Are integrated amps technically better than separates?

Would you fill in some of the details of what you know about power cords? You've told us that a power cord that's not up to the job is going to hurt the sound, but what are the basic requirements for an adequate power cord? One that will not degrade the sound in a system the average 'Goner might own.  

As a general rule of thumb I go with heat. If the cord makes any heat at all, in the middle or on either end after several hours of operation, then its not up to the task of working with the component its powering. At that point a heavier power cord can be very nice. The connections on either end can be important- if they heat up it won't matter much about how good the cord itself is as the voltage drops on the connectors (hence the heat) will overpower that.


I have a cheap solid state amp in my bedroom system that makes about 5 watts per channel on a good day. It does not need much of a power cord. But with a 60watt tube amp you can notice differences (and measure differences too) quite easily. So there is a scale issue here due to Ohm's Law; consequently what will work varies from product to product. 

Agree with “erniesch", his statement is 100% correct: "Never mind that just inside the wall is the cheapest copper lines the contractor could buy.” Last two feet power cable 10 or 18 AWG difference is drop in the ocean, if power lines inside house are bad.  

OK- if that is true you have to explain how I could measure a voltage drop across a power cord and hear the difference in my system- when my house at the time still had knob and tube wiring.


All things electrical have to obey Ohm's Law and its a simple fact that solid core copper does better then multistranded wire, which is why its used in building wiring. Keep also in mind that in most houses there is a wiring code and while there is cheap wire, it has to meet that code.


Now the measurements I made point to the idea that the more current draw, the greater the voltage drop across the power cord; this suggests that its more of a problem with equipment that has a higher current draw and far less of an issue with equipment that does not. Ohm's Law again.

But one other area that should be mentioned is high frequency response of the power cord- another thing that solid core does really well. This issue here is that in most equipment there is a power transformer, rectifiers and filter caps. The rectifiers can only turn on (commutate) when the cap voltage is less than that from the transformer; most of the time this means that conduction only occurs at the very top of the AC waveform. So the current draw has to happen over a fairly short period of time, even less than a millisecond. If the bandwidth does not exist in the cable, the power supply will not charge properly- it will round the charging pulse. Again, Ohm's Law.


None of this says the cable has to be expensive. It does say that the cable has to have the bandwidth and the current capacity, and good enough connections at either end such that those ends don't warm up over time- if they do, you know the connections are robbing power from the system.


This is all measurable and audible. In the case of an amplifier simply measure the full power output, the output impedance and the distortion. You'll find that it varies with the input voltage. This should be totally non-controversial; I think the only reason it is is because there tends to be a knee-jerk reaction to the idea that a power cord can make a difference, associated with a failure to cause one's hand to move and actually make the measurement!

But pream has to have output stage that drives interconnect and relatively low impedance (compared to tube input inside amp) of power amplifier input. The interstage driver in integrated amplifier is much easy to design and doesn’t have to have a feedback and output transformer or big value capacitor or sophisticated DC cancelling circuit.

The first statement is true. The second needs more explanation; as far as I can see its false. Regarding the first statement, over the years I've found that a lot of the improvements I've been able to make in our gear have related to power supplies. Arguably that's one of the trickier things that goes on in an integrated amp; IME it outweighs many of the other circuit topography issues! I totally get the connectivity issue- that is an advantage. But in most cases, its not *enough* of an advantage.

It’s so interesting that folks think they need a high quality power chord and expensive interconnects. Never mind that just inside the wall is the cheapest copper lines the contractor could buy.

If you are suggesting that a power cord does not obey Ohm's Law... Ohm's Law isn't like a speed limit; it can't be broken. By comparison, the solid core wiring in the wall is fairly high performance in terms of voltage drops along its length compared to a power cord, but it is illegal to use Romex or the like for power cords. The effect of the power cord is measurable and with relatively unsophisticated instruments. So the onus is on you to explain why they can't have an effect.

I've seen power cords rob a 140 watt amp of 40 watts of its output power. If you are suggesting that is not audible as well... sheesh! Its pretty laughable.

Each amplification stage, capacitor and cable add distortions and noise.

All those gain stages are in an integrated amp too. Usually though you have a greater chance of lower noise with separates since crosstalk and intermodulations from the other channel won't exist.


To see how profound this is, this morning I was working on an integrated amp, and noticed a very distorted signal at the input of the power amp- on the first stage and throughout that channel. Chased it back to the input, which is all passive components on that amp, and realized that since I was only driving one channel, the unused channel had that distorted input (and output) going on as a result. There is nothing malfunctioning about this amp; but apparently this sort of signal (the result of crosstalk) is adding distortion to the amplifier, which it would not have were the channels merely separated. The amplifier was an older solid state amp made by Trio (Kenwood/Pioneer).

Taking away the flexibility factor of separates, is my line of thinking correct?

No. If you want to build an integrated amplifier that has the same performance as separates, you would have to have individual power transformers for each power amp in the box as well as whatever preamp is in the box. In a nutshell you're not likely to see that. Keeping all the ground currents straight is another engineering nightmare that has to be solved too; with individual boxes that's a lot easier!

There is also the issue of speaker cables. If you are running low impedance speakers, the cables become pretty critical and length matters! So you wind up being limited as to how far apart you can place the speakers. I like an expansive soundstage so the speakers need to be far apart; pretty tricky with everything in one box unless you have 16 ohm speakers (where the cables make less difference).


They are a compromise, plain and simple. The only reason this is a topic is because there are good integrated amps and fair separates so there is some crossover. But the best separates are not bettered by an integrated amp.