Economical replacement period for power plant batteries in cell towers

[Sahib]

We have a lot of cell towers. The diesel generator in each site runs for a long time due to frequent power failures. My plan to reduce diesel consumption is to avail the power plant batteries to the maximum extent possible as the cost of running generator during power failure is three to four times compared to power plant battery emergency supply. So if power plant battery supply is used to the extent possible during power failure instead of generator supply, the saving due to reduction in diesel fuel may be enough for replacement worn out batteries used in that way. The power plant batteries are of valve regulated type. Are their charge and discharge cycle suitable for any pay back period? Thanks.

 

[kingpb]

How often are you talking about, and what is the duty cycle rating of your batteries? What if the plant needs the emergency supply?

I would say you need to have your own battery system, but then how long of a duration outage are you talking about? That could require a hefty battery depending on duration, and then the diesel is still going to be needed for long periods.

Power plant batteries are used for specific purposes; and duty cycles would need to be studied.

 

[iceworm]

We have a lot of cell towers. The diesel generator in each site runs for a long time due to frequent power failures. My plan to reduce diesel consumption is to avail the power plant batteries to the maximum extent possible as the cost of running generator during power failure is three to four times compared to power plant battery emergency supply. So if power plant battery supply is used to the extent possible during power failure instead of generator supply, the saving due to reduction in diesel fuel may be enough for replacement worn out batteries used in that way. The power plant batteries are of valve regulated type. Are their charge and discharge cycle suitable for any pay back period? Thanks.

Interesting concept. Some years back (~15 years) ATT Canada did a research project concerning powering remote sites. The cost of delivering fuel to the remote sites was horribly expensive. The generaor was sized for the high peak loading and ran lightly loaded for most of the time. There were problems with wet stacking and poor efficiency.

The project installed a battery bank with sufficient capacity for 24 hours (discharge to 50% DoD). The battery charger was sized to fully load the generator and recharge the batteries in 4 hours. So, state of charge controls monitor the batteries and when they hit 50%, the gen started and the charger hit the batteries as hard as it could with out over heating the batteries - about once a day.

The efficiency gained from running the gen flatout saved enough fuel the system paid for the new batteries in one year.

They had some trouble with the charging algolrithm at first. The batteries were not topping off and wer left a bit less than full charge. And the battery capacity dropped over time. They fixed the algolrithm and that issue went away - small increase in fuel consumption.

I think you could apply this to your installations. For example (just an example - I don't know if the numbers are good or not):
Size the batteries, for 50% state-of-charge, to cover 90% of your outages (1 day to 3 days maybe - I don't know). If at 50% SOC, the power is not back on, start the gen. Size the gen (and charger) for a 12 hour recharge plus required plant loads. As long as you don't go below 50%, you won't hurt the batteries. It will take some juggling to make the best use of the existing equipment.

Excellent science fair project. Once you put together a good proposal, the management will come up with the money. Let us know how it goes.

ice

 

[kingpb]

Interesting concept. Some years back (~15 years) ATT Canada did a research project concerning powering remote sites. The cost of delivering fuel to the remote sites was horribly expensive. The generaor was sized for the high peak loading and ran lightly loaded for most of the time. There were problems with wet stacking and poor efficiency.

The project installed a battery bank with sufficient capacity for 24 hours (discharge to 50% DoD). The battery charger was sized to fully load the generator and recharge the batteries in 4 hours. So, state of charge controls monitor the batteries and when they hit 50%, the gen started and the charger hit the batteries as hard as it could with out over heating the batteries - about once a day.

The efficiency gained from running the gen flatout saved enough fuel the system paid for the new batteries in one year.

They had some trouble with the charging algolrithm at first. The batteries were not topping off and wer left a bit less than full charge. And the battery capacity dropped over time. They fixed the algolrithm and that issue went away - small increase in fuel consumption.

I think you could apply this to your installations. For example (just an example - I don't know if the numbers are good or not):
Size the batteries, for 50% state-of-charge, to cover 90% of your outages (1 day to 3 days maybe - I don't know). If at 50% SOC, the power is not back on, start the gen. Size the gen (and charger) for a 12 hour recharge plus required plant loads. As long as you don't go below 50%, you won't hurt the batteries. It will take some juggling to make the best use of the existing equipment.

Excellent science fair project. Once you put together a good proposal, the management will come up with the money. Let us know how it goes.

ice

All well and good, I agree with your synopsis. However, the question was related to using existing power plant batteries, not a new install of batteries dedicated to the application. I apologize in advance if I missed something in the post.

 

[Phil Corso]

Sahib,

Have you looked into PV-Solar or Fuel-Cell technology to augment Gen-set?

Regards, Phil Corso

 

[iceworm]

... For example:
Size the batteries, for 50% state-of-charge, to cover 90% of your outages (1 day to 3 days maybe - I don't know). If at 50% SOC, the power is not back on, start the gen. Size the gen (and charger) for a 12 hour recharge plus required plant loads. As long as you don't go below 50%, you won't hurt the batteries.. ...

