Single Connect:AMP driving 6 Bose cube speakers, 2 Yamaha outdoor speakers and a subwoofer?

it's too big, clunky and complicated for the space and what I need it to do.
Perhaps not; the big clunky box may well be the one with the power to drive all your speakers - no small thing - that a Connect Amp may struggle with. It seems like you are only using two channels of the multi channel receiver though; what is the power deliver spec per channel?

With the impedances of the combined load another question mark, the safe thing to do for a system that seems to be working well is to disturb it as less as possible. That would mean replacing the Connect once you establish that the earlier unit has failed. But be sure to do that by ruling out network issues as the reason for the old Connect to have hiccups.

The Bose double cubes are effectively a single speaker each, so you have three Bose satellites plus (I hope) the Bose sub. If it's the Bose Acoustimass 10 kit then the satellites are fed via the sub. The sub is usually fed from the centre, left, right, surround left, surround right speaker outputs of a 5.1 AV receiver. There is no dedicated sub connection to the AV receiver. The sub filters its sound from the 5 main speaker feeds.

It's said that each channel of the Bose Acoustimass system provided a 4-8 Ohm load to the amp. Unfortunately Bose aren't forthcoming with exact specs so it's speculative about the minimum load. There's also no information about speaker efficiency. That info is incredibly important, and the fact that Bose won't say what the dB efficiency of their speakers is tells you volumes about the quality (or lack thereof) of the product.

This info is important because if the old Pioneer pumps out a lot of current, and can increase its power output by 50% when the impedance halves from 8 Ohm to 4 Ohm, then it's a bit of a monster and so speaker efficiency isn't so much of an issue. It could be using a big old wire-wound transformer and beefy capacitors so can supply lots of current which the Bose speakers will lap up. Modern amps don't do that. They use switch mode power supplies (sometimes referred to as Digital Amplification to make it sound high tech and new) which aren't so good at driving difficult loads. However, Digital amplification is very efficient and cheaper to build so it keeps the green lobby and the bean counters happy while providing the marketing department with a high Wattage figure to shout about. The problem is Watts mean sod all if there's no current to back it up.

What we need to know then is which Pioneer receiver, and how exactly are the receiver speaker outs connected to the speaker switch, and (most importantly) which connections on the speaker switch are connected to what speakers in your home. The oddment here is the third one of the Bose cubes. Speakers generally go in pairs, so wiring one speaker in the mix is strange.

Sorry, I meant to type “3 double cube pairs” so it’s six effective bose speakers, 12 cubes, plus the Yamahas. It is NOT a Bose subwoofer and it’s connected to a dedicated subwoofer output on the Pioneer AV receiver. I’m not at the house now and I don’t remember the model of the Pioneer AV receiver. It’s old, it came with the house and it came without a manual. Thanks!

Okay, so it sounds like you have 4 pairs of speakers plus an active sub. The Bose satellites on their own have a minimum impedance of about 5 Ohms. That's not a disaster. Unofficial figures suggest an efficiency of around 85dB and a frequency range from 280Hz to 13,500Hz. That is poor for a speaker system costing £800/$1000 complete. I can get better than that from £50 stereo speakers. But you've got what you've got so let's just go with the flow.

The connect amp will drive a 4-8 Ohm load. Depending on your speaker selector and how its current protection works try to configure it so the amp sees a 6-8 Ohm load or higher. This will help prevent the speakers trying to draw too much current.

The old Pioneer receiver could be anything from a base model through to something old but quite high-end picked up for a song via Craigs List / Ebay / Gumtree. The low-end receivers weighed in at around 10kg and were rated at 50-80W/ch but that was measure at 1kHz and with 1% THD (distortion). Real world power translates to around 30-40W/ch and enough current drive to power relatively inefficient 5.1 speaker packages to 95dB. The higher-end AV amps were rated at 100W/ch (stereo mode) but measured at full audio spectrum 20Hz-20kHz and at less than 0.1% THD. They weighed in at 18-20kg. That weight translated to bigger power transformers, more current and an ability to drive just about any speaker to ear splitting volumes. The Sonos Connect Amp @ 55W/ch from a Class D amp circuit sits somewhere between the two and closer to the budget amp than the high-end one.

