Just wondering what others have been doing to power multiple radios located within the same shelter? Seems that Motorola wants to sell a separate charger and battery for every radio in the rack. Each radio has an 120VAC power supply with battery revert, and a separate charger (Argus) with battery. This seems excessive if there are a handful of Quantars or MTR2000s all with their own backup battery and charger.
Can one simply parallel a few batteries and the ARGUS charger to reduce numbers/cost? This assumes the backup time and battery capacity... correct wiring harness, and DC circuit breakers, etc, etc... anyone done this?
Of course, one can also buy a DC power supply (24 or 48V for Quantars, and 24V for MTR2000) for the radios instead of the AC versions. What have others done for a power plant (rectifiers and battery)? Too bad the MTR doesn't have 48V... most of the cellular industry has now gone to -48V.
Just wondering... Any ideas or comments?
RFDude
Power Plants for Quantars and MTR2000
Moderator: Queue Moderator
There is a thread here on what I recommend for backing up Quantars; search for "Heart Interface."
I am not an advocate of Motorola's backup harness and routine, primarily because the battery charger on the Quantar (and every other base station I've encountered) is 20-years old technology and will cook the batteries to lifelessness, even if there is never an outage, in two years or so.
Most installations I deal with have auto-start backup generators, so the function of the battery backup is limited: (a) to prevent the radios from rebooting while the generator starts, warms and transfers load, and (b) to avoid a disaster if the generator fails to start or hold load, just long enough to get the Highway Department to the site with a skid-mount generator.
In any event, I strongly advocate that each base station have its own independent battery-backed standby power supply. Quantars take 600W on transmit, and you can figure an average draw at 12VDC of 7.5 to 10A, depending on duty cycle. To be sure that you can keep them running for a couple of hours, you should have available about 180 AH @ 12VDC. For my machines, this is two G31 gel cells, which power the Heart Interface, which is a combined inverter/smart 3-stage charger/very fast (6 ms) transfer switch.
Say you had four base stations running. Since you cannot eliminate the possibility that they'd all be transmitting at once, you'd need at least 8 batteries, an inverter (if you go that route) with not less than 3 kW capacity (and 4 kW would be better), and not less than 200W of charging capability. Such equipment is rare and expensive. Moreover, the entire site then becomes vulnerable to a common cause failure (such as any one of the batteries developing a bad cell).
The power supply I use can be assembled for about $1,500 total cost (including installation), per radio. The batteries should last at least 5 years and require no maintenance. (The oldest one I have going is 9 years, and the batteries still test good, so I have no empiracal data on battery life. "No maintenance" means no adding of water to the batteries, but they should be visited and given a brief load test at least twice a year.) Since the power supply outputs 110VAC to the Quantar, the radio never sees loss of power, doesn't reboot, doesn't cut back on power during revert, and doesn't require the Quantar backup option and cable. Should any one power supply fail, the rest of the radios continue working. The Heart is programmed to define "loss of utility power" as including voltage above 135V (RMS) or below 95V (RMS), or frequency off by more than 10%, so the unit also protects against brownouts and loss-of-legs. The only negative of the whole setup is that you also don't have access to the Quantar "on battery" alarm, but if that function is important, it can be rigged in different ways.
I am not an advocate of Motorola's backup harness and routine, primarily because the battery charger on the Quantar (and every other base station I've encountered) is 20-years old technology and will cook the batteries to lifelessness, even if there is never an outage, in two years or so.
Most installations I deal with have auto-start backup generators, so the function of the battery backup is limited: (a) to prevent the radios from rebooting while the generator starts, warms and transfers load, and (b) to avoid a disaster if the generator fails to start or hold load, just long enough to get the Highway Department to the site with a skid-mount generator.
In any event, I strongly advocate that each base station have its own independent battery-backed standby power supply. Quantars take 600W on transmit, and you can figure an average draw at 12VDC of 7.5 to 10A, depending on duty cycle. To be sure that you can keep them running for a couple of hours, you should have available about 180 AH @ 12VDC. For my machines, this is two G31 gel cells, which power the Heart Interface, which is a combined inverter/smart 3-stage charger/very fast (6 ms) transfer switch.
Say you had four base stations running. Since you cannot eliminate the possibility that they'd all be transmitting at once, you'd need at least 8 batteries, an inverter (if you go that route) with not less than 3 kW capacity (and 4 kW would be better), and not less than 200W of charging capability. Such equipment is rare and expensive. Moreover, the entire site then becomes vulnerable to a common cause failure (such as any one of the batteries developing a bad cell).
The power supply I use can be assembled for about $1,500 total cost (including installation), per radio. The batteries should last at least 5 years and require no maintenance. (The oldest one I have going is 9 years, and the batteries still test good, so I have no empiracal data on battery life. "No maintenance" means no adding of water to the batteries, but they should be visited and given a brief load test at least twice a year.) Since the power supply outputs 110VAC to the Quantar, the radio never sees loss of power, doesn't reboot, doesn't cut back on power during revert, and doesn't require the Quantar backup option and cable. Should any one power supply fail, the rest of the radios continue working. The Heart is programmed to define "loss of utility power" as including voltage above 135V (RMS) or below 95V (RMS), or frequency off by more than 10%, so the unit also protects against brownouts and loss-of-legs. The only negative of the whole setup is that you also don't have access to the Quantar "on battery" alarm, but if that function is important, it can be rigged in different ways.
