Mixing Batteries on Float-- Mex Test Update 2

[BFL13]

UPDATE 1 on 23 Dec post
UPDATE 2 on 25 Dec post
----------

Mex is driving me crazy! He said do not mix my used 12s (Scrubber batteries, he calls them--humph! They are respectable golf car batteries I'll have you know. Humph!) with my regular bank of 6s or they would evilly interact, and I would ruin my good 6s. I asked if that were still true if they were all on a Float charge.

Mex said they would still have evil interactions even when all on a Float. So I have gone ahead at some inconvenience to keep the used 12s and the good 6s apart.

Now all of a sudden, Mex pipes up in another thread with this!!!

MEXICOWANDERER wrote:
Batteries should be separated after the charging protocol. Connect all batteries together and apply 1 nominal 14.2 charging volts and there is no way any of the batteries can detect it one battery or a hundred are online, or the chemistry of any of them. The only requirement made of a charging system is that it not under or overcharge the battery. Voltage requiments for flooded 5% antimony batteries and pure lead AGM is similar enough for multiple warhead work.

It's when the charging ceases that the problems arise.

So go ahead and use a smart solenoid to parallel charge the batteries. I have done this too many times to count.

What is going on here? Are there evil interactions or not as long as they are charging or on a float? Does this leave the time they are discharging together as when the evil interactions get going?

If they are banked on solar, they discharge at night and charge during the day. So the evil interactions come out after dark? Can I just call Ghost Busters?

[NinerBikes]

BFL13 wrote:
I have done some 10 hr capacity load tests on these T-1275s using their 20hr rate. Last time it took 9 hours to get down from full to 50% (as measured after the test using voltage and SG) and AH capacity came to 270ish out of 300 (by Trimetric) rated on the pair (150 each) so that would be 270/300 = 90%. (matches the 9hrs out of 10 too) 2.5 year old batteries.

Looks like I can float them and by that suffer antimony poisoning or else not float them and get sulfation setting in from their voltage dropping too fast (when SG catches up falling). The only way out of this is to do neither and just go camping and never come back! 🙂

Assuming your T1275's were 2 years old when you got them, and had what 400-600 or more charge cycles on them on golf carts, how much life can be expected at 90% or 80% capacity, in terms of recharge cycle from 40-90% with occaisional top off to full 100% float cycles?

From my perspective, if the T1275 I am to receive has 90%, that's still 135 AH and if 85% 127.5 Amp hours, still more than the group 24 75AH that's a year old now.

Does daily solar charging while camping mask or hide a lot of the losses between cells when you have one bad cell?

[BFL13]

I have done some 10 hr capacity load tests on these T-1275s using their 20hr rate. Last time it took 9 hours to get down from full to 50% (as measured after the test using voltage and SG) and AH capacity came to 270ish out of 300 (by Trimetric) rated on the pair (150 each) so that would be 270/300 = 90%. (matches the 9hrs out of 10 too) 2.5 year old batteries.

Looks like I can float them and by that suffer antimony poisoning or else not float them and get sulfation setting in from their voltage dropping too fast (when SG catches up falling). The only way out of this is to do neither and just go camping and never come back! 🙂

[MEXICOWANDERER]

You can certainly use simple amp hours to establish a basic maintenance recharge curve.

Using only square wave resistance testing is sort of like determining the top speed of a car by how splattered bugs get hitting the windshield. True testing of an accumulator has to involve actual performance capacity as compared to lifespan.

A 100 ampere battery than exhibits 60% of usable amp hour capacity after 300 cycles is a hell of a lot better battery than a competitor than has 55% after 180 cycles.

Take note of the comments about antimonial poisoning being caused by constant float maintenance. The esteemed writer however neglected o take into consideration that float voltage values can be adjusted. Absolute minimum voltage needed to maintain OEM electrolyte density is best, followed by periodic top charging.

[BFL13]

Mex, nothing is ever simple to simple folks! 🙂

I think you want to compare the amp hours taken to get from say 12.5 to 14.4 using straight amps and time (or a "corrected" Trimetric AH figure taking out their heat factor) with some sort of standard amount that it should take?

We are looking for an unusual heat loss amount?

The low charging amps (1-3amps) for the test is to preclude voltage "spiking" on start-up so you get a smooth rise in voltage?

