Needed - Advise for budget charger/inverter for Marine "block" NiCd batteries
35 views
Skip to first unread message
Bob F
Nov 9, 2020, 11:46:57 AM11/9/20
to
Several years ago, I acquired 45 SAFT 43 Amp Hour SBM43-2 "block" NiCd
batteries which were being retired from use in a ferry. I finally
started playing with them yesterday, charging 2 banks of 4 of the 2 cell
batteries for several hours with 4 amp and 6 amp car battery chargers.
They seemed to charge well.
I would like to assemble these into a backup supply for a couple desktop
computers and associated equipment in case of power outages. My primary
concern is finding a charger that can properly charge and maintain the
NiCd batteries. Chargers listed for such batteries seem to be not easily
found. I suspect that most Lead acid battery chargers would not be
suitable for such batteries for long term usage, although it is possible
that the right number of cells might be usable with them.
I would appreciate discussion with anyone with experience or knowledge
of using such batteries.
I do have a ProMatic 30-3 40 amp 3 stage charger which could be used in
this project if it proved to be suitable, and another larger marine
charger/inverter I have yet to dig out of a stack of stuff to identify.
batteries which were being retired from use in a ferry. I finally
started playing with them yesterday, charging 2 banks of 4 of the 2 cell
batteries for several hours with 4 amp and 6 amp car battery chargers.
They seemed to charge well.
I would like to assemble these into a backup supply for a couple desktop
computers and associated equipment in case of power outages. My primary
concern is finding a charger that can properly charge and maintain the
NiCd batteries. Chargers listed for such batteries seem to be not easily
found. I suspect that most Lead acid battery chargers would not be
suitable for such batteries for long term usage, although it is possible
that the right number of cells might be usable with them.
I would appreciate discussion with anyone with experience or knowledge
of using such batteries.
I do have a ProMatic 30-3 40 amp 3 stage charger which could be used in
this project if it proved to be suitable, and another larger marine
charger/inverter I have yet to dig out of a stack of stuff to identify.
Jim Wilkins
Nov 9, 2020, 1:07:17 PM11/9/20
to
"Bob F" wrote in message news:robrpv$qlk$1...@dont-email.me...
If you intend to use them as the battery of a computer backup UPS that was
designed for the usual sealed lead-acid AGMs you may find the project more
trouble than it's worth unless you have electronic design skills and test
equipment. A charger not meant for NiCds could destroy them by not stopping
when they are fully charged.
The fast, difficult ways to charge them:
https://batteryuniversity.com/learn/article/charging_nickel_based_batteries
Trickle charging is safer but much slower:
https://www.powerstream.com/NiCd.htm
"The cheapest way to charge a nickel cadmium battery is to charge at C/10
(10% of the rated capacity per hour) for 16 hours.. So a 100 mAH battery
would be charged at 10 mA for 16 hours. This method does not require an
end-of-charge sensor and ensures a full charge. Cells can be charged at this
rate no matter what the initial state of charge is."
"In a standby mode you might want to keep a nickel cadmium battery topped up
without damaging the battery. This can be done safely at a current of
between 0.05 C and .06 C. The voltage required for this is dependent on
temperature, so be sure to regulate the current in the charger."
In any case you need a charger with current control, and preferably Volt and
Amp meters. A laboratory-type power supply can do the job but they aren't
cheap, so I build my own, powered by the DC from my solar panels.
https://www.amazon.com/dp/B083SC2NP1/ref=twister_B0881DXP9L?_encoding=UTF8&psc=1
Some power supplies can tolerate battery voltage on the output when they
aren't turned on, or if the grid goes down, some can't. The easy fix is a
series diode in the output.
Jim Wilkins
Nov 10, 2020, 3:45:28 PM11/10/20
to
"Jim Wilkins" wrote in message news:roc0gj$alc$1...@dont-email.me...
...
=====================================================
"The cheapest way to charge a nickel cadmium battery is to charge at C/10
(10% of the rated capacity per hour) for 16 hours.
==============================
The equivalent slow-but-sure charging method for lead-acids is to limit the
charging voltage to the "float" level, which is just below where the battery
begins to gas, typically around 13.6V. The UPSs I've checked do this, and
it's a selectable option for my solar charging controllers. The disadvantage
is slow charging time after the battery reaches around 80% of its Amp-hour
capacity, because the battery draws less and less current at that voltage as
it approaches full charge.