-
ASTROMODAT
- Posts: 1825
- Joined: Tue Nov 05, 2002 12:32 am
I suggest you use the standard Motorola P/S on your Quantars, with the Motorola Battery Revert Option. Connect it to a nice bank of 24 V sealed gel cells. We have done this for years, with and without power outages on our Quantars, and we have had excellent results. Add-on "Rube Goldberg" type stuff is OK for Hams, etc, but for commercial operations where lives are at stake, stick with the factory integrated solution.
Larry
Larry
While I know absolutely nothing about the Quantar, here's something to keep in mind regarding the MTR2000:
On the MTR2000, the batteries should not be connected to the station when the station has AC power. Motorola is quite clear on this in the station documentation, warning of damage to both the station and battery banks.
As far as I can tell, from looking at their block diagrams, it looks like Motorola just tied the battery backup connector onto the supply rails between the step-down and regulator stages in the supply, without the benefit of isolation diodes. As a result, there is always station voltage on the backup power connector.
As a result, any backup system (including the factory specified Argus unit) for the MTR2000 must include a disconnect relay for the battery bank. The station backplane includes both logic level and N.O. contact closure outputs for A.C. fail that can be used for this purpose. If you shared a battery bank you would almost certainly have to provide individual disconnect relays for each repeater.
I think that, to some degree, the optimum solution depends on how much run time you expect to need. If you are looking for a short run time, until your generator kicks in, I think RKG is 100% dead on. The solution that he suggests is utterly invisible to the repeaters, and is very well thought out. I also share his lack of enthusiasm for the charging technology found in many revert systems, including the Argus.
Where longer run times are needed, I am a little jealous about the power lost in the DC-AC inversion process, and then again in the station's AC-DC power supply. Overall, I think that if a smooth transition is the big concern, then RKG's solution is tough to beat. If maximum run time is the goal, then I would vote for direct battery operation.
On a related note, I am currently putting together a battery backup for a station that includes both a 24V MTR2000, and a number of 12V control and link components. As I mentioned above, I'm not nuts about the Argus' battery maintenance, and I need to support 12V and 24V operation.
After some time with my local MSS and the folks at DuraCom, I currently have on my bench the parts for a rather interesting solution. Essentially, I have two linked three-stage chargers that can charge and maintain two 12V banks in series, drawing 12V off of the "lower" bank, and 24V off of the full bank. Since the banks are managed individually I can increase the capacity of the lower 12V half if I find that it is drawing down significantly faster than the rest of the bank. For now, the bank consists of four 6V 200Ah AGM batteries in series. We may double that if the run time is ultimately not to our liking. (The site sees frequent outages, and a generator is a political site issue)
My goal is to build and install the thing some time this week. After letting it get up a good charge, I am going to go up, look at my watch, kick the plug out of the wall, and see how we do. I'll try to report back sometime next week.
This looks like it has a lot of potential, we'll see how it goes in the real world...
On the MTR2000, the batteries should not be connected to the station when the station has AC power. Motorola is quite clear on this in the station documentation, warning of damage to both the station and battery banks.
As far as I can tell, from looking at their block diagrams, it looks like Motorola just tied the battery backup connector onto the supply rails between the step-down and regulator stages in the supply, without the benefit of isolation diodes. As a result, there is always station voltage on the backup power connector.
As a result, any backup system (including the factory specified Argus unit) for the MTR2000 must include a disconnect relay for the battery bank. The station backplane includes both logic level and N.O. contact closure outputs for A.C. fail that can be used for this purpose. If you shared a battery bank you would almost certainly have to provide individual disconnect relays for each repeater.
I think that, to some degree, the optimum solution depends on how much run time you expect to need. If you are looking for a short run time, until your generator kicks in, I think RKG is 100% dead on. The solution that he suggests is utterly invisible to the repeaters, and is very well thought out. I also share his lack of enthusiasm for the charging technology found in many revert systems, including the Argus.
Where longer run times are needed, I am a little jealous about the power lost in the DC-AC inversion process, and then again in the station's AC-DC power supply. Overall, I think that if a smooth transition is the big concern, then RKG's solution is tough to beat. If maximum run time is the goal, then I would vote for direct battery operation.
On a related note, I am currently putting together a battery backup for a station that includes both a 24V MTR2000, and a number of 12V control and link components. As I mentioned above, I'm not nuts about the Argus' battery maintenance, and I need to support 12V and 24V operation.
After some time with my local MSS and the folks at DuraCom, I currently have on my bench the parts for a rather interesting solution. Essentially, I have two linked three-stage chargers that can charge and maintain two 12V banks in series, drawing 12V off of the "lower" bank, and 24V off of the full bank. Since the banks are managed individually I can increase the capacity of the lower 12V half if I find that it is drawing down significantly faster than the rest of the bank. For now, the bank consists of four 6V 200Ah AGM batteries in series. We may double that if the run time is ultimately not to our liking. (The site sees frequent outages, and a generator is a political site issue)
My goal is to build and install the thing some time this week. After letting it get up a good charge, I am going to go up, look at my watch, kick the plug out of the wall, and see how we do. I'll try to report back sometime next week.
