Speaker Level=line source for difficult load?

[ro9397]

This is a low-fidelity application.

I desire to power preferably up to about 10-12 (8 would work) small identical 7Wpc power amps with one source. I'll hopefully get the amp input impedance tomorrow...likely in the normal range of about 20k Ohms.

Am I correct that the headphone output of an iPod music source (and any regular line level) has too little current to drive several power amps to maximum level?

But what about the following? One iPod drives one of the above 7wpc amps, whose 7Wpc speaker output from one channel (yes, it's mono but no matter) leads to drive the other 10-12 slave amp inputs? Each slave amp has a resistor network at its input, estimated 2k series followed by 5k parallel, minimizing the speaker level down to a line level to properly drive the slave amps.

Yes, no, maybe? What are recommended R values? What is the theoretical maximum number of slave amps?

One larger power amp with line-matching transformers is undesired because of extra cost and complexity.

[AndrewT]

What is the recommended load impedance for your source?
What is the output voltage of your source?

With these two pieces of information you can design the multiple little power amps to have a compatible input impedance and gain.
Simple design exercise if you find out the two answers we would all need to get started.

[macboy]

Headphone output of the iPod is designed to drive headphones, so maybe as low as 32 ohms. You should have no problem driving several 20 kohm line inputs.

[ro9397]

Originally Posted by AndrewT

What is the recommended load impedance for your source?
What is the output voltage of your source?

With these two pieces of information you can design the multiple little power amps to have a compatible input impedance and gain.
Simple design exercise if you find out the two answers we would all need to get started.

I don't know where to get the iPod's headphone output voltage spec. My humblest apologies for being such a dunce w/ Ohm's Law. Will not the iPod's voltage be split among each of the many slave amps, thereby making the iPod's voltage too low for maximum amp power?

I perceive too little iPod source voltage regardless of its current reserves...same as a 15W pure class A amp having too little voltage/power even though it has plenty of current reserves powering a loudspeaker...am I on the right track here?

[ro9397]

Originally Posted by macboy

Headphone output of the iPod is designed to drive headphones, so maybe as low as 32 ohms. You should have no problem driving several 20 kohm line inputs.

Yes, I thought the same thing, plenty of current reserves, and this is apparently correct. But a PhD. in Materials Science mentioned (IIRC) that the source voltage potential of the headphone output will diminish/be split among each of the slave amp inputs. The result is too little source voltage to power the amps to maximum output...if I'm correct.

I think a good, possibly perfect analogy to the scenario above is a low-power amp with pure class A topology running out of voltage powering a loudspeaker load, even while the amp has plenty of current reserves.

That's why I thought of using a speaker output source with a resistor pad at each amp input, but I'm unsure whether or not this will drive each amp to maximum output.

Again, the goal is something simpler and less costly than one higher powered amp and line-matching transformers.

[AndrewT]

Your PhD friend has left you confused.
The voltage is not split between the multiple loads.

The voltage is split between the internal output impedance of the source component and the effective load you attach to your source.

Let's suppose your iPod has a voltage output of 1.2Vac and assume it has an output impedance of 200r.

Let's hang a 200r load on that output.
The internal impedance sees a voltage drop of half of the 1.2Vac available and the load sees the remaining 0.6Vac. Vout = Rload / [Rload+Rs] * Vac

Now change the load to 1k0.
The internal impedance causes a voltage loss of 200 / [200+1k0] * Vac = 0.2Vac
The remaining 1.0Vac is applied to the 1k0 load.
That 1k0 load could be a single high impedance headphone of 1kohms or ten amplifiers each with an input impedance of 10k or twenty amplifiers each with an input impedance of 20kohms.

To start the process, you need to know the source impedance and the source voltage. It would be helpful to know what the maximum output current and/or the minimum recommended load impedance are.

BTW,
a 1Vac applied to a 1kohm load passes a 1mAac current to the load.
This is equivalent to 1.4mApk passing to the load.
If your iPod cannot manage ten times that, then it is only fit for the refuse bucket

Another BTW,
we can talk you through an easy way to measure the source voltage and source impedance.

[ro9397]

Andrew
I'll contact Apple for your requested specs.

Thanks! I may send you a PM.

[BobEllis]

What Andrew was telling you is that if the i-pod can drive headphones it can probably drive your string of amplifiers. Somehow I doubt that Apple will release specifications for such an off label use.

Just try it. If it meets your needs, great.

If not, it is a fairly trivial matter to build a buffer from a NE5532 that could both mix both channels of your output and provide a little gain if needed. NE5532 is designed to drive a 600R load so if your amp input impedance is 20K, each side of the opamp can drive 33 of your amps.

Given the low output power of your amps, you could probably steal power from one of them to power the buffer.

Edit: How long are the runs between the I-pod and the amps? If long, you may be better off with your original idea, increasing the "line" level to minimize the effect of induced noise. Start with a 20K resistor in series with the amp input and add a shunt to get the required attenuation when paralleled with the input impedance of your amp. If the amp goes to full power on .5V RMS and your "driver" outputs 7.5 volts RMS, you need to reduce the voltage by a factor of 15. A 1.5K resistor across the amp input would be pretty darned close. Do this at each remote amp.

If very long, you might want to use a NE5532 buffer with one half as an inverting buffer to drive balanced lines to your amps. If your amps dont' accept a balanced input you'd then need to either modify the amps to accept balanced input or add an opamp balanced to unbalanced converter. None of this is very hard to do if you have basic soldering skills.

[ro9397]

Bless you all for the help!

Bob
I'll have the amps here within a week and try it. Duh! That's why they pay you guys the big bucks, to dream up such remarkable ideas!

If not, is such a device like your suggested NE5532 available commercially? I need to make this work for multiple users (luthiers, instrument repair shops) and building each of them such a device, regardless how easy, is not an option.

BTW, the amp inputs are 3.5mm stereo phone jacks. Please let me know if you've ever seen a 1-2 splitter to daisy-chain one amp's input jack to the next. Each amp has its own 3.5mm audio interconnect cable, so pigtails not required on the adapter, just a simple adapter to daisy chain...I suppose an adapter made to power two headphones from one 3.5mm jack would be perfect.

[BobEllis]

The NE5532 is an opamp available for less than a dollar almost anywhere you can buy electronic parts. You would need another dollar or two worth of parts to make the buffer or receiver. If you won't build something at each station, don't worry about it.

You might be able to build the pad into your amplifier so that it would appear to users that it was a normal input.

By multiple uses, do you mean that each room will have the option of disconnecting from central and having their own background music? Is it a mobile system?

How about this? Gigaware™ Headphone Splitter for iPod® - RadioShack.com