Power Draw of Ruby Amp?

[kevinng]

I'm trying to find the power draw of Ruby amp.

I'm planning to wire 6 AA batteries in series - giving 9 volts with a combined 2000-3000 mAh x 6 = 12000-18000 mAh.

Including a 2-inch speaker, how long do you think the amp will run?

[Rob Strand]

I'm planning to wire 6 AA batteries in series - giving 9 volts with a combined 2000-3000 mAh x 6 = 12000-18000 mAh.

Since the batteries are in series the voltage goes up but the total mAh is only that for a singe battery ie.  2000-3000 mAh.   If you wired the batteries in parallel then the voltage is the single cell voltage but the mAh would multiply by the number of cells;  <insert many caveats about paralleling batteries>.

The next issue is when the batteries run down the voltage drops and the amp will clip earlier, perhaps to a point where is doesn't sound good or isn't loud enough.    The reason I mention this is battery mAh ratings assume the battery is quite depleted.  For Alkaline batteries the end-point voltage which gives the rated mAh is 0.9V or so.    That means 6 * 0.9V = 5.4 V which means the amp's power is going to be somewhat reduced.

As a very simple calculation.  Look at the LM386 datasheet.   See output power graph 8 ohm, and 9V curve.    Output power 0.7W  Power dissipation 0.5W.   So that's 0.7 + 0.5 = 1.2W going in at 9V.   Average current = 1.2W / 9V = 130mA.

An optimistic estimate for run time is 2500 = 2500 / 130 = 20 hrs.    A realistic estimate could be less than half of that.

A trick to get more life is to start of with a higher voltage, say 12V.   Then when the battery dies off to 0.9V you have 8*0.9 = 7.2V.
(I haven't checked to see if the LM386 will overheat at 12V.)


Checkout,

Notice how each loss of 0.1V in the end-point voltage throws away a lot of capacity,



Here, we have an example showing how the low impedance of a battery can keep a flat voltage curve over usable period,
https://www.ronburyswildlife.com/2015/07/which-aa-batteries-to-use-in-ltl-acorn.html

[PRR]

The '386 was actually designed to give "decent" battery life with 16Ω load and 9V radio battery.

The idle current can be read right off the datasheet, your first stop.

Ah.... some junior engineer has totally screwed-up the output vs dissipation charts (lost the impedance labels). Kids today! OK, Fig 3 shows 9V battery in 8Ω makes up to 6V peak to peak, which is 2V RMS, squared is 4/8Ω or a half-watt. At this low supply voltage we expect 50% sine efficiency at best, so 1 Watt total demand on 9V battery. Or 1/9 Amps, 110mA.

Can you actually play FULL power long enough to drain a battery? No. For heavy-clipped speech/music, the average may be 1/3rd of max. For heavy-metal geetar, say 1/2. So 55mA.

As Rob corrected, your battery is 2,000 mA. 2000/55 is 36 Hours.

In EXTREME clipping we could guesstimate 4V in 8Ω half the time or 250mA. 8 Hours.

[Rob Strand]

Can you actually play FULL power long enough to drain a battery? No. For heavy-clipped speech/music, the average may be 1/3rd of max. For heavy-metal geetar, say 1/2. So 55mA.

Playing clean or cranked will change the results for sure.

Something which can save a lot of battery power is to roll-off the low end, either with a high-pass filter in the preamp or using a small output cap on the LM386.   The small speakers don't output much low-end anyway even if the low-end electrically makes it to the speaker.   If you cut the low end it's prevents wasting power due the strong low-end part of the spectra.   Some tweaking is required to balance the sound but I wouldn't be surprise to get 2 or more times the battery life.

[Mark Hammer]

An octet of AA batteries gives a decent lifespan, with an LM386 amp, and decent headroom.  I will also make a case for putting the whole thing into the most efficient and largest cab you can, where feasible.  I realize your plan is to power a 2" speaker, but even 2" speakers can have more and less bass (and resulting volume) from better and worse cabs.