This looks like it has a lot of potential, we'll see how it goes in the real world...
It is a good idea to keep an eye on the batteries from time to time, particularly if you are concerned about adequacy in bank capacity, and here is a "down and dirty" load test that I perform. (Did it last Friday, in fact, on one system powering a 110W Quantar and a Zetron 38M.)
First, I threw the AC breaker. Transfer took place and radio did not reboot.
Then I went away for 2 hours and came back and watched system voltage. It was stead in the mid 12.7VDC range.
Then I keyed the transmitter locally and watched voltage. Dropped quickly to about 12.5VDC and held there steady.
Unkeyed, and voltage recovered promptly to 12.7 range.
Repeated this test once every 15 minutes for 1-1/2 hours. Same results.
Got bored; reset the breaker and went home.
As for the adequacy of your 200 AH bank, tell me the power consumption of the MTR transmitting and not transmitting, and of the other things connected to the bank, and I'll run some calcs. A good rule of thumb is that the average load on the bank should be kept to or below 5% of nominal capacity, and if you do that, the system should be good for 10 hours. At that point, the batteries are 50% discharged. Anything greater than a 50% discharge will shorten the cycle life of deep-cycle storage batteries, though AGMs (and gel cels) can take (a few) deep hits with less lifetime impact than flooded cells.
Note that the nominal capacity of a storage battery meant for cycling assumes a 5% discharge rate. If the discharge rate is higher than 5%, the effective capacity of the battery (or bank) (the starting point in the calc) is reduced. Somewhere I have a book with values, but it suffices to observe that the reduction in effective capacity as average load increases over 5% of nominal is more than linear.
Last observation: if you are using AGMs, be sure whatever you are using to charge them has an AGM curve and that you have selected it. AGMs will take a higher current rate than flooded cells when discharged, but are much more sensitive to overvoltage.
If I were going to run the radios on batteries directly, I'd specify the versions with DC input, disable their charging function (if they have one), hook them up to the battery bank, and then wire in a "smart" charger (like the Heart Interface or the StatPower) to the battery bank. While line power is available, and with the batteries charged, the charger will maintain a float voltage (about 13.4 for a 12 volt system and about 26.8 for a 24V system), regardless of load, up to its capacity. If there is an excess load (such as the station transmitting), the load will be shared with the batteries, and voltage will be less than float but higher than with the batteries alone. If there is an outage, the batteries carry the system by themselves, but as soon as offsite power is restored, the charger will recharge the batteries without operator intervention. So long as the batteries are in good shape, the system will never see a material voltage drop in the event of an outage, and therefore will not reboot. Like my system, the direct DC supply and the intervention of the charger will insulate the radios from over- or under-voltage conditions on the offsite power (but if power gets too far from norm, the charger will shut down and the batteries will be on their own).
The advantage of such a system is that, as noted, you avoid the inefficiencies of the DC-to-AC-to-DC conversions. (For what it is worth, the efficiency of the 1 kW 12V Heart Interface is spec'd at 85% full load, 93% peak.) The disadvantages are (a) that the batteries won't last as long, since they are discharging/recharging as a result of every transmission cycle and (b) the radios will see a greater voltage drop on keyup than if they are supplied by 110V utility power. What the effect of the latter is over time I do not know, but I doubt it is zero.
First, I threw the AC breaker. Transfer took place and radio did not reboot.
Then I went away for 2 hours and came back and watched system voltage. It was stead in the mid 12.7VDC range.
Then I keyed the transmitter locally and watched voltage. Dropped quickly to about 12.5VDC and held there steady.
Unkeyed, and voltage recovered promptly to 12.7 range.
Repeated this test once every 15 minutes for 1-1/2 hours. Same results.
Got bored; reset the breaker and went home.
As for the adequacy of your 200 AH bank, tell me the power consumption of the MTR transmitting and not transmitting, and of the other things connected to the bank, and I'll run some calcs. A good rule of thumb is that the average load on the bank should be kept to or below 5% of nominal capacity, and if you do that, the system should be good for 10 hours. At that point, the batteries are 50% discharged. Anything greater than a 50% discharge will shorten the cycle life of deep-cycle storage batteries, though AGMs (and gel cels) can take (a few) deep hits with less lifetime impact than flooded cells.
Note that the nominal capacity of a storage battery meant for cycling assumes a 5% discharge rate. If the discharge rate is higher than 5%, the effective capacity of the battery (or bank) (the starting point in the calc) is reduced. Somewhere I have a book with values, but it suffices to observe that the reduction in effective capacity as average load increases over 5% of nominal is more than linear.
Last observation: if you are using AGMs, be sure whatever you are using to charge them has an AGM curve and that you have selected it. AGMs will take a higher current rate than flooded cells when discharged, but are much more sensitive to overvoltage.
