explaining how the FLA’s function.

Right now, I’m wanting to get some more life out of my nephew’s battery bank and any advice you boys can send my way, I really appreciate it.

And finally, am I thinking correctly that since we’re able to get the battery to 1.285 SG suggest there is more life years in the GB?

Paul in Georgia
Living the SG Free Dream!

Battery Diagnostics

Comments

Categories