Initially my UPS charges a heavily discharged battery at 6 Amps and the
battery's state of charge determines the voltage. When the voltage reaches
13.6V it stops increasing and soon the current begins to decrease gradually,
taking several/many hours to stabilize at or below 1% of the battery's
Amp-hour rating, C, for instance 180mA for a 12V 18A-H AGM.
That's what to expect from the battery charger built into a UPS, though you
should check it yourself as the settings are different for gel, agm and
flooded batteries. A UPS with swollen batteries may be available cheaply or
free if you look around. Good cheap used DC-AC inverters to build your own
UPS are much rarer.
...
I would like to assemble these into a backup supply for a couple desktop
computers and associated equipment in case of power outages...
=====================================================
"The cheapest way to charge a nickel cadmium battery is to charge at C/10
(10% of the rated capacity per hour) for 16 hours.
The equivalent slow-but-sure charging method for lead-acids is to limit the
charging voltage to the "float" level, which is just below where the battery
begins to gas, typically around 13.6V. The UPSs I've checked do this, and
it's a selectable option for my solar charging controllers. The disadvantage
is slow charging time after the battery reaches around 80% of its Amp-hour
capacity, because the battery draws less and less current at that voltage as
it approaches full charge.
Initially my UPS charges a heavily discharged battery at 6 Amps and the
battery's state of charge determines the voltage. When the voltage reaches
13.6V it stops increasing and soon the current begins to decrease gradually,
taking several/many hours to stabilize at or below 1% of the battery's
Amp-hour rating, C, for instance 180mA for a 12V 18A-H AGM.
That's what to expect from the battery charger built into a UPS, though you
should check it yourself as the settings are different for gel, agm and
flooded batteries. A UPS with swollen batteries may be available cheaply or
free if you look around. Good cheap used DC-AC inverters to build your own
UPS are much rarer.
Bob F
Nov 11, 2020, 12:27:02 AM11/11/20
to
research and thought. I may have to put together my own charge
controller to match up to the NiCd battery requirements.
This will be a slow project, but I will appreciate any additional
comments from anyone any time along the way.
Jim Wilkins
Nov 11, 2020, 8:41:33 AM11/11/20
to
"Bob F" wrote in message news:rofsn3$ous$1...@dont-email.me...
I appreciate your thoughts. I definitely need to give this a lot of
research and thought. I may have to put together my own charge
controller to match up to the NiCd battery requirements.
This will be a slow project, but I will appreciate any additional
comments from anyone any time along the way.
=======================================
How much AC power does your expected load need? If you don't hack a UPS
you'll need a DC power supply large enough to provide it plus enough extra
to recharge the batteries after an outage at the voltage appropriate to the
batteries and inverter. These voltage and current regulators might help by
expanding your choice of power supply to higher voltage switchers with
little or no output adjustment.
https://www.amazon.com/Digital-Control-Adjustable-Step-down-Regulated/dp/B07QHBYTWY
https://tech.scargill.net/dps5020-diy-power-supply/
20A at 12V is only 240 Watts, which a desktop and peripherals can easily
exceed. The battery voltage has to be within the limits of the UPS or DC-AC
inverter input.
I changed to recycled business laptops with their own backup batteries after
measuring the high power demand of my desktops and monitors. When the grid
fails the laptops run on battery long enough to connect and power up my
large UPS, a free salvaged 900W APC1400 with 2 KWH of external marine
batteries, enough to run my (small) refrigerator and freezer overnight. The
batteries don't gas while charging (I checked) because both the APC and my
solar controllers are set to limit at the 13.6V float voltage instead of
using the faster charging method that raises the voltage to 14.4V
temporarily. Neither has a setting for NiCds, only Lead-acid or Lithium.
Like all the true-sine inverters I've looked at, the APC consumes
significant power internally, about as much as my freezer, so the UPS I run
constantly is a much smaller modified sine one that supports the antenna TV
distribution amp and the laptop that's set up to play and record TV, plus
one lamp and my power recliner. The laptops' external monitors, the audio
system and the power-hungry printer don't really need to be on a UPS if you
have an alternate AC source to manually switch to.
Before buying a UPS, confirm that it can handle extended run time on large
external batteries. The smaller, cheaper ones generally overheat if you try.
One of mine shuts itself off after 30 minutes.
You might look for an orphaned 12V or 14.4V NiCd power tool charger to
experiment with. Anderson Powerpole connectors are nice for battery systems
where either or both sides of a connection may be live.