EDIT--thanks for the article explaining antimony poisoning.

[MEXICOWANDERER]

HERE IS A RATIONAL DISCUSSION OF ANTIMONY AND OTHER PLATE ALLOYING MATERIALS


LEAD-ANTIMONY, LEAD-CA
LCIUM, LEAD-SELENIUM, VRLA, NI-CD.
WHAT’S IN A NAME?
M. S. (Steve) Clark
Knoxville, TN 37922
ABSTRACT
Users are continuously bombarded by manufacturers and their re
presentatives extolling the virtues of their products. This
includes not only the products, but also th
e technologies employed. Without extensiv
e research it can often be difficult for a
user purchasing a new or replacement battery to separate the marketing hype from the facts.
Among many U.S. manufacturers the current buzz word is lead-selenium. Most U.S. manufacturers want us to refer to these
batteries as low-antimony designs. The reason is that antimony is
the primary alloy metal and selenium is a grain refiner. In
truth, the same arguments can be applied to nearly all U.S. pr
oducts where tin is the primary
alloy metal and calcium is used
as a hardener. The same is true of one U.
S. manufacturer who markets their VRLA designs as “pure lead” when in reality it is
a lead-tin alloy.
The attempt to paint lead-selenium with the same brush as lead-antimony makes sense in marketing space. The majority of all
batteries sold are for standby service. Si
nce lead-antimony designs are not suitable for standby service, this is done in an
effort to promote lead-cal
cium for its superior standby characteristics. Just
as lead-selenium is not the perfect battery for
every application, we shall se
e lead-calcium has its own issues that are generally “overlooked” during marketing.
In simple terms, if the perfect battery existed, then we would not have the number of battery companies, types and
technologies that exist. It is the job of the user to find th
e best battery for the applicati
on with the minimum life cycle cos
t.
What this paper will attempt to do is to separate fact from
fiction and present a reasonably unbiased look at the available
technologies and how they fit into typical applications.
INTRODUCTION
At Battcon 2008, two papers were presented that discussed the relative advantages of Lead-Calcium technologies over Lead-
Antimony and Lead-Selenium
1,2
. Another paper was presented extolling the virtues of a new Absorbed Glass Mat/Gelled
Electrolyte hybrid valve-regulated-
lead-acid (VRLA) design as the solution to all our battery needs
3
. The purpose of this
paper is not to challenge what was presented, but rather to lo
ok at the relative advantages and disadvantages of the various
battery types available today and how to select a battery based
on the application. The primary focus of this paper is on
stationary batteries used in standby and cycling applications.
BATTERY HISTORY
The most common battery used today has been in commercial use for over 130 years. First demonstrated by Gaston Planté in
1860, the venerable lead-acid battery is still the mainstay of ener
gy storage. Over the years there have been many evolutions
in the technology, but the basic chemistry has not changed.
Lead-acid battery physical plate designs ha
ve changed from solid lead to include Ma
nchex, pasted and tubular plate designs.
Separator technology has gone from wood to natural rubber, synthetic rubber and fiberglass and other synthetic fibers. Plate
chemistry has changed from pure lead, to include lead-antimony
, lead-calcium, lead-selenium (a
nd its relatives) and lead-tin.
The old open tank and glass battery jars have been replaced by
vented cells in various plastic containers and valve-regulated
designs.
12 - 1
The most recent lead-acid battery technology introduced was
VRLA. Despite initial claims to the contrary, early valve-
regulated designs suffered from short lifetimes and low reliability
. As a result, designs have evolved rapidly. We have seen
changes in plate alloys, internal arrangement, post seals, the addition of catalysts,
the removal of catalysts, etc. all in an
attempt to improve the longevity and reliability of the cells.
At the end of the 1800’s, the nickel-cadmi
um battery became a viable product. While
U.S. users are most familiar with the
small AAA and AA batteries, flooded Ni-Cd designs have been in commercial use in Europe for over 100 years. Even though
Ni-Cd has many advantages over lead-acid, U.S. users have only recently started embracing this technology.
More recently, new battery technologies such as the flow battery and high temperature sodium-sulfur have been developed
and deployed for large energy storage applications.
The new comers to the stationary battery market are large
format nickel-metal-hydride and lithium-based chemistries for
stationary applications. Battery companies around the worl
d are in research and development on large format lithium
batteries for stationary applications.
The first commercial lithium based stationary battery was the Av
estor lithium-metal-polymer battery. The battery is currently
unavailable, but may return in the future. Other companies have subsequently entered the market with lithium-ion type
batteries. The main draw back to lithium batteries is the need fo
r very exact charge control. The other draw back is that all
components of the battery are hi
ghly flammable and manufacturin
g tolerances are critical to prevent catastrophic failure.
The need for water cooling systems for nickel-metal-hydride batteries operating in a continuous charge mode seems to be