If I were going to run the radios on batteries directly, I'd specify the versions with DC input, disable their charging function (if they have one), hook them up to the battery bank, and then wire in a "smart" charger (like the Heart Interface or the StatPower) to the battery bank. While line power is available, and with the batteries charged, the charger will maintain a float voltage (about 13.4 for a 12 volt system and about 26.8 for a 24V system), regardless of load, up to its capacity. If there is an excess load (such as the station transmitting), the load will be shared with the batteries, and voltage will be less than float but higher than with the batteries alone. If there is an outage, the batteries carry the system by themselves, but as soon as offsite power is restored, the charger will recharge the batteries without operator intervention. So long as the batteries are in good shape, the system will never see a material voltage drop in the event of an outage, and therefore will not reboot. Like my system, the direct DC supply and the intervention of the charger will insulate the radios from over- or under-voltage conditions on the offsite power (but if power gets too far from norm, the charger will shut down and the batteries will be on their own).
The advantage of such a system is that, as noted, you avoid the inefficiencies of the DC-to-AC-to-DC conversions. (For what it is worth, the efficiency of the 1 kW 12V Heart Interface is spec'd at 85% full load, 93% peak.) The disadvantages are (a) that the batteries won't last as long, since they are discharging/recharging as a result of every transmission cycle and (b) the radios will see a greater voltage drop on keyup than if they are supplied by 110V utility power. What the effect of the latter is over time I do not know, but I doubt it is zero.
Power Plants for Quantars and MTR2000
Interesting discussion....
The MTR AC power supply puts out ~ 14.4VDC which is more than most battery float voltages. So leaving a battery directly on the DC revert terminals will cause overcharging and damage. While looking at the block diagram, I couldn't see any other reason why this is a bad idea... although some birdies suggest there might be a latch up issue with the buck converters for the lower voltages that would result in blown semiconductors. Sounds weird? One of the guys in the shop blew a PS up when connecting it to the battery. Can't say much more without a schematic and you know that Motorola isn't much help either... if they even know... they subcontract the PS out to third parties.
At sites where public safety is installed and lots of radios + room for more, I've been using the cellular approach: A large North American name brand 600A power plant in a 23" rack that provides +24V, with N+1 rectifiers (each is at least 60A, some are 130A each) and LVD. Most also have a small -48V 5A DC-DC converter for a few loads that require it. Batteries are standard "front access" 12V 150 AH that sit on trays. Despite the generator, there is a minimum 4 hours of battery based on 50% transmitter duty cycle. Of course, all the radios have DC only power supplies. Yes, this is expensive, but it is bullet proof.
The issue is with smaller sites that have 4 or less radios. This is where the big power plant solution above is too costly. And the Motorola 120VAC with DC revert is such a shame... for it looks like it is 95% there to be able to simply do away with an external ARGUS charger. The Argus adds significant costs, and having separate batteries... well, that just isn't good "trunking" efficiency for lack of a better term. You end up having to supply every radio based on worst case. Hence the question to the group. Its one of those "catch-22" things... I don't like the Motorola solution, but...I don't want to kludge something together either.
Define KLUDGE... For discussion: The best way I could come up with to deal with the 14.4V DC in multiple MTRs.... It might even work.... Buy the AC PS for all radios and parallel the DC sides via a DC distribution bus with circuit breakers. The battery would be connected to this distribution bus with an isolation diode. For charging, use a good quality SOLAR CHARGE CONTROLLER (STECA or MORNINGSTAR, etc). These function in a ON/OFF duty cycle fashion, hence when the battery voltage would rise to an appropriate level (say 13.5V for an AGM), the charge controller would disconnect the battery from charging. Current limit charge through a suitable resistor. This idea could use some refinement... but in principle it would allow one to get rid of the ARGUS battery charger.
Be gentle! This is only an idea thrown into the wind for discussion.
RFDude.
The MTR AC power supply puts out ~ 14.4VDC which is more than most battery float voltages. So leaving a battery directly on the DC revert terminals will cause overcharging and damage. While looking at the block diagram, I couldn't see any other reason why this is a bad idea... although some birdies suggest there might be a latch up issue with the buck converters for the lower voltages that would result in blown semiconductors. Sounds weird? One of the guys in the shop blew a PS up when connecting it to the battery. Can't say much more without a schematic and you know that Motorola isn't much help either... if they even know... they subcontract the PS out to third parties.
At sites where public safety is installed and lots of radios + room for more, I've been using the cellular approach: A large North American name brand 600A power plant in a 23" rack that provides +24V, with N+1 rectifiers (each is at least 60A, some are 130A each) and LVD. Most also have a small -48V 5A DC-DC converter for a few loads that require it. Batteries are standard "front access" 12V 150 AH that sit on trays. Despite the generator, there is a minimum 4 hours of battery based on 50% transmitter duty cycle. Of course, all the radios have DC only power supplies. Yes, this is expensive, but it is bullet proof.
The issue is with smaller sites that have 4 or less radios. This is where the big power plant solution above is too costly. And the Motorola 120VAC with DC revert is such a shame... for it looks like it is 95% there to be able to simply do away with an external ARGUS charger. The Argus adds significant costs, and having separate batteries... well, that just isn't good "trunking" efficiency for lack of a better term. You end up having to supply every radio based on worst case. Hence the question to the group. Its one of those "catch-22" things... I don't like the Motorola solution, but...I don't want to kludge something together either.
