needed idiots guide to powering with batteries

[looboo]

Hello
after recently acquiring a Pro-ject RS preamp and being thoroughly disappointed with it I decided to try powering it with some 6v sla batteries that I had knocking about, all I did was wire 3 in parallel to get the 18v required by the pre wired them to a plug and stuck it into the preamp,well the difference was not subtle so now I would like to make a proper supply that can charge the batteries also I keep reading references to smoothing caps? so does anybody fancy writing me up a how to , I am useless with electronics and maths but I can read a schematic and use a soldering iron.
any help would be greatfully received.

 

[Rundmaus]

all I did was wire 3 in parallel to get the 18v [...] well the difference was not subtle

If you really wired them *in parallel*, you supplied the preamp with 6V instead of 18V, which would account for a huge difference in sound...

also I keep reading references to smoothing caps?

Smoothing caps are only useful with AC-based supplies, where they filter out the ripple remaining after rectification of the AC source. With a pure DC supply like a SLA battery, no smoothing caps are required.

Rundmaus

 

[looboo]

Hi .thanks for the info,

If you really wired them *in parallel

I said I was useless but I did manage to get 20.4 volts out of them so should of said in series.

 

[Rundmaus]

Ok, sounds reasonable then...

I have been using SLA batteries on a regular basis for years now, and I usually do a simple constant-voltage charging, with an additional upper current limit.

What capacity are your batteries? And how much current does the preamp draw? I think it is useful to estimate first how long a set of batteries will be able to power the amplifier.

If you need recharging within some hours, some automated charging circuit will be needed for convenience. If we're talking about weeks, a simple manual constant voltage charger will do - maybe along with a second set of SLA batteries.

Rundmaus

 

[kevinkr]

Make sure you install a fuse quite close to the batteries to protect against shorts. You do not want to see what happens when you inadvertently short the output connections together.

You can charge the batteries in series connection provided that you limit current to a reasonable value (consult the specifications for the battery for charge rate) and make sure that the supply delivers between 2.2 - 2.3V per cell for full charging - if you intend to leave the charger connected in float mode make sure that you do not exceed the recommended charging voltage for this mode.

A simple charger based around something like an LT1084 with current limiting to 1/5 - 1/10th C (rated capacity of the battery) should work fine. Limit voltage to ~ 20V across the battery terminals. Make sure there is a high current diode (Schottky) in series with the output of charger circuit to protect the charger circuitry.

This is very similar to a circuit I designed about 30yrs ago to charge SLA batteries which worked quite well in a large scale budget conscious application. Note that there is no attempt at temperature compensation.
Lead Acid Battery Charger Circuit

Add that schottky diode in series with the output as I mentioned - use at least a 3A diode and include a fuse!

R1 0.5 ohms will provide a charging current of about 1.2 - 1.3A change to 1 ohm for half that.

Change R2 to 1K - with no battery connected adjust the voltage to ~20.3V to compensate for the series diode drop, add a 2A medium or fast blow fuse in series with the output for charging currents of <2A.

Use an LT1084 not the LM317 shown unless charging currents are intended to be less than 1.5A..

Input voltage should be at about 23V while charging. The input capacitor can be much smaller that 10000uF without any penalties in performance.

 

[looboo]

hi Rundmaus, the batteries I have are 6v 1ah
the amps power supply is a plug top thingy the qutpus 18v dc 1amp
the amp manual only states a power consumption figure of less than 1 watt at idle
and a big warning to only use the supplied power supply, OOPS there goes me warranty

 

[kevinkr]

You can use the circuit I referenced above to charge these batteries, but limit the charge rate as stated so you'd want no more than 200mA I'd say. An R1 of 3 - 4 ohms should be fine..

 

[looboo]

Hi kevinkr,
thanks for the circuit and component changes that looks like just what I need.

 

[kevinkr]

Let me know what the maximum charging rate is for those batteries and I will give you the requisite value for R1..

 

[gootee]

<snipped>

Smoothing caps are only useful with AC-based supplies, where they filter out the ripple remaining after rectification of the AC source. With a pure DC supply like a SLA battery, no smoothing caps are required.

Rundmaus

Actually, that probably isn't true, and also sort-of misconstrues the reasons why we need power supply capacitors.

If we call them reservoir caps, instead, we will probably see that we still need them. And we would definitely still want decoupling caps, which are just local reservoir caps, located close to the active devices.

The caps are there to accurately provide current, on demand, and act as a very low-impedance source for that current, which enables them to provide large currents with very fast rise-times. The capacitor currents ARE the music signals that are heard from the speakers, most of the time, when a regular power supply is used.

The power rail voltage waveform is relatively unimportant (unless PSRR is a big problem). It's the ability to accurately supply current, with the needed amplitude and bandwidth, that matters, for an audio power amp's power supply.

A battery might have a significant resistance or inductance, compared to a bank of reservoir capacitors, in which case you would probably want the capacitors, to help prevent the rail voltage from being pulled low (ripple) by large music currents, and to prevent having voltage spikes induced when fast-rising currents are needed for the speakers, and to prevent transient and harmonic distortion if the battery cannot supply current fast-enough. Most amplifiers need to be able to pull current accurately at up to hundreds of kiloHertz, to keep the internal feedback loops happy, and also to be able to supply the worst-case transient currents for the output (e.g. zero to rail at maximum slew rate).