Yes, the Connect Amp will drive the speakers via the impedance matching switch. How loud it will go though depends on what it's being compared to in the Pioneer amp.

However, Digital amplification is very efficient and cheaper to build so it keeps the green lobby and the bean counters happy while providing the marketing department with a high Wattage figure to shout about. The problem is Watts mean sod all if there's no current to back it up.

Although a small digression, this allows me to raise a question that has never been explained well enough to me in simple terms a non electrical engineer can understand.
But I do know enough to understand that Watts/Volts/Ohms/Amps measure different aspects of the phenomenon and are inter-related.
So my question is about this oft heard current comment. Let's say there are two amps, both which truly measure 50 watts per channel, rms, at the same THD number, into a measured 8 ohm load. Is it then possible for them to be different where outputs in current is concerned? If one of these two puts out less current than the other, how does it then manage to deliver the same Watts into the same Ohm load, with the same THD? What part of the inter relationships am I missing? Will the current difference not translate into a Watts difference as well?

For what it is worth, Google tells me that : VOLTS x AMPERES = WATTS, and amperes, or amps, are a measure of current.
So, if the same watt spec is to mean different amps (current), volts have to change in an inverse way to the change in amperes for a given watts number, going down if the amperes are to be higher than in cases where the amperes or current is lower - for the same watts number. Where/when does the change in volts happen in audio kit for the claimed variation of current for a given watt specification?
PS: some more googling gave me most of the answer before my eyes glazed over, but I picked up enough to confirm that the Connect Amp, given that it can also double its 55 wpc to 110 wpc with the same THD+N of 0.02%, into a 4 ohm load, isn't by any means a wimp as far as its ability to deliver clean current is concerned, when compared to any typical modern day stereo amp of 50 wpc; indeed it is probably better than many such.
But both will struggle to meet loads that drop significantly below 4 ohms.
Circling back to the OP situation then, I still have my doubts about the suitability of any such amp, given the probable nature of the total load. The big clunky Pioneer is probably so for good reasons as seen by how well it is driving the load.

However, Digital amplification is very efficient and cheaper to build so it keeps the green lobby and the bean counters happy while providing the marketing department with a high Wattage figure to shout about. The problem is Watts mean sod all if there's no current to back it up.

Although a small digression, this allows me to raise a question that has never been explained well enough to me in simple terms a non electrical engineer can understand.
But I do know enough to understand that Watts/Volts/Ohms/Amps measure different aspects of the phenomenon and are inter-related.
So my question is about this oft heard current comment. Let's say there are two amps, both which truly measure 50 watts per channel, rms, at the same THD number, into a measured 8 ohm load. Is it then possible for them to be different where outputs in current is concerned? If one of these two puts out less current than the other, how does it then manage to deliver the same Watts into the same Ohm load, with the same THD? What part of the inter relationships am I missing? Will the current difference not translate into a Watts difference as well?

I'm not an electrical engineer either, so I'm not sure I'll be able to explain it in layman's terms that will also satisfy those who are engineers. The following then is very simplistic. I acknowledge that things are far more complicated in real amplifier circuits, but that's not what the question from Kumar is asking.

When I was at school our physics teacher taught us about transformers, voltage and current by melting a 6" nail. It was a graphic demonstration of what happens to current when using a transformer to convert high voltage to low voltage and what happens to the current generated as a result. There was something like 1000 Amps on tap when the nail shorted out the connections.

At their heart amplifiers have the same kind of voltage reduction going on. We're not melting nails but we are reducing mains voltage down to something that won't fry the speakers.Amplifiers then are voltage sources. Wattage comes from what happens when the voltage is put across a resistance.

What separates a good amp from a mediocre one is how far the designers are willing to go to over-spec the transformer rating to deliver a lot of current in to a low Ohms load. A transformer rated just high enough to cope with the stated spec plus a bit of headroom is going to run up to its maximum current rating far sooner than one with a huge amount of headroom. So, while both might be advertised as 50W in to 8 Ohm, one of them will cope far better running at high volume or dealing with a difficult load.

So, what is it about the transformer power specs than makes one better than another?

