Charging/maintaining 18v of SLA battery with 20v laptop PSU?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm about to set up an 18v SLA supply for my 2 SMSL SA36A Pros using new 12V and 6V batteries (both will be new 12AH Yuasas) in series.

(Why not 3 x 6V? tidier and simpler, the Yuasa's are the same dimensions except depth).

I have a largish (6A) 20V laptop power supply which I've been using to power the SMSL's. it's very tightly regulated - 20.1V open and almost no drop under load, which is to say with the 2 SMSL's playing bass-heavy music at or near flat-out, (2-3A RMS?).

Yes, the SMSL's sound fine with the brick, but having had nice results using a 12V SLA with both them and various Tripaths, I'd like to try them with 18V (actually nearer 19V) battery power.

So:- coincidentally, 20V happens to be as near as dammit exactly the float for 18V of SLA at the temperatures they'd likely be operating at.

Assuming I was *careful* and didn't let them discharge to less than c.50% (i.e. keep them over 18V), which would easily allow an evening (maybe an entire day) of listening at quite high volume, would the laptop brick effectively 'trickle-charge' and then float-charge them if I left it on overnight?

I've been trying to figure out how much current the batteries would draw from the PSU with that 2V differential.

IOW, could the batteries draw too much current for their own or the PSU's good?

Conversely, could they draw too little from 20V to charge overnight from c. 50%?

The reasoning for all this is that it adds up to a simple and **cheap!** topology. It would save having to buy smart/maintenance chargers which would otherwise be the most expensive components.

Even if I didn't already have the brick, good ones can be had for £10-20. Nothing more needed than a few spade connectors, a couple of inline fuses and switches (which I already have) and I would never be without power for the amplifiers (well, as long the grid's still up).

I'm going to try it anyway and if it *seems* to work, initially taking frequent voltage readings (and maybe even current readings), but I'm interested in hearing in advance from anyone more experienced/knowledgeable as to whether it's feasible (I won't have the batteries until into the new year) .
 
Last edited:
Administrator
Joined 2007
Paid Member
The risk with series connecting is that charge inequality will result in some cells in the chain seeing to much voltage as they reach a fully charged state. This will also reduce the current available to the other less charged cells.

Not a good idea imo.

As to charge currents, a discharged or partially discharged battery in good shape (with low internal resistance) will draw pretty much whatever the charger can supply, with the batteries drawing up to 10's of amps if it were available. So if the charger isn't limited then you could well run into problems there too.
 
The risk with series connecting is that charge inequality will result in some cells in the chain seeing to much voltage as they reach a fully charged state. This will also reduce the current available to the other less charged cells.

Not a good idea imo.
Yes, I've considered that. But the Yuasa's will be identical in everything but number of cells, as close as you can get to a single 'true' 18V SLA battery (ETA > apart from 3 identical 6V, of course).
As to charge currents, a discharged or partially discharged battery in good shape (with low internal resistance) will draw pretty much whatever the charger can supply, with the batteries drawing up to 10's of amps if it were available. So if the charger isn't limited then you could well run into problems there too.

This is the nub. Do you have an idea of how much current an 18V/12AH battery would draw at c. 18.2V from 20V?

What I might try at first is charging from gradually diminishing battery charge/voltage and see if there are signs of distress from the brick. I assume the worst that's likely to happen is waving it bye-bye.

I guess that I'd pretty much have the PSU on all the time, acting as a float charger (which I'm pretty certain would be fine), only switching it off for 'critical' listening (assuming the amps do actually sound better with the batteries), hopefully rarely getting below c. 80%.
 
Last edited:
I'm trying to apply my pretty basic understanding of electricity (V/I/R), but battery physics/chemistry complicates things beyond it;

So (JonSnell) bulk-charging of an 18V 12AH SLA would be at about 22V but current limited to 1.2A (optimally, for long battery life - I know occasional fast charging isn't disastrous).

The question is whether a lower voltage, i.e. the tightly-regulated 20V from the brick, intrinsically current-limits as long as the battery is only partly discharged?
 
Administrator
Joined 2007
Paid Member
The Yuasa document you want is this one View attachment Yuasa Little Red Book of Batteries.pdf

Even reducing the charge level by a few percent on a SLA battery means that it will still draw huge current for a short time from the charger.

One possible solution would be to include a shottky diode in the feed from the battery to the amp and then parallel that diode with a low value resistor to limit the charge current when its charging. Its a compromise solution but would safeguard the charger.
 
The Yuasa document you want is this one View attachment 520469

Even reducing the charge level by a few percent on a SLA battery means that it will still draw huge current for a short time from the charger.

One possible solution would be to include a shottky diode in the feed from the battery to the amp and then parallel that diode with a low value resistor to limit the charge current when its charging. Its a compromise solution but would safeguard the charger.


Ah - you pre-empted. I'd heard about the use of diodes. That would drop the voltage the amp sees from the battery by 0.5-0.6v?

I'm beginning to see the issues, and why so much effort goes into designing good charging systems - noticed that internal resistance in SLA's is measured in mΩ, hence even a couple of volts can lead to a LOT of current flowing.