Current shunts are safer than the internal current range on a meter.
https://www.amazon.com/LM-YN-Current-Resistor-Ammeter/dp/B075WNL8C4/ref=sr_1_3?dchild=1&keywords=current+shunt+75a&qid=1605100803&sr=8-3
The full scale voltage drop is 75mV so that shunt gives 1mV per Amp.
I appreciate your thoughts. I definitely need to give this a lot of
research and thought. I may have to put together my own charge
controller to match up to the NiCd battery requirements.
This will be a slow project, but I will appreciate any additional
comments from anyone any time along the way.
How much AC power does your expected load need? If you don't hack a UPS
you'll need a DC power supply large enough to provide it plus enough extra
to recharge the batteries after an outage at the voltage appropriate to the
batteries and inverter. These voltage and current regulators might help by
expanding your choice of power supply to higher voltage switchers with
little or no output adjustment.
https://www.amazon.com/Digital-Control-Adjustable-Step-down-Regulated/dp/B07QHBYTWY
https://tech.scargill.net/dps5020-diy-power-supply/
20A at 12V is only 240 Watts, which a desktop and peripherals can easily
exceed. The battery voltage has to be within the limits of the UPS or DC-AC
inverter input.
I changed to recycled business laptops with their own backup batteries after
measuring the high power demand of my desktops and monitors. When the grid
fails the laptops run on battery long enough to connect and power up my
large UPS, a free salvaged 900W APC1400 with 2 KWH of external marine
batteries, enough to run my (small) refrigerator and freezer overnight. The
batteries don't gas while charging (I checked) because both the APC and my
solar controllers are set to limit at the 13.6V float voltage instead of
using the faster charging method that raises the voltage to 14.4V
temporarily. Neither has a setting for NiCds, only Lead-acid or Lithium.
Like all the true-sine inverters I've looked at, the APC consumes
significant power internally, about as much as my freezer, so the UPS I run
constantly is a much smaller modified sine one that supports the antenna TV
distribution amp and the laptop that's set up to play and record TV, plus
one lamp and my power recliner. The laptops' external monitors, the audio
system and the power-hungry printer don't really need to be on a UPS if you
have an alternate AC source to manually switch to.
Before buying a UPS, confirm that it can handle extended run time on large
external batteries. The smaller, cheaper ones generally overheat if you try.
One of mine shuts itself off after 30 minutes.
You might look for an orphaned 12V or 14.4V NiCd power tool charger to
experiment with. Anderson Powerpole connectors are nice for battery systems
where either or both sides of a connection may be live.
Current shunts are safer than the internal current range on a meter.
https://www.amazon.com/LM-YN-Current-Resistor-Ammeter/dp/B075WNL8C4/ref=sr_1_3?dchild=1&keywords=current+shunt+75a&qid=1605100803&sr=8-3
The full scale voltage drop is 75mV so that shunt gives 1mV per Amp.
Bob F
Nov 11, 2020, 1:09:15 PM11/11/20
to
The DPS5020 is exactly the kind of device I was thinking I would need to
make my own charger. With it's RS232 capability, I should be able to
regulate the charge states using my computer or a separate micro controller.
The NiCds I have are 43AH cells, for which specified charge current is
8.6A, so this unit should be able to handle them well, even if I
parallel banks.
Jim Wilkins
Nov 11, 2020, 4:27:04 PM11/11/20
to
"Bob F" wrote in message news:roh9c8$g26$1...@dont-email.me...
Excellent!
The DPS5020 is exactly the kind of device I was thinking I would need to
make my own charger. With it's RS232 capability, I should be able to
regulate the charge states using my computer or a separate micro controller.
The NiCds I have are 43AH cells, for which specified charge current is
8.6A, so this unit should be able to handle them well, even if I
parallel banks.
============================
http://basicpi.org/Archive/DPS5020%20CNC%20power%20communication%20protocol.pdf
Excellent!
The DPS5020 is exactly the kind of device I was thinking I would need to
make my own charger. With it's RS232 capability, I should be able to
regulate the charge states using my computer or a separate micro controller.