It's this thing called headroom - The additional capability of the transformer to cope with peaks in demand. If we want an amp that can deliver 50W in to a static 8 Ohm load then there's a way of working out what the voltage for the secondary coil of the transformer has to be to deliver that. In this case it's 20 Volts. From that we can work out the current that the transformer will supply (W/V = I) so we get 2.5 Amps. That mean our transformer is rated as 50VA (Volt-Ampere) if we want a maximum of 50W in to a straight 8 Ohm resistor.

That's all fine and dandy for a fixed resistance but speakers aren't that cooperative. Their resistance varies with frequency. So we need to consider what happens when the resistance changes. More specifically, we need to look at what happens then the resistances falls.

If the resistance drops to 4 Ohms then a perfect power supply will deliver 5 Amps and the power being used will raise to 100W. However, our transformer is designed with a fixed 20 Volt output from the secondary coil for an 8 Ohm load, so unless we can change the the VA rating of the transformer then it's never going to supply anything more than 2.5 Amps. To make a transformer that can supply 5 Amps in to a 4 Ohm load then we'll need to double the VA rating.


Hang on a second.... Doesn't that make it a 100 Watt amp then?

It would do if we increased the secondary turnings to give 28 Volts for an 8 Ohm load. Then we'd have W/V = I so 100/28 = 3.5 Amps of current. That's still not enough to satisfy the 5 Amp demand for the 4 Ohm speaker.

What is done instead then is to keep the secondary windings for 20 Volts. Now we have 100/20 = 5A. Bigger transformer, running well under its maximum rating - in car engine terms the top speed is the same but the engine is far more torquey - it provides the headroom to cope with lower resistance loads despite the paper wattage being the same.


So why don't all amps with low rated transformers blow up or melt down or clip horribly then?

Firstly, the power figures assume the amp is driven full whack, but that's rarely the case. If the speakers are sensitive (say 90dB) then all it takes is 2.83V in to an 8 Ohm speaker to produce 90dB measured at 1m from the speaker, and 90dB is pretty loud. So we aren't using the maximum that the transformer can generate all the time. If our amp can produce 20 Volts but all we are using is 3 V for moderate listening levels then there's a reserve of power still on tap. At 8 Ohms our 2.83 V means that the speaker is drawing about 1/3 of an amp, so there's some headroom ready to handle the dips in speaker impedance (complex resistance). If we keep the same 2.83 Volts then at 4 Ohms the speaker is drawing about 0.7A and at 2 Ohm it is 1.4A. This is all well within the capabilities of the 2.5A transformer

Secondly, if the rated power (50W) is RMS and measured at the full 20Hz-20kHz audio range and at very low THD then the amp's transformer will be larger than the VA rating to deliver exactly that power. That means some power reserve has already been factored in to the design.

In pure mains AC circuits the RMS value is 0.707 of the peak power value. Amp circuits don't quite follow the rules for pure AC though, but for the sake of convenience we'll use 0.7 as a factor. So our 50W in to 8 Ohm low THD amplifier is running at about 70% of the transformers VA rating - this is ignoring all the thermal losses, resistances and other complications - so our budget 50W amp have say a 75VA transformer. It's still running on a 20 Volt secondary output, so W/V = I gives us 75/20 = 3.75 Amps. Oops.... This transformer is still going to run out of gas with a 4 Ohm load. It can't supply 5 Amps. But we only need 5A if the amplifier is running at its rated 50W output.


Okay, so as long as we avoid running at high volumes then any amp will do just fine then?

Not quite. There are some things working against the amp. No transformer or amplifier is 100% efficient. Cabling to speakers has an effect and if it's underspec'd (cross-sectional area for the length) then it will limit current flow. Speaker sensitivity varies quite a lot; cheaper speakers achieve 85-86dB on average. They'll demand more current than better speakers at 88-90dB. Sound dissipates over distance: With a 90dB sensitivity 8 Ohm speaker we need 1W to make 90dB of sound pressure. But sit 3 metres away and we need 9W to make the same sound pressure. Swap the speaker for one with 86dB sensitivity @ 8 Ohms and now we need 22W to make 90dB at 3 metres.