Still, worth a try - at worst I'll be forced to proper smart chargers.

ETA >> forgot; thanks for the link, very useful indeed.
 
Last edited:
Can you explain your mode of operation for the charger and battery?

Were you going to turn off and disconnect the laptop supply from the mains at the time of listening? Or were you always going to have the battery connected to the charger when listening? If so, then it may be easier just to use a power resistor (eg. 2.2 ohm 30W).

The recharge current of a battery is dependant on the state of charge, and battery impedance, and charging voltage. In your case, if the battery is sufficiently discharged then the battery recharge current will initially be limited by the charger current (6A). Does the manufacturer say anything about recharge current limit - these may be UPS application batteries but may have a recommended max recharge current of 20-30% capacity rating.

Nowadays you can buy quite cheap and small 6V/12V chargers for car/motorbike batteries that are nicely regulated and automated for identifying the battery voltage. The point being, that a periodic connection of each of your monoblocs to that charger for the purpose of a maintenance charge to ensure full state of charge is reached would be of benefit, otherwise a weak cell may slowly eventuate in a monobloc and not be recovered by your laptop charger alone.
 
I'm about to set up an 18v SLA supply for my 2 SMSL SA36A Pros using new 12V and 6V batteries (both will be new 12AH Yuasas) in series.

I'd would have the 20.1 v PS attached to the batteries all the time - whether the amps are in use or not. It will not hurt the batteries. They will remain charged with the PS supplying the amps' operating current and the batteries providing any extra amps for music peaks - much a like a capacitor. Here's my reasoning ...


SLA-AGM batteries have calcium added to the chemistry. This increases their operating voltage by 5%. See the attached "State-of-Charge" chart.

SoC By Voltage - AGM and Flooded (1).jpg

So at full charge, a 12v battery is at 13.3 v and the 6v battery is at 6.65v for a total of 19.95v. The PS at 20.1v is only .15V higher. This is negliagble and won't hurt the battery. 12V float charges have tolerances greater than this w/o battery damage. I just measured my SLA charger on float and it showed 13.8v. The flooded lead-acid battery in your, car when fully charged, is at 12.65 and the alternator consistently puts out 14.4v w/o ill effects.

Try it. If your skeptical, you can always test for a day and check to see if the batteries get warm.

If you routinely run on batteries alone and use the PS to charge, then limiting the charge current with a resistor would be warranted as others have stated. I'd use the 25% SOC value for the discharged batteries (19.59v) against the 20.1v PS for a voltage differential of 0.5v and limit the current to 2 amps for 12ah batteries (1/6th charge current is my rule-of-thumb). A 1 ohm resistor would give 0.5 amps but could take more than a day to charge. You can have a resistor in-line with a bypass switch - open for charging, closed for continuous operation/hibernation.
 
Last edited:
Ok, so your aim is to provide additional power supply smoothing, over and above what the PS can provide, and what any filter caps within your amp modules can provide.

Did you look at what filtering/storage is in your amp modules, and what the transient current performance of your PS is, and what the voltage sag or ripple was at the power terminals of your module when you apply a high power transient or test signal?

The battery will only provide filtering support 'current' to any significant degree when its terminal voltage is lowered by quite a substantial level. Yes it has some capacitive performance, but I doubt that capacitance would be noticeable in your system.

As such, I doubt the battery would 'discharge' to any level at all during a playing session, nor would it noticeably contribute to your existing power system's performance.

Battery cell strings left on continuous float can slowly exhibit cells that walk down in capacity - depending on quality of battery, duty of battery, temperature and float voltage, and number of cells in string. The comment about using a fit-for-purpose charger is that it is likely to include a maintenance equalise voltage charge to somewhat address weak cells developing.
 
I'm about to set up an 18v SLA supply for my 2 SMSL SA36A Pros using new 12V and 6V batteries (both will be new 12AH Yuasas) in series.

(Why not 3 x 6V? tidier and simpler, the Yuasa's are the same dimensions except depth).

I have a largish (6A) 20V laptop power supply which I've been using to power the SMSL's. it's very tightly regulated - 20.1V open and almost no drop under load, which is to say with the 2 SMSL's playing bass-heavy music at or near flat-out, (2-3A RMS?).

Yes, the SMSL's sound fine with the brick, but having had nice results using a 12V SLA with both them and various Tripaths, I'd like to try them with 18V (actually nearer 19V) battery power.

So:- coincidentally, 20V happens to be as near as dammit exactly the float for 18V of SLA at the temperatures they'd likely be operating at.

Assuming I was *careful* and didn't let them discharge to less than c.50% (i.e. keep them over 18V), which would easily allow an evening (maybe an entire day) of listening at quite high volume, would the laptop brick effectively 'trickle-charge' and then float-charge them if I left it on overnight?

I've been trying to figure out how much current the batteries would draw from the PSU with that 2V differential.

IOW, could the batteries draw too much current for their own or the PSU's good?

Conversely, could they draw too little from 20V to charge overnight from c. 50%?