Define KLUDGE... For discussion: The best way I could come up with to deal with the 14.4V DC in multiple MTRs.... It might even work.... Buy the AC PS for all radios and parallel the DC sides via a DC distribution bus with circuit breakers. The battery would be connected to this distribution bus with an isolation diode. For charging, use a good quality SOLAR CHARGE CONTROLLER (STECA or MORNINGSTAR, etc). These function in a ON/OFF duty cycle fashion, hence when the battery voltage would rise to an appropriate level (say 13.5V for an AGM), the charge controller would disconnect the battery from charging. Current limit charge through a suitable resistor. This idea could use some refinement... but in principle it would allow one to get rid of the ARGUS battery charger.
Be gentle! This is only an idea thrown into the wind for discussion.
RFDude.
Creative, but you'll find that your batteries will never charge.
In order to charge deep cycle storage batteries, you need to apply different charge curves to them, depending on the progress of the charge. The accepted curves are as follows (referencing a 12V battery):
Stage 1 (Bulk): The charger is current regulated, at about 90-95% of its rated capacity, and voltage sensed. The voltage trip point is set to the point just at the point where the battery "gasses," which for a flooded cell battery is 14.4 VDC at 70 degrees F. When battery terminal voltage reaches this point and stays there for 2 minutes or so, the charger shifts to the next stage.
Stage 2 (Acceptance): The charger is voltage regulated, at the bulk setpoint, and current sensed. The battery's internal resistance will determine the level of current that flows into the battery. When this current decays to 2% of the battery's (or bank's) nominal capacity, and holds there for 2 minutes or so, the charger shifts to the next stage.
Stage 3 (Float): The charger is voltage regulated at the float setpoint (about 13.4-13.5VDC at 70 degrees F), and voltage sensed at that point. Output will be adjusted to maintain that point as external loads are added or removed. Essentially, the charger is acting as a voltage eliminator at this point.
If you were to graph current and voltage, you'd see:
Bulk: current rises to the setpoint current and stays there until just before the end of the cycle, while voltage rises quickly (from 12.whatever to about 13.7VDC) and then more slowly to the bulk setpoint (14.4 nominal).
Acceptance: voltage holds at the bulk setpoint and current slowly decays to the 2% point.
Float: initially, the charger shuts down completely, as battery terminal voltage decays from 14.4 to 13.5, then voltage holds at this point and current starts at about 5-6A and slowly decays to (for a bank of 200-500 AH) about 400-500 mA. Voltage will momentarily fluctuate as loads are added and removed.
It is accepted that for any state of discharge up to 50% of nominal bank capacity, with a charger of adequate size (between 25% and 40% of bank nominal capacity), the bank will be restored to 80-85% of capacity by the end of the Acceptance cycle, which will usually take about 1 hour, and will be completely recharged within 1-2 hours after the shift to float cycle. A battery or bank attached to this type of charger can remain on the float cycle indefinitely, without damage to the battery.
With flooded cells, some electrolyte will have to be added periodically, since a small amount will be lost due to gassing during the Acceptance cycle. For gel cels and AGMs, the bulk setpoint is set at least 0.1-0.2VDC below gassing, and no maintenance is necessary (or possible).
"Smart" chargers are relatively recent devices. The first commercially successful one of which I am aware was brought to market by Cruising Equipment Co., in Washington, in the late 1980s and was designed for use on cruising yachts with high capacity alternators that employed external regulation. Cruising Equipment was acquired by Heart Interface Co. about 5 years later, and Heart was acquired by Xantrex a few years after that. The original Cruising Equipment regulator is sold today by Heart under the names "Alpha" and "In-Charge."
AC-supplied chargers that accurately follow the "smart" charge profiles are available under the names Heart and StatPower (both now owned by Xantrex), Trace, Newmar, and maybe a few others. Some that claim to be smart chargers (notably those from Tripp-Lite) do not in fact perform as required.
Now what happens if you try to charge a deep cycle battery with fixed voltage regulation? One of two things. If the voltage setpoint is too low, the battery will never charge. I once saw a calc that concluded that if you hooked a 100AH 12V battery discharged by 50AH to a fixed-voltage charger set at 13.8VDC, it would take two weeks to achieve the 80% charged state and you would never get charge higher than about 90%. After a few cycles of not being fully recharged, some of the lead paste on the plates changes to crystalline lead sulfate, and the battery's nominal capacity is permanently reduced. Under-charging is the major cause of premature failure of deep-cycle batteries in uses (like marine) where substantial discharging occurs on a regular basis.
On the other hand, if the fixed-voltage setpoint is too high, the battery rises to the 80% charged point (albeit quite slowly) and then remains above the gassing point as long as the charger is attached. This causes the evaporation of electrolyte, exposing of the plates, and physical destruction of the plates. Overcharging is the major cause of premature failure of deep cycle batteries in uses where substantial discharging does not normally occur (such as standby power supplies).