slowing development in this area and further development is not expected near term. This technology seems more at home in
cycling or transportation applications such as hybrid vehicl
es where the battery does not operate in a continuous charge
mode.
Changes in battery technology have been spurred by the desire to improve over one or more characteristics of the original
lead-acid battery. As each change in lead-acid battery technology has achieved performance improvements, it has also
introduced undesirable characteristics. The development of alternate technologies seeks to address the inherent limitations in
lead-acid technology but as the
technologies are deployed, we fi
nd that each has its pros and
cons and so the search for the
perfect battery continues.
The advanced battery technologies discussed above have been covered in multiple papers at Battcon and the expertise in
these areas is still primarily the province of the manufactur
ers. The focus of this paper on the three most common
technologies, the venerable vented lead-acid ba
ttery, the VRLA battery and the Ni-Cd battery.
LEAD-ACID BATTERY TECHNOLOGY REVIEW
Plate Configurations
There are five basic plate configurations
used to produce lead-acid batteries
1.
Pasted – The active material is contained in a supportin
g grid that provides the current path (Faure-1881)
2.
Tubular – The active material is containe
d in an insulated electrolyte permeable tube through which extends a lead alloy
rod that provides the current path. Tubular plates
were originally referred to as “ironclad” plates.
3.
Planté – A solid lead plate where the activ
e material is formed on the surface of
the plate (Gaston Planté 1860). Modern
plates are normally scored with
a grid to increase surface area.
4.
Manchester – An alloy plate, either lead-antimony or lead-calcium, with round openings into which are inserted rolled up
corrugated pure lead strips. The active mate
rial is formed on the corrugated lead
strips. Batteries using this plate design
are typically referred to as Manchex batteries
5
.
5.
Round cells – Pure lead conically
shaped grids are stacked horizo
ntally in a cylindrical jar.
Pasted and tubular plate designs dominate th
e market. Round, Planté and Manchester pl
ates are limited to use in pure lead
batteries which have a very small market share.
12 - 2
Today, the most common negative plate is the pasted plate. Even if a battery is marketed as a tubular plate, Manchex or
Planté battery, it nearly always has a pasted negative. The reas
on is simple economics. Pasted plates are inexpensive and easy
to manufacture.
The solid lead Planté and the round cell have the longest life and
are the most resistant to damage by elevated temperatures.
The Manchester plate is a hybrid design. Wh
ile typically marketed as a pure lead desi
gn the grid is normally either a lead-
antimony or lead-calcium alloy. The gr
id alloy results in a Manchex battery ha
ving operational char
acteristics somewhere
between a Planté and a pasted plate design using the same grid a
lloy. The advantage of Manchester plates is the larger surface
area for active material formation which gives them larger cap
acity and higher current capabilities than a similar size Planté
battery.
Pasted plates are the most commonly
used plates. The aging characteristics
and resistance to damage by elevated
temperatures are directly proportional to th
e thickness of the plate grid and to the type of alloy used in the grid. In pasted
plates the lead oxide is mixed with expanders and binding agen
ts to improve utilization of the active material. The paste is
then forced into the plate by a rolling mill.
Tubular plates are composed of a series
of small vertical pockets of non-conductiv
e electrolyte permeable materials. A thin
lead-alloy rod runs down the center of each pocket and the pockets
are filled with the active material before they are sealed.
The life and resistance to elevated temperat
ure damage of tubular plate designs depends on the thickness of the rod, the alloy
type and the ability of the pocket material to resist damage from
internal pressure. Tubular plates are typically filled with d
ry
lead oxide powder without expanders.
Plate Alloys
There are five basic plate alloys used today.
1.
Pure Lead used in standby long duration batteries with low current demands
2.
Lead-Antimony used for cycling applications and often for heavy equipment starting batteries
3.
Lead-Calcium (dominates the U.S. market
) for flooded standby and VRLA designs
4.
Lead-Selenium (dominates the European market)
for flooded standby and cycling applications
5.
Lead-Tin for VRLA
Pure Lead
The pure lead battery has the advantage of being very long lived
and reasonably resistant to da
mage by elevated temperatures
and over charging. However, it has very low power density and is not able to produce large currents.
There is only one pure lead VRLA on the market, and it is limited to very small sizes.
Note: There are companies that market “
pure lead” VRLA batteries in large sizes. The “pure lead” designation in this case
refers to the purity of the lead used to produce the battery
4
. In reality, the batteries use a lead-tin plate alloy and pasted plates.
Lead-Antimony (Antimony c
ontent greater than 2%)
In an effort to improve the power density and current capability