The NiCds I have are 43AH cells, for which specified charge current is
8.6A, so this unit should be able to handle them well, even if I
parallel banks.
http://basicpi.org/Archive/DPS5020%20CNC%20power%20communication%20protocol.pdf
Jim Wilkins
Nov 13, 2020, 6:39:35 AM11/13/20
to
Data sheets for battery charge controller ICs are a concentrated wealth of
information even if you don't intend to use the device.
https://www.ti.com/lit/ds/symlink/bq24401.pdf?ts=1605220440805&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FBQ24401
I have industrial experience with Lithium and Lead-acid battery charge
controllers, but not with NiCds.
information even if you don't intend to use the device.
https://www.ti.com/lit/ds/symlink/bq24401.pdf?ts=1605220440805&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FBQ24401
I have industrial experience with Lithium and Lead-acid battery charge
controllers, but not with NiCds.
Bob F
Nov 13, 2020, 10:41:46 AM11/13/20
to
and similar ones does not seem to discuss "fast charging" for these
flooded cells at all. It seems that charging is generally recommended at
.2 C5 for various times. I have found nothing listing faster charge specs.
Info does not seem easy to find.
Jim Wilkins
Nov 13, 2020, 12:50:17 PM11/13/20
to
"Bob F" wrote in message news:rom9fn$1v7$1...@dont-email.me...
------------------------------------------------------
Can you determine how they were charged and discharged on the ferry? The
make and model of the charger should help.
https://www.saftbatteries.com/contact-us
Since I don't have cable TV my Internet access is limited and I can't burn
through a lot of data searching.
Can you determine how they were charged and discharged on the ferry? The
make and model of the charger should help.
https://www.saftbatteries.com/contact-us
Since I don't have cable TV my Internet access is limited and I can't burn
through a lot of data searching.
Jim Wilkins
Nov 13, 2020, 6:25:39 PM11/13/20
to
"Bob F" wrote in message news:rom9fn$1v7$1...@dont-email.me...
================================
file:///C:/Users/Net/AppData/Local/Temp/Block+battery_datasheet_EN_0215_LR-1.pdf
"Faster [than what?] recharge time enables at least 80% recovery of capacity
from fully discharged conditions at float in 15 hours."
certainly suggests they aren't suited to 4 hour fast charging, else they'd
advertise it.
The graph labeled
SBM Range - Available capacity after constant voltage charge 1.40 VPC [Volts
Per Cell)] at + 20C (+ 68F)
may reveal an open-loop charging method you could use, if you have enough
Amp-hour capacity to compensate for the reduced charge level.
I can't state that the obvious simple solution WILL work, but it may be
worth investigating.
file:///C:/Users/Net/AppData/Local/Temp/Block+battery_datasheet_EN_0215_LR-1.pdf
"Faster [than what?] recharge time enables at least 80% recovery of capacity
from fully discharged conditions at float in 15 hours."
certainly suggests they aren't suited to 4 hour fast charging, else they'd
advertise it.
The graph labeled
SBM Range - Available capacity after constant voltage charge 1.40 VPC [Volts
Per Cell)] at + 20C (+ 68F)
may reveal an open-loop charging method you could use, if you have enough
Amp-hour capacity to compensate for the reduced charge level.
I can't state that the obvious simple solution WILL work, but it may be
worth investigating.
Jim Wilkins
Nov 20, 2020, 9:39:08 AM11/20/20
to
"Bob F" wrote in message news:rofsn3$ous$1...@dont-email.me...
...
Since no one else has commented yet;
A series string of your used batteries should be matched for capacity.
https://www.icmm.csic.es/jaalonso/velec/baterias/aboutn~1.htm
See "Cell Reversal" and the following section.
Measuring cell capacity requires discharging them individually while
recording the time and current. Electronic loads are available to do this,
for a price:
https://www.amazon.com/OTGO-Constant-Electronic-Discharge-Capacity/dp/B07GZG83MR
Car headlight bulbs can be used for 12V but I don't know of a safe
substitute for power resistors at 1.2V. My ceramic resistors can reach
500-600F at their max rated power and need to be on a thermal and electrical
insulator like firebrick.
Measuring the charge works only for batteries that charge at 100%
efficiency, like Lithiums.
At home I use PC-interfaced DVMs to measure and record voltage and current,
and this and smaller rheostats as the load:
https://www.alliedelec.com/product/ohmite/rus1r0e/70022560/
Naturally I didn't buy new at that price.
When I was the battery tech on an electric vehicle project the necessary
special test equipment consisted of a digital storage scope with a DC
current probe and a programmable electronic load that dumped its heat into a
barrel of water.