The reasoning for all this is that it adds up to a simple and **cheap!** topology. It would save having to buy smart/maintenance chargers which would otherwise be the most expensive components.

Even if I didn't already have the brick, good ones can be had for £10-20. Nothing more needed than a few spade connectors, a couple of inline fuses and switches (which I already have) and I would never be without power for the amplifiers (well, as long the grid's still up).

I'm going to try it anyway and if it *seems* to work, initially taking frequent voltage readings (and maybe even current readings), but I'm interested in hearing in advance from anyone more experienced/knowledgeable as to whether it's feasible (I won't have the batteries until into the new year) .

How about using 18V cordless drill batteries and a charger? Get a broken drill with them and cut off the battery receiving part so you can just wire it up whilst being able to constantly swap batteries. Many modern li-ion batteries have balancing circuits to make sure their cells are charged correctly. I'm sure you could get some secondhand Ryobi +One batteries and charger for not too much.
 
Battery cell strings left on continuous float can slowly exhibit cells that walk down in capacity - depending on quality of battery, duty of battery, temperature and float voltage, and number of cells in string. The comment about using a fit-for-purpose charger is that it is likely to include a maintenance equalise voltage charge to somewhat address weak cells developing.

I don't disagree with the continuous float comment. However, given the voltages involved, the PS at 20.1 is not really a float charge for SLA-AGM, not a high enough voltage. Telco's routinely keep voltage on their battery backup infrastructure in their central offices.
 
The battery will come with a recommended float voltage. If the manufacturer provides good installation and maintenance documentation then that voltage should identify a temperature and tempcomp levels, and range of float, and recommended maintenance charging etc

Have you located that info, or are you just using generic info for predesign purpose.

Telcos use a wide range of batteries, but not for the purpose you are using them for. If you wanted a battery to provide current for transient loading from your amps, then LiIon provides a much narrower voltage window between no current (idle voltage) and a lower voltage that causes a substantial discharge current.
 
Can you explain your mode of operation for the charger and battery?

Were you going to turn off and disconnect the laptop supply from the mains at the time of listening? Or were you always going to have the battery connected to the charger when listening? If so, then it may be easier just to use a power resistor (eg. 2.2 ohm 30W).

The recharge current of a battery is dependant on the state of charge, and battery impedance, and charging voltage. In your case, if the battery is sufficiently discharged then the battery recharge current will initially be limited by the charger current (6A). Does the manufacturer say anything about recharge current limit - these may be UPS application batteries but may have a recommended max recharge current of 20-30% capacity rating.

Nowadays you can buy quite cheap and small 6V/12V chargers for car/motorbike batteries that are nicely regulated and automated for identifying the battery voltage. The point being, that a periodic connection of each of your monoblocs to that charger for the purpose of a maintenance charge to ensure full state of charge is reached would be of benefit, otherwise a weak cell may slowly eventuate in a monobloc and not be recovered by your laptop charger alone.

I am considering the chargers you suggest, but I'd need 2 - a 12v and a 6V. Decent ones come in at about £40 in the UK, so not prohibitive, but I often look back at the what I've spent on gear I've bought over the years on impulse .......

AS I said, I'll get the batteries first and see how it goes (with some careful experimentation).
 
I am considering the chargers you suggest, but I'd need 2 - a 12v and a 6V. Decent ones come in at about £40 in the UK, so not prohibitive, but I often look back at the what I've spent on gear I've bought over the years on impulse .......

AS I said, I'll get the batteries first and see how it goes (with some careful experimentation).


If your not using the batteries for portable operation, maybe you should consider just adding large capacitors instead of batteries to further smooth the PS.

I use batteries myself so I appreciate the absence of PS issues. But my application is always mobile and the batteries are charged after use (and its two 12v SLA-AGMs which simplifies the whole battery maintenance issue). If you haven't bought the batteries yet, maybe two 12v SLAs is a better solution. After all, you're now looking at replacing the 20.1v PS so no need to stay in that voltage range.

That would boost overall output and simplify charging. I charge my two 12V SLAs in parallel with one 3-stage charger. One lasted 9 years after being discharged at least 75% every use.
 
I am considering the chargers you suggest, but I'd need 2 - a 12v and a 6V.

Sorry for the confusion - I'll try to clarify:

I was making the comment that rather cheap chargers are available for general maintenance charging - such a charger was not intended to be the power source for your amps or be part of the permanent power system (PS, battery, amps). Eg. for a monthly equalise charge. The cheap charger I have is used for 6V and 12V monoblocs - it detects both types of battery. Link to manual
ALDI AUTO XS CPL 2054 USER MANUAL Pdf Download.

With respect to batteries for experimentation, you may easily be able to obtain second hand 12V 7Ah, or even 6V, batteries that are left over when UPS batteries are swapped out for commercial and IT equipment. Smallish UPS are used for many many applications, and the battery pack is often replaced every few years. In many instances, only a few of the 6V or 12V monoblocs are poor, and the rest are fine for more years of service (especially when charged from just a 6V or 12V charger). I have about 4 such monoblocs that I use for testing (heater powering; preamp powering, USB isolator powering,....).
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.