, early developers experimented
with different plate designs
and types of alloys. The first successful
alloy was lead-antimony. The addition of
antimony makes lead stronger and so
allowed for taller plates. It also accelerated the de
velopment of pasted and
tubular plate designs.
Today, antimony alloys dominate the market for cycling applications and due to ease of production are common in
developing nations. The dominance of antimony alloys in the cyc
ling market is based on the fact that the alloy is extremely
resistant to distortion or damage from repeated discharge and recharge cycles. In our daily lives we encounter these most
frequently as “deep cycle batteries” fo
r boats, golf carts, fork lifts etc.
12 - 3
Lead-Antimony alloys are not well suited for stand-by service. The phenomenon of antimony-poisoning where antimony
from the grid alloy forms small discharge points on the negative
plate surface is a direct result
of continuous charging. This
results in a continuously increasing float current and water
consumption over the life of the battery. The rate of antimony-
poisoning is directly related to the op
erating temperature, chargi
ng voltage and the antimony content of the alloy.
The decline in production of antimony designs in Europe and the U.S. has been driven by the steady increase in reliability of
the electrical grid. With increased reliability of the normal power source the number of discharge and recharge cycles drops.
This results in the battery spending the majority of its time
on float charge. Under these conditions, antimony-poisoning takes
over and causes a rapid increase in maintenance and shortened battery life.
The high float current of lead-antimony alloys makes them unsuitable for use in VRLA batteries.
Lead-Calcium
In 1935, manufacturers began the successful development of lead-calcium alloys
6
. Calcium alloys are not susceptible to a
similar phenomenon like antimony-poisoning and so perform better on continuous charge in standby applications. Lead-
calcium alloys currently domi
nate the American market.
The major disadvantage is that lead-calcium
alloys are unable to withstand a large nu
mber of discharge and recharge cycles.
As such they are unsuitable for cycling service. Even repeated
shallow cycles (less than 20%) can rapidly age the battery.
Calcium does not readily mix with lead making lead-calcium allo
ys difficult to manufacture. If the process is not controlled
properly, the alloy content between plates ca
n vary. This can result in some of the grids having too high of a calcium content.
Calcium tends to deposit at the grain boundaries and over time co
rrosion of the calcium results in plate growth and ultimately
determines the life of the battery. If the content of the calcium
is too high, grid corrosion is accelerated which results in
accelerated plate growth, shortened life and
in extreme cases, the grids can grow to
the point they crack the container.
Another issue a manufacturer faces is large variations in
individual cell voltage (ICV) during float operation
9
. Lead-antimony
designs typically float within a very narrow
ICV range. This is most often attributed
to the relatively high float current of
lead-antimony alloys and the fact that antimony stabilizes the po
larity of the positive plates. Lead-calcium alloys have very
low float currents and it is believed the low float current and lack of antimony are responsible for the large variation in
negative plate polarization that causes large variations in ICVs.
In extreme cases ICV variation can result in a gradual loss
of capacity in low voltage cells. To achieve acceptable ICV
variation, some manufacturers add platinum salts to the high voltage cells in a string to depolarize the negative plate. The
user needs to be aware of this fact, because after about 8 year
s the platinum becomes locked in
the negative plate. If a new
cell is added to an existing string that is more than 8 years old and platinum was added to the new cell, then new cell ICV
may be below that required to maintain charge and the cell will
sulfate. If platinum was used in manufacture and the battery
is more than 8 years old, replacement cell orders should specify that no platinum is to be added to the cell.
The limited cycle ability and ICV variations have driven near
ly continuous development in lead-calcium alloys. Gone are
binary alloys composed of lead and calcium. The primary cons
tituents of today’s alloys are lead and tin (tin content is
typically less than 2%). The calcium content has been remained at less than 0.1% and is typically around 0.05%. To improve
the alloy properties, many modern alloys add aluminum to the mix as a grain refiner.
The extremely low float current of lead-cal
cium alloys made them the first choice for VRLA battery designs and they are still
the most common alloys used today.
Lead-Selenium (Antimony content less than
2%) also referred to as “Low-Antimony”
While American manufacturers focused on lead-calcium alloys for standby applications, European manufacturers developed
low antimony alloys. In order to reduce the antimony-poisoning issue and improve standby operation, developers reduced the
antimony content of the alloy. In order to reduce the antimony
-poisoning rate to an acceptab
le level for standby service,
developers were forced to reduce the antimony content to less than 2%.
12 - 4