A DC-AC inverter with a 120V load will draw constant power from your
batteries as the voltage drops, unlike the constant current of an electronic
load or constant resistance of my rheostats (unless adjusted), so it tells
you how long your battery will support your load better than other capacity
measures. An inverter has the advantage of an automatic low voltage cutoff
that allows it to safely run unattended but you still need to record the
voltage and current at frequent intervals. The summed current gives
discharged Amp-hours, the voltage drop shows rising battery impedance that
may limit the inverter run time to less than full battery capacity.
...
This will be a slow project, but I will appreciate any additional
comments from anyone any time along the way.
-------------------------
comments from anyone any time along the way.
Since no one else has commented yet;
A series string of your used batteries should be matched for capacity.
https://www.icmm.csic.es/jaalonso/velec/baterias/aboutn~1.htm
See "Cell Reversal" and the following section.
Measuring cell capacity requires discharging them individually while
recording the time and current. Electronic loads are available to do this,
for a price:
https://www.amazon.com/OTGO-Constant-Electronic-Discharge-Capacity/dp/B07GZG83MR
Car headlight bulbs can be used for 12V but I don't know of a safe
substitute for power resistors at 1.2V. My ceramic resistors can reach
500-600F at their max rated power and need to be on a thermal and electrical
insulator like firebrick.
Measuring the charge works only for batteries that charge at 100%
efficiency, like Lithiums.
At home I use PC-interfaced DVMs to measure and record voltage and current,
and this and smaller rheostats as the load:
https://www.alliedelec.com/product/ohmite/rus1r0e/70022560/
Naturally I didn't buy new at that price.
When I was the battery tech on an electric vehicle project the necessary
special test equipment consisted of a digital storage scope with a DC
current probe and a programmable electronic load that dumped its heat into a
barrel of water.
A DC-AC inverter with a 120V load will draw constant power from your
batteries as the voltage drops, unlike the constant current of an electronic
load or constant resistance of my rheostats (unless adjusted), so it tells
you how long your battery will support your load better than other capacity
measures. An inverter has the advantage of an automatic low voltage cutoff
that allows it to safely run unattended but you still need to record the
voltage and current at frequent intervals. The summed current gives
discharged Amp-hours, the voltage drop shows rising battery impedance that
may limit the inverter run time to less than full battery capacity.
Bob F
Nov 20, 2020, 10:57:46 PM11/20/20
to
Jim Wilkins
Nov 21, 2020, 7:18:10 AM11/21/20
to
"Bob F" wrote in message news:rpa37n$ant$1...@dont-email.me...
Again, thanks for the great input. More food for thought.
-----------------------------
Apparently you can take your time to decide what to do with them.
https://batteryuniversity.com/learn/article/how_to_store_batteries
"Nickel-based batteries can be stored in a fully discharged state with no
apparent side effect."
Again, thanks for the great input. More food for thought.
Apparently you can take your time to decide what to do with them.
https://batteryuniversity.com/learn/article/how_to_store_batteries
"Nickel-based batteries can be stored in a fully discharged state with no
apparent side effect."
Jim Wilkins
Dec 3, 2020, 11:30:09 AM12/3/20
to
"Jim Wilkins" wrote in message news:ron4lh$5fc$1...@dont-email.me...
...
The Waterpik I just repaired uses two AA NiMH cells in series, float-charged
with 3VAC through a series diode. I didn't want to recommend such a simple
unregulated charging circuit until I had seen it used in a product.
It's a slow charge, perhaps adequate for emergency backup but not daily deep
cycling. In my home solar battery backup system the fast charge to the float
voltage is automatic while higher voltage acceptance and equalizing charges
are manual.
...
The graph labeled
SBM Range - Available capacity after constant voltage charge 1.40 VPC [Volts
Per Cell)] at + 20C (+ 68F)
may reveal an open-loop charging method you could use, if you have enough
Amp-hour capacity to compensate for the reduced charge level.
---------------------------
SBM Range - Available capacity after constant voltage charge 1.40 VPC [Volts
Per Cell)] at + 20C (+ 68F)
may reveal an open-loop charging method you could use, if you have enough
Amp-hour capacity to compensate for the reduced charge level.
The Waterpik I just repaired uses two AA NiMH cells in series, float-charged
with 3VAC through a series diode. I didn't want to recommend such a simple
unregulated charging circuit until I had seen it used in a product.
It's a slow charge, perhaps adequate for emergency backup but not daily deep
cycling. In my home solar battery backup system the fast charge to the float
voltage is automatic while higher voltage acceptance and equalizing charges
are manual.
Reply all
Reply to author
Forward
0 new messages