The biggest surprise though is how much extra power is needed to make a perceptible difference on the sound pressure level. It's said that a minimum of 3-5dB is required for listeners to hear any sort of increase. For our 86dB 8 Ohm speaker that means changing the sound pressure from 90dB to 93dB requires a power increase from 22W up to 45W. Fortunately for us we listen in stereo rather than single speaker mono. So running two speakers together means we need 11wpc for 90dB, 23wpc for 93dB, and about 36wpc for 95dB. To get the sound to be twice as loud requires a 10dB increase which means over 110Watts. Clearly that's impossible with a 50W @ 8 Ohm amp.




In summary then for transformer based amps

We don't use all of the amplifier's power capability all the time so the hard limits of the transformer aren't encountered every time the amp is switched on. However, there's less power in reserve to deal with difficult loads where an amp has an under-spec'd transformer. That translates to a limit in the ability of the amp to cope with peaks in the music programme and real world speaker loads. A bigger transformer works better.



What about 'digital' amps?

Class D amps don't have the same sort of current limits as transformer-based amps. But they do have their own limitations: Distortion due to high levels of feedback (a consequence of how they work) and lots of Radio Frequency Interference if they aren't designed very carefully.

Class D and the other variants of the switching designs (T-amps etc) are cheap to build poorly, but expensive to build well.

but that's not what the question from Kumar is asking.

If the resistance drops to 4 Ohms then a perfect power supply will deliver 5 Amps and the power being used will raise to 100W.
First, thank you very much for the considerable effort behind the detailed answer; much appreciated, though even that was enough to have some glazing of the eyes happening at my end.

To the second quoted sentence : this also is exactly what the Connect Amp does, doubling its 55 wpc @ 8 ohms to 110 wpc @ 4, with no change in the 0.02% THD+N measurement; if it does that, it has to have a power supply that also supplies double the current when the impedance drops from 8 to 4 ohms; one goes with the other in this case, current and watts. Which is probably why I find that it is a very adequate replacement to drive passive speaker pairs ranging from Quad 11Ls to Dali Zensors, in place of any third party stereo amp of a similar spec.

But this is also why I prefer play units now. Like for any other active speaker, all the hard work of matching the amp to the speaker has been done by Sonos and if this is done competently, with active crossovers, and amps dedicated to each driver in the box, sound quality from active speakers is as good as from any very well matched separates system and better than any ill matched one. And it leaves these headaches to those best equipped to solve them so I can leave these aside and focus on what matters, the music.

Re: eyes glazing over. I suppose it depends how much validation one needs before accepting an explanation from a stranger on the web.

Also, without some background as to why the output voltage of the transformer is fixed and what that means for the current capability then you'd drop back to the basic W = I x V equation in which both current and voltage are free to alter i.e. 100W = 1A x 100V = 2A x 50V.... and so on until all the permutations of I x V that produce 100W are satisfied. That would undermine an explanation based on a fixed voltage output, and so damage the credibility of the argument, but it would be based on an incorrect application of electrical theory.

Had it not been for supplying the reasoning behind the transformer choice would you have accepted the equivalent of the "because they put a bigger engine in the car" argument without it raising any questions, especially since you already said that it hadn't been explained satisfactorily in the past?


As for the Connect Amp, has anyone measured the output to verify the claims?


The amps aren't really digital, they're switching amps, but "digital" has a nice buzz to the word that folk automatically (but incorrectly) attribute as better. One of the main attributes of Class D amps is that they tap in to the almost unlimited voltage and current available from the mains outlet and have the potential in their design to pass most of that to the speakers. So doubling the power at half the resistance isn't anything surprising in Class D.

It's entirely subjective, but I still prefer the sound of the original ZP100 with its transformer-based amp to the ZP120/Connect Amp. That might be down to the way the analogue amp has a grip on the bass, or it could be that I prefer the warmth (distortion?) of a linear amp compared to the more sterile sound of the Class D.

In a similar fashion, I'd always take a Connect hooked up to a good linear amp and a pair of bookshelf speakers for serious listening over a pair of Plays. There's something about the way a good stereo system digs in to the music to reveal timing and subtlety that's missing from a lot of systems. Then again, I've heard what should be very good budget stereo systems in very bad rooms and all the resolving power evaporated.