[MEXICOWANDERER]

To make this understandable as possible, the easiest formula is to COMPARE your battery with a new battery. But with like construction (% antimony) I look for similarly inefficient recharge kWh PERCENTAGE. I.E. 20 amp hours hours for restoring a 100 amp hour 5% antimony battery is the same as 44 amp hours for a 220 amp hour Golf Car battery. It's the percentage that matters. The total loss during the period matters the most.

[BFL13]

Thanks Mex. So it is safe (even if not ideally efficient) to parallel them with each other and even with a bank of 6s?

I will run this another week, then I can try that recharge test if my equipment will suit. (Batts are in the garage at 35-40F if that makes any difference.)

I don't have a Killawatt gizmo like some people here have. I do have a Trimetric set to record amp hours. I can use a low amp charger to raise the battery voltage from whatever in the 12s to the mid 14s and note how many AH and the time that took (subject to the Trimetric's default allowance for heat loss)

Is that the idea? If so perhaps some scientist can translate that into that into KWh. Then what? What is good and bad wrt to the battery's 20hr rate capacity and the recharging amps used? (150AH for a single T-1275)

Now I have to read up on antimony poisoning of negative plates! Yipes --how long will they live with that going on? Use em or lose em time. 😞

EDIT--Salvo, I don't have much luck with battery resistance testing based on past attempts. I can't organize my measurement timing right and get enough samples up and down like you explained is needed, to get anything but a very rough R. Might try again. I still have your instructions.

[Salvo]

I bet the battery internal resistance will also tell a good story.

Load battery with several 2 to 4 amp loads. Toggle one of the loads on/off. Measure battery voltage and current.

R_bat = delta V / delta I

Sal

[MEXICOWANDERER]

Ain't it fun? Welcome to the world of trends-and-tendencies.

A >.200vdc cell slump is the culprit that would cause an internal draw, I didn't see that kind of value in your report. But as long as you are in-test-mode, let them set for another week and see what happens. You have great quality acid and what I suspect is a measurable amount of antimony poisoning of the negative plates. There is no cure for that, sadly.

Batteries that droop to 12.4 volts after 2 weeks in 20C temperatures, are energy wasters.

Battery OEM's hated me when I hit them with these kinds of tests. They want smileys and pure BCI parameters for tests.

Basically, a pure voltage sag test that you're doing BFL13 can be augmented with a controlled 1 amp recharge rate, measuring kWh needed to raise voltage to 14.4 which is a standard bulk charge and automotive alternator recharge value.

This is great that you guys are interested in tests like this. I call the process of learning "B.S. PROOFING" battery condition.

1. Load Test

2. Hydrometer Test

3. Voltage slump and kWh recharge test.

The one amp (2 or 3 amps is acceptable) recharge kWh test is easy to do. Stop charging when voltage stabilizes at 14.4 then see how many kWh have accumulated.

This test can be performed on VRB accumulators as well. albeit with different values used.