In an automobile, a fixed-voltage regulator regulates the alternator. However, the life of an automobile starting battery is very different from that of a deep cycle storage battery. The starter, while imposing a load as high as 200-300A, runs for only a few seconds, and it takes less than 1 AH out of the battery. That is quickly restored by a fixed voltage regulator set at, for instance, 13.7 or 13.8VDC, since current will flow at a high rate during the first few minutes of charge; after that, battery terminal voltage will rise to the setpoint, and net current flow into the battery will be quite low (usually less than 1-2A). At this rate, the battery would be cooked if left on the charger indefinitely, but in general the engine will not be running for more than a couple of hours before being shut down. The fixed voltage setpoint in automotive alternator regulators is a compromise. However, it is considered acceptable if the battery lasts for 3-5 years.
You can now see that, if you left your station's standby batteries attached to a charger with a fixed voltage set at 14.4, the batteries would quite quickly be cooked to death. (In fact, if they were flooded cell batteries, you'd also generate a dangerous amount of hydrogen gas, and most likely overheat the batteries to the point where the cases cracked and spilled the remaining acid on the floor.) If you hook it up to a fixed voltage at a 13.5VDC setpoint, the batteries will never recover from a charge.
I do not claim to have bench tested every charger built into every radio that claims to be able to charge and maintain revert batteries. But I have looked at a lot of them, and I have never encountered a true multi-stage smart charger. That is why I set out to build my chargers.
In order to charge deep cycle storage batteries, you need to apply different charge curves to them, depending on the progress of the charge. The accepted curves are as follows (referencing a 12V battery):
Stage 1 (Bulk): The charger is current regulated, at about 90-95% of its rated capacity, and voltage sensed. The voltage trip point is set to the point just at the point where the battery "gasses," which for a flooded cell battery is 14.4 VDC at 70 degrees F. When battery terminal voltage reaches this point and stays there for 2 minutes or so, the charger shifts to the next stage.
Stage 2 (Acceptance): The charger is voltage regulated, at the bulk setpoint, and current sensed. The battery's internal resistance will determine the level of current that flows into the battery. When this current decays to 2% of the battery's (or bank's) nominal capacity, and holds there for 2 minutes or so, the charger shifts to the next stage.
Stage 3 (Float): The charger is voltage regulated at the float setpoint (about 13.4-13.5VDC at 70 degrees F), and voltage sensed at that point. Output will be adjusted to maintain that point as external loads are added or removed. Essentially, the charger is acting as a voltage eliminator at this point.
If you were to graph current and voltage, you'd see:
Bulk: current rises to the setpoint current and stays there until just before the end of the cycle, while voltage rises quickly (from 12.whatever to about 13.7VDC) and then more slowly to the bulk setpoint (14.4 nominal).
Acceptance: voltage holds at the bulk setpoint and current slowly decays to the 2% point.
Float: initially, the charger shuts down completely, as battery terminal voltage decays from 14.4 to 13.5, then voltage holds at this point and current starts at about 5-6A and slowly decays to (for a bank of 200-500 AH) about 400-500 mA. Voltage will momentarily fluctuate as loads are added and removed.
It is accepted that for any state of discharge up to 50% of nominal bank capacity, with a charger of adequate size (between 25% and 40% of bank nominal capacity), the bank will be restored to 80-85% of capacity by the end of the Acceptance cycle, which will usually take about 1 hour, and will be completely recharged within 1-2 hours after the shift to float cycle. A battery or bank attached to this type of charger can remain on the float cycle indefinitely, without damage to the battery.
With flooded cells, some electrolyte will have to be added periodically, since a small amount will be lost due to gassing during the Acceptance cycle. For gel cels and AGMs, the bulk setpoint is set at least 0.1-0.2VDC below gassing, and no maintenance is necessary (or possible).
"Smart" chargers are relatively recent devices. The first commercially successful one of which I am aware was brought to market by Cruising Equipment Co., in Washington, in the late 1980s and was designed for use on cruising yachts with high capacity alternators that employed external regulation. Cruising Equipment was acquired by Heart Interface Co. about 5 years later, and Heart was acquired by Xantrex a few years after that. The original Cruising Equipment regulator is sold today by Heart under the names "Alpha" and "In-Charge."
AC-supplied chargers that accurately follow the "smart" charge profiles are available under the names Heart and StatPower (both now owned by Xantrex), Trace, Newmar, and maybe a few others. Some that claim to be smart chargers (notably those from Tripp-Lite) do not in fact perform as required.
Now what happens if you try to charge a deep cycle battery with fixed voltage regulation? One of two things. If the voltage setpoint is too low, the battery will never charge. I once saw a calc that concluded that if you hooked a 100AH 12V battery discharged by 50AH to a fixed-voltage charger set at 13.8VDC, it would take two weeks to achieve the 80% charged state and you would never get charge higher than about 90%. After a few cycles of not being fully recharged, some of the lead paste on the plates changes to crystalline lead sulfate, and the battery's nominal capacity is permanently reduced. Under-charging is the major cause of premature failure of deep-cycle batteries in uses (like marine) where substantial discharging occurs on a regular basis.