Oh yeah - you likely will also have to resize the utility powered charger as well.

As I said, it will take some juggling to make the best use of the existing equipment.

ice

 

[Sahib]

iceworm:
How the 50% state of battery charge relates to its voltage, because the sensor to start the engine is to depend on it? Thanks.

 

[iceworm]

iceworm:
How the 50% state of battery charge relates to its voltage, because the sensor to start the engine is to depend on it? Thanks.

If you are using lead acid (wet or VRLA), the voltage is pretty linear with the state of charge - other chemistery not so much. Do some intenet research I suspect there is plenty of stuff out there. Should be applications for EVs, solar at least.

If you are looking at this - you are the engineer of record and you will be doing development. Like I said great science fair project.

ice

 

[Sahib]

Thanks, ice. One more question, if you do not mind .

Size the batteries, for 50% state-of-charge, to cover 90% of your outages (1 day to 3 days maybe - I don't know). If at 50% SOC, the power is not back on, start the gen. Size the gen (and charger) for a 12 hour recharge plus required plant loads.

If the POCO power supply is available but at moderately long intervals, how many hours the generator is generally required to be run per day, to avoid, perhaps, partial charged state of batteries? In short, is the running of generator to bring the batteries up to fully charged condition required or not?

 

[Sahib]

If you are using lead acid (wet or VRLA), the voltage is pretty linear with the state of charge

ice

So if it is a 400Ah, 48V battery, 50% state of charge relates approximately to 24V output voltage of the battery. But there would be a provision for Automatic Isolation (by a DC contactor) of the battery from the load (44.4V for 48V units). So to reach 50% state of charge of battery does not seem to be possible.

 

[iceworm]

... If the POCO power supply is available but at moderately long intervals, how many hours the generator is generally required to be run per day, to avoid, perhaps, partial charged state of batteries? In short, is the running of generator to bring the batteries up to fully charged condition required or not?

I don't understand the question:
If the poco is available, power the battery charger with that. You will want to keep the batteries floating at 100% SoC. The gen might run an hour a month to exercise.

ice

 

[iceworm]

So if it is a 400Ah, 48V battery, 50% state of charge relates approximately to 24V output voltage of the battery. But there would be a provision for Automatic Isolation (by a DC contactor) of the battery from the load (44.4V for 48V units). So to reach 50% state of charge of battery does not seem to be possible.

For your 400AH, 48V (nominal), assuming VLRA, you could see (open circuit):
100% 51.2V
75% 50.4V
50% 49.2V
25% 48.0V
0% 47.2V

Fo 50% SoC you are only going to get 200AH before you start the generator to recharge. But that is all you can get out of them if you want them to last. Even then, keeping the DoD or 50%, they may only last 1000 discharge/charge cycles.

You have an opportunity for an excellent science fair project. Might even get a paper out of it.

"Optimizing Generator Fuel Consumption for High Reliability Useage in Areas of Low Utility Power Reliability" by Sahib

Time for you to get researching:
Battery chemistry
Charging algolrithms
Automatic generator controls
Generator fuel efficiency

Lots of interesting stuff

ice

 

[Sahib]

Excellent information, ice :thumbsup:.

Regarding

Fo 50% SoC you are only going to get 200AH before you start the generator to recharge. But that is all you can get out of them if you want them to last. Even then, keeping the DoD or 50%, they may only last 1000 discharge/charge cycles.
ice

those 1000 discharge/charge cycles of VRLA batteries, at what ambient temperature?

The discharge/charge cycles of VRLA batteries at other ambient temperatures?

 

[iceworm]

... those 1000 discharge/charge cycles of VRLA batteries, at what ambient temperature?

The discharge/charge cycles of VRLA batteries at other ambient temperatures?

Sahib -
It's time for you to go to work. Buck up - you are the engineer of record. You start by writing a Detailed Design Document. This is not the physics or design, this is the list of specifications of how you expect the system to behave.

Concurrently, start your research on the basic physics.

You have to do this - not me.

ice

 

[Sahib]

Update: Here is a draft presentation on fuel reduction study for your comments please.

 

[Sahib]

I located a mistake in the attachment of my last post: the cost of the battery is really much higher than the saving due to non-operation generator and so could not be replaced with that saving.

 

[GoldDigger]

I located a mistake in the attachment of my last post: the cost of the battery is really much higher than the saving due to non-operation generator and so could not be replaced with that saving.

One case where the intermittent operation of the generator at full output has more than just fuel savings compared to continuous operation of the generator at lower output is for a diesel prime mover.
If the continuous load is 50% or less of the full rated output of the engine/generator combination, a phenomenon called "wet stacking" can greatly reduce the prime mover life and increase maintenance requirements.
This can increase the cost of generator-only operation by far note than the extra fuel costs from reduced efficiency of operation.
If you start with a generator sized for continuous operation without too much excess capacity you can avoid this problem.