[BFL13]

Yes, the two T-1275s are separate. The initial difference was 0.03v but grew to 0.05. We'll see if that gets to be more. I too can't see any reason not to keep paralleling the two 12s but was also hoping to be able to parallel them with a pair or two of 6s without harm while camping (Maybe Mex will see this and comment if he can notice anything from these numbers)

The comment above that you could measure the current is interesting. If there are no evil interactions, but just a draw on the good batteries, then the weaker batts would be just another parasite draw while camping- no big deal

[mena661]

That is a pretty quick voltage drop. Also not sure whether or not that is due to them being deep cycle 12's. BUT since your voltage difference between the two batts is only .04, looks like you're in the clear for paralleling those two.

EDIT: Weird that your voltages dropped a lot but your SG's didn't. They were separated right?

[BFL13]

Got some results from doing Mex's test at the four day point. I can see why he said to wait four or five days.

(Trojan says 100% is 12.73v and 1.277 SG)

Starting /Day Four

Battery A 13.43v/12.66
Battery B 13.40v/12.61

Battery A 1.275,75,70,90,75,75/ 75,75,70,85,75,75
Battery B 75,75,1.300,75,75,75/ 70,70,90,70,70,70

I don't know what all that means exactly 😞 Opinions welcome!

I am a bit concerned that the voltage has dropped so quickly. The 27DCs were from 12.78 to 12.86 at this point and were still at the 12.71-80 level after three weeks sitting disconnected. However the 27s were starting /deep cycle and these T-1275s are deep cycle, so that might account for some of the difference there.

For additional comparison I am running a similar test on my four 6s. Two days after the day they were disconnected) they are still above 13v. All SGs are 1.295-1.300. (subtract 0.0025 if you are fussy about temperature) They have been on 7355 float (13.7v) for the last several weeks before being disconnected to do this test.

I intend to leave them all disconnected for another week to see if the T-1275s stop dropping in voltage so fast. (They have clean tops and all that)

They do "hold a charge" ok while camping. They have been running the 2000w inverter as its battery bank, taking 100a draws with no trouble, doing great. I am surprised to see their voltage drop so quickly when not doing anything!

[BFL13]

AFAIK, the problem occurs where the one battery is in worse enough shape than the other. That could be with same type or different types. If they are different types, but all in good shape then nothing bad happens supposedly.

My test here is to see if the two same type batts are different enough in "shape" to matter, using Mex's test for that.

[Wayne_Dohnal]

I'll note that there are a lot of RVers out there who have been and still are paralleling dissimilar battery banks for charging. Any motorhome with a b.i.r.d. (automatic battery interconnect) does this by default. I don't have numbers, I'll just say "a lot" of Fleetwood motorhomes are built this way. Whenever driving or plugged into shore power the coach and chassis batteries are connected in parallel with no time limit, in some cases for months at a time on shore power. When the charge current is removed, the parallel connection doesn't drop immediately, because the battery voltage has to go below a certain threshold to trigger the disconnect. From the factory my rig has a single FLA battery for the chassis and 2 FLA golf cart batteries for the coach. Some have replaced one or the other with an AGM battery. I've never heard of any problems or shortened battery life because of this system.

[BFL13]

I should do this same thing with my four 6s in series parallel, now all on a float as one big 12.

I found the test where I had done this before to check self-discharge rates on my 27DCs I had back then. Could be used for comparison with these used T-1275s, to help verify the results in a few days:

Posted: 11/29/10 09:21am Link | Print | Notify Moderator

So three "identical" 27s banked for a month and floated at 13.4v did ok, SG held up. Did a cycle down to 75% anyway (not needed)and recharged to 16+ volts recovery method, now trying other way. Separated batteries and let sit. Only 10 days in but interesting results so far.

Voltages by batteries No 1,2,3 where 1 and 3 are from 5/08 and 2 is from 3/10. (EDIT Perhaps I had that mixed up and the last was the newer battery)
Was 13ish volts next day after removing charge to 16+ so start with day after that by Day number:

1- 12.86, 12.87, 12.89
2- 12.83, 12.82, 12.88
3- 12.80, 12.78, 12.86
4- 12.79, 12.76, 12.85
5- 12.78, 12.74, 12.85
6- 12.78, 12.74, 12.84
7- 12.78, 12.74, 12.84
8- 12.77, 12.73, 12.83
9- 12.77, 12.73, 12.83
14- 12.76, 12.72, 12.82
20- 12.74, 12.71, 12.80

All SGS still ok, no sign of lower than when started. (1.275ish--slight variations among cells--the SG of the higher voltage battery 3 is same as the others)