On the other hand, if the fixed-voltage setpoint is too high, the battery rises to the 80% charged point (albeit quite slowly) and then remains above the gassing point as long as the charger is attached. This causes the evaporation of electrolyte, exposing of the plates, and physical destruction of the plates. Overcharging is the major cause of premature failure of deep cycle batteries in uses where substantial discharging does not normally occur (such as standby power supplies).
In an automobile, a fixed-voltage regulator regulates the alternator. However, the life of an automobile starting battery is very different from that of a deep cycle storage battery. The starter, while imposing a load as high as 200-300A, runs for only a few seconds, and it takes less than 1 AH out of the battery. That is quickly restored by a fixed voltage regulator set at, for instance, 13.7 or 13.8VDC, since current will flow at a high rate during the first few minutes of charge; after that, battery terminal voltage will rise to the setpoint, and net current flow into the battery will be quite low (usually less than 1-2A). At this rate, the battery would be cooked if left on the charger indefinitely, but in general the engine will not be running for more than a couple of hours before being shut down. The fixed voltage setpoint in automotive alternator regulators is a compromise. However, it is considered acceptable if the battery lasts for 3-5 years.
You can now see that, if you left your station's standby batteries attached to a charger with a fixed voltage set at 14.4, the batteries would quite quickly be cooked to death. (In fact, if they were flooded cell batteries, you'd also generate a dangerous amount of hydrogen gas, and most likely overheat the batteries to the point where the cases cracked and spilled the remaining acid on the floor.) If you hook it up to a fixed voltage at a 13.5VDC setpoint, the batteries will never recover from a charge.
I do not claim to have bench tested every charger built into every radio that claims to be able to charge and maintain revert batteries. But I have looked at a lot of them, and I have never encountered a true multi-stage smart charger. That is why I set out to build my chargers.
-
ASTROMODAT
- Posts: 1825
- Joined: Tue Nov 05, 2002 12:32 am
The Emergency Revert Battery Option on the Quantar is only $300 (not including the 24 V battery bank, of course). This will be the best $300 you have ever spent. Factory designed and integrated to work, when lives are at stake! We have used our for years without a hitch. Via the Motorola Quantar remote RSS, you can remotely select Float and Equalize. Works like a champ, and our original batteries are in great shape.
Larry
Larry
"Equalizing" means a special charging cycle designed to force as much of the hardened lead sulfate on the surface of the plates back into useable lead paste. It involves subjecting the batteries to an extended charge that is current regulated at 3-4% of the battery's nominal capacity, for a timed period (generally about 3-4 hours). During an equalization charge, the battery terminal voltage will rise to 15-16 volts, and the batteries will gas heavily. The conditions for running an equalization charge are:
1. All DC loads must be removed from the batteries (as the high voltage involved can damage some solid state devices).
2. The electrolyte must be checked and topped as necessary before the cycle is started, and after the cycle has ended (following a period of about 1 hour of cool down).
3. Hydrocaps should be removed from the cells and the cell openings covered with paper towels, to catch any spattering electrolyte.
4. The battery containment should be power vented and all ignition sources excluded.
5. Voltage should be constantly monitored and the cycle terminated if voltage rises above 16.5-17.0VDC.
6. An equalization charge can only be applied to flooded cells with open cells, and never to gel cels, AGMs or so-called "sealed" or "maintenance free" flooded cells.
These conditions preclude equalizing remotely.
In general, equalizing charges should be performed once or twice a year (or season) to batteries (or banks of batteries) that regularly see deep discharges. An equalization charge does nothing useful to a battery (or bank) that spends virtually all of its time in standby.
The Quantar manual refers to equalizing batteries for "typically 48 to 72 hours." (68P81085E35-R, p. 4-140.) Long before one got to 48 hours into an equalizing charge, the electrolyte would have entirely boiled away, the batteries would have cracked or exploded, and the building might well have been set on fire. This reference is an indication that, however much Motorola knows about radios, it is not a reliable reference about battery maintenance.
See http://www.xantrex.com/support/docserve.asp?id=736
1. All DC loads must be removed from the batteries (as the high voltage involved can damage some solid state devices).
2. The electrolyte must be checked and topped as necessary before the cycle is started, and after the cycle has ended (following a period of about 1 hour of cool down).
3. Hydrocaps should be removed from the cells and the cell openings covered with paper towels, to catch any spattering electrolyte.
4. The battery containment should be power vented and all ignition sources excluded.
5. Voltage should be constantly monitored and the cycle terminated if voltage rises above 16.5-17.0VDC.
6. An equalization charge can only be applied to flooded cells with open cells, and never to gel cels, AGMs or so-called "sealed" or "maintenance free" flooded cells.
These conditions preclude equalizing remotely.
In general, equalizing charges should be performed once or twice a year (or season) to batteries (or banks of batteries) that regularly see deep discharges. An equalization charge does nothing useful to a battery (or bank) that spends virtually all of its time in standby.
The Quantar manual refers to equalizing batteries for "typically 48 to 72 hours." (68P81085E35-R, p. 4-140.) Long before one got to 48 hours into an equalizing charge, the electrolyte would have entirely boiled away, the batteries would have cracked or exploded, and the building might well have been set on fire. This reference is an indication that, however much Motorola knows about radios, it is not a reliable reference about battery maintenance.