[PRR]

> Something which can save a lot of battery power is to roll-off the low end

Remember: musicians! Roll-off the low end only so the mids can play louder. Especially guitarists will be running into clipping much of the time. Clipped 500Hz-5KHz will eat battery about as fast as clipped 80Hz-5KHz, but louder on the ear. (Especially, as you say, with tiny speakers.)

> An octet of AA batteries gives a decent lifespan, with an LM386 amp, and decent headroom.

Thanks for practical experience. 

> even 2" speakers can have more and less bass (and resulting volume) from better and worse cabs.

Everything is relative, but the bass limit of a direct radiator is ALL about its size. If the midrange is strong the bass will seem weak. If the midrange is weak it may be "flat" to very low frequency. A twelve is a happy size for guitar-band efficiency. A two-inch must either lack bass below 480Hz (2.6 octaves) or have its midrange efficiency dropped from ~~1% to 0.03%, a 15 dB hurt.

Ah, yes, an open-back two-inch will suck very bad, and very far up the scale.

[kevinng]

Thanks for the reply, I'd study and consider them for my design.

[kevinng]

Can you actually play FULL power long enough to drain a battery? No. For heavy-clipped speech/music, the average may be 1/3rd of max. For heavy-metal geetar, say 1/2. So 55mA.

Playing clean or cranked will change the results for sure.

Something which can save a lot of battery power is to roll-off the low end, either with a high-pass filter in the preamp or using a small output cap on the LM386.   The small speakers don't output much low-end anyway even if the low-end electrically makes it to the speaker.   If you cut the low end it's prevents wasting power due the strong low-end part of the spectra.   Some tweaking is required to balance the sound but I wouldn't be surprise to get 2 or more times the battery life.

Rob, where in the circuit and what capacitors do you think I should do this?

See image - do I add a cap to the circle in the diagram? What value cap do you think will work?

[kevinng]

Will 8 x AA batteries, 4 in series work for Ruby Amp?

I plan to connect a 1-inch speaker to it. That's 14.8v, will it fry the circuit?

[Rob Strand]

Can you actually play FULL power long enough to drain a battery? No. For heavy-clipped speech/music, the average may be 1/3rd of max. For heavy-metal geetar, say 1/2. So 55mA.

Playing clean or cranked will change the results for sure.

Something which can save a lot of battery power is to roll-off the low end, either with a high-pass filter in the preamp or using a small output cap on the LM386.   The small speakers don't output much low-end anyway even if the low-end electrically makes it to the speaker.   If you cut the low end it's prevents wasting power due the strong low-end part of the spectra.   Some tweaking is required to balance the sound but I wouldn't be surprise to get 2 or more times the battery life.

Rob, where in the circuit and what capacitors do you think I should do this?

See image - do I add a cap to the circle in the diagram? What value cap do you think will work?

The best place to start is C2 on your schematic, the one going to the speaker.  Try 100uF.   You might even go to 47uF but I suspect that might be going a bit far.   The schematic I have for the Ruby has some roll-off before the power amp already.

[To protect the innocent, the C2 cap has different designators on the various web schematics.]

[Rob Strand]

Will 8 x AA batteries, 4 in series work for Ruby Amp?

I plan to connect a 1-inch speaker to it. That's 14.8v, will it fry the circuit?

8xAA batteries should end-up being less than 12.8V.

A 1" speaker isn't going to have much lows at all.  Maybe the 47uF output cap is worth trying.

[Mark Hammer]

Will 8 x AA batteries, 4 in series work for Ruby Amp?

I plan to connect a 1-inch speaker to it. That's 14.8v, will it fry the circuit?

I ran a 386-based practice amp off that for years.  Never harmed anything.  Just make sure your electrolytic caps are rated at 25V or more.  A 16V rating (usually the next one down from 25V) may not be enough.

Small speakers certainly don't have a flat response down to deep bass, but neither are they "brickwall" filters that eliminate all bass content.  If a cab can provide some acoustic reinforcement to lower frequencies (e.g., in the 200-300hz range), then don't be shy about using a larger-value cap to let them through to the  speaker.  Note as well, that a nice stiff 12V supply will provide a little better headroom such that content below 350hz doesn't fart out on you.