There are plots that I have seen, on line, that show that batteries by themselves do not even perform as well as three-terminal regulators.

Batteries are not perfect "pure DC" power sources, and can provide significant noise, as well as cause voltage ripple, transient inaccuracy, and harmonic distortion, when supplying dynamic load currents through a typical power amplifier, which the caps would help to mitigate.

 

[paulb]

You could try something like this. Based on the UC3906.

 

[Rundmaus]

Gootee,

we're talking about a preamp here. For the usual current demands of a preamp circuit, the low impedance of a SLA battery is as close to a perfect DC source as needed in any practical situation.

For a power amp circuit, you are probably right, although I doubt that an electrolytic capacitor has a significantly lower impendance than a sufficiently beefy lead-acid battery. Local decoupling at the load site makes sense, of course, and should always be done.

Conventional lead-acid and to a certain extent SLA batteries have very low internal resistances and can supply high currents without significant voltage drop.

Battery 'noise' and voltage ripple from a DC source belong to the world of audio myths, IMHO.

Rundmaus

 

[gootee]

You might be right. But it couldn't hurt the performance, to include some reservoir capacitance. And it would also be easy and cheap to try it both ways, to check whether or not any difference could be measured, or heard.

 

[Rundmaus]

But it couldn't hurt the performance, to include some reservoir capacitance.

Of course it wouldn't. Preferably by using some low ESR caps close to the load for local decoupling, if there's enough space available.

Rundmaus

 

[looboo]

Thanks for all the replies

Let me know what the maximum charging rate is for those batteries and I will give you the requisite value for R1..

Hi Kevinkr ,there is no max charge rate specified only an initial charge rate of <0-3A.

Originally Posted by gootee
But it couldn't hurt the performance, to include some reservoir capacitance.
Of course it wouldn't. Preferably by using some low ESR caps close to the load for local decoupling, if there's enough space available.

Rundmaus

If decoupling caps are beneficial how many do I need ? what value? and how would I wire them in?
thanks again for the help

 

[AndrewT]

Actually, that probably isn't true, and also sort-of misconstrues the reasons why we need power supply capacitors.

If we call them reservoir caps, instead, we will probably see that we still need them. And we would definitely still want decoupling caps, which are just local reservoir caps, located close to the active devices. .........

Once again Gootee is right.
Always use a cap across a battery supply. Even if all it does is attenuate the noise coming from the battery.

Conventional lead-acid and to a certain extent SLA batteries have very low internal resistances and can supply high currents without significant voltage drop.

Battery 'noise' and voltage ripple from a DC source belong to the world of audio myths

is all about promulgation of myths.

 

[looboo]

Hello ,
I am still struggling with this

Once again Gootee is right.
Always use a cap across a battery supply. Even if all it does is attenuate the noise coming from the battery.

how would I wire the caps in and what value/type should they be ,the preamp needs 18v 1ah .thanks for any help.

 

[gootee]

The value is probably not critical, except that it probably shouldn't be tiny. The cap voltage rating should be a little higher than the maximum possible actual battery voltage.

At the battery connection, use an electrolytic cap with its + lead connected to the + battery connection and its negative to battery negative. If you have several, paralleling them would lower the impedance and might be better. You could use as much capacitance as would physically fit. But a few hundred uF would probably be plenty, and a huge amount might have inrush current problems at power up time.

Keeping all connections and wiring short will help. And keeping the positive and negative always close together is important, so you don't make antennas. (That is especially important for your input signal and input signal ground conductor pair. Twist them tightly together.)

Right at any active devices, you need both high frequency bypass and some decoupling. So you want one tiny cap, extremely close to each device (within a couple of mm) and a small electrolytic in parallel with that, within a centimeter or so.

The tiny bypass cap should be physically small, between its leads, and is usually anything from 0.01 to 0.1 uF. Usually, the best type for that is an X7R ceramic but you could try a film cap.

The paralleled electrolytic value would normally be calculated to have enough uF for low frequency current demands and low-enough inductance (ie lead spacing plus round trip of mounting distance from device) but for a preamp i would guess that anything at or above 47 uF should always be sufficient.

If you had the space, it would probably be better to use four or five 10 uF in parallel. That would give a lower inductance and resistance. It would be overkill but this is diyaudio and it should "theoretically" improve the high frequency transient response and also improve the feedback's ability to kill HF harmonic distortion.

 

[paulb]

This bears repeating:

Make sure you install a fuse quite close to the batteries to protect against shorts. You do not want to see what happens when you inadvertently short the output connections together.

 

[Rundmaus]

Besides any capacitor discussion: Keep in mind that SLA batteries can deliver significant currents into low-resistance loads. Enough to turn a shorting capacitor or other components into impressive fireworks. A fuse should be the first thing coming after the battery terminals.

Rundmaus

EDIT: Cross-posted with paulb. But the fuse thing is important enough to be said repeatedly.