See http://www.xantrex.com/support/docserve.asp?id=736
Last edited by RKG on Thu Jan 01, 2004 8:13 pm, edited 1 time in total.
-
ASTROMODAT
- Posts: 1825
- Joined: Tue Nov 05, 2002 12:32 am
We have Equalized our 24 V battery bank with the stock Quantar PS many times, without ever boiling away any electrolyte. During blackouts at our site, the battery bank has always pereformed flawlessly. Follow Motorola's instructions, and you should be fine.
Motorola has a nice System Planner on the Float/Equalize functions of the Quantar PS, and appropriate battery bank sizing guidelines. I will try to find the PDF and post it---it's chalk full of good information in this area.
Larry
Motorola has a nice System Planner on the Float/Equalize functions of the Quantar PS, and appropriate battery bank sizing guidelines. I will try to find the PDF and post it---it's chalk full of good information in this area.
Larry
Power Plants for Quantars and MTR2000
Hey RKG:
None of the 600A power plants I look after (hundreds?) have three stage charge levels like you mention. To your point, Statpower, Heart Interface and others have introduced this clever way of safely fast charging batteries that are used in cyclic applications. But all my telecom power plants have only two setpoints: FLOAT and EQUALIZE. One can set how often the EQ happens and for how long... or after any power outage. The power plant controllers are pretty slick nowadays. Since we use VRLA/AGM at remote sites, none of the battery vendors recommend use of EQUALIZE. Just leave on FLOAT. The charge tapers off exponentially and can take a long time to get to 100%, but it does get there.
The switch sites have flooded cells, and that is a completely different beast as you mention. But there is no way we would disconnect the load from the battery. In some cases we can isolate a string before EQ. I tend to build my switch sites this way. But then you need to bring in a separate rectifier to do the EQ, since all resident ones are powering the load.
BTW... Powerware makes lots of nice, big UPS systems. Last few years, they are using a charging agorithm that basically equalizes the battery, then disconnects the charger for a period of weeks. The battery is monitored. Any outages, or if certain voltage thresholds are reached, the charge cycle is repeated. They claim this prolongs battery life over that of a constant float current.
So the QUANTAR AC PS is slick in it's capabilities with the battery option. Too bad the MTR is not!
RFDude
None of the 600A power plants I look after (hundreds?) have three stage charge levels like you mention. To your point, Statpower, Heart Interface and others have introduced this clever way of safely fast charging batteries that are used in cyclic applications. But all my telecom power plants have only two setpoints: FLOAT and EQUALIZE. One can set how often the EQ happens and for how long... or after any power outage. The power plant controllers are pretty slick nowadays. Since we use VRLA/AGM at remote sites, none of the battery vendors recommend use of EQUALIZE. Just leave on FLOAT. The charge tapers off exponentially and can take a long time to get to 100%, but it does get there.
The switch sites have flooded cells, and that is a completely different beast as you mention. But there is no way we would disconnect the load from the battery. In some cases we can isolate a string before EQ. I tend to build my switch sites this way. But then you need to bring in a separate rectifier to do the EQ, since all resident ones are powering the load.
BTW... Powerware makes lots of nice, big UPS systems. Last few years, they are using a charging agorithm that basically equalizes the battery, then disconnects the charger for a period of weeks. The battery is monitored. Any outages, or if certain voltage thresholds are reached, the charge cycle is repeated. They claim this prolongs battery life over that of a constant float current.
So the QUANTAR AC PS is slick in it's capabilities with the battery option. Too bad the MTR is not!
RFDude
By definition, equalizing flooded cells requires raising their terminal voltage well above the gassing point. If you are truly equalizing, therefore, you must be boiling away some electrolyte, and any charge cycle, however labelled, that does not boil away any electrolyte cannot equalize the cells.
In the laboratory, I have watched G27 batteries intentionally abused by subjecting them to an excessive 4% current overcharge, in order to see what happens. Within 6-8 hours (and in some cases as quickly as 2 hours), the electrolyte above the plate level was completely gone. After about 12-16 hours, more than half the plate area was exposed. The chamber smelled awful, and the battery cases were too hot to touch. Before 24 hours, the battery cases had either split or melted (depending on the material used in their construction); those batteries that had any remaining electrolyte spilled it into the containment. All of the batteries were physically destroyed beyond repair. (In one or two cases, as I recall, an intercell connector physically separated from the cell plate, by warpage, before the battery case failed. This prevented a failure of the case, but it rendered the battery destroyed.)
In the laboratory, I have watched G27 batteries intentionally abused by subjecting them to an excessive 4% current overcharge, in order to see what happens. Within 6-8 hours (and in some cases as quickly as 2 hours), the electrolyte above the plate level was completely gone. After about 12-16 hours, more than half the plate area was exposed. The chamber smelled awful, and the battery cases were too hot to touch. Before 24 hours, the battery cases had either split or melted (depending on the material used in their construction); those batteries that had any remaining electrolyte spilled it into the containment. All of the batteries were physically destroyed beyond repair. (In one or two cases, as I recall, an intercell connector physically separated from the cell plate, by warpage, before the battery case failed. This prevented a failure of the case, but it rendered the battery destroyed.)