Cambridge Audio CXA81 Review (Sample 2)

[Kosimo]

What do you mean by this? It has 2 channels. It is a 2 channel amp. Where does it say the amp has 4 channels?

And adding bold does not make you correct.

It was not to be correct, just to highlight what I think it is important, sorry, it was not to be rude

Ok but where is said that is two channels? I see four channels here. Specifications does not say. But 750W as the max power consumption
would be coherent with 4 independent powered channels of 120W@4Ohms.

 

[Snoopy]

You don't have 4 channels. You have 2 channels and 2 sets of binding posts that receive the same signal. It's no different than plugging two headphones into a headphone amp

 

[Jimbob54]

I see four channels here.

Where? On the back? Or in the spec? It's a 2 channel amp with 4 black and 4 red speaker bindings. If you ran 2 sets of identical 8 ohm speakers independently they aren't getting 80w pc each concurrently.

To passively bi amp from a one box amp it would have to be a multichannel amp. So think a 5.1 Avr and using (for eg) the front and center bindings.

 

[Kosimo]

@Snoopy @Jimbob54

I think we are a bit speculating here. I do not know the topology of outputs and do cannot any measurements.
Looking at the internals:

I see four capacitors that could drive the four output channels but could be also be linked (two for a single channel), so honestly I cannot say.
Are you saying that LA/LB and RA/RB are connected directly and not driven indipendently ?
That could make the difference, power levels aside, between bi-wiring/binding-posts and passive bi-amping.

 

[audio_tony]

@Snoopy @Jimbob54

I think we are a bit speculating here. I do not know the topology of outputs and cannot any measurements.
Looking at the internals:


I see four capacitors that could drive the four output channels but could be also be linked (two for a single channel), so honestly I cannot say.

The four capacitors are the smoothing capacitors for the + and - PSU lines. Each channel has a pair of capacitors, because the amplifier requires a + and - supply voltage for each channel - this is known as a 'split' supply.

Are you saying that LA/LB and RA/RB are connected directly and not driven indipendently ?

As I explained in my previous post - each of the two* channels / amplifiers drives two pairs (4) of speaker posts.

These are switched as A and B.

*NOTE two NOT four channels / amplifiers (this amplifier only has two separate amplifiers - one for the left channel and one for the right channel).

Please understand this.

That could make the difference, power levels aside, between bi-wiring/binding-posts and passive bi-amping.

Power output does not change with bi-wiring. The amplifier is still sending the same amount of power down each set of cables.

 

[Jimbob54]

Are you saying that LA/LB and RA/RB are connected directly and not driven indipendently ?

I'm no EE so wouldnt have a clue looking at the internals. But yes, that is exactly what we are saying. A and B take from the same amp channel. If you select only A, you get 80w pc into 8ohm. Ditto only selecting B. Selecting both halves the output to each speaker/ terminal pair.

 

[RandomEar]

@Kosimo The Cambridge Audio website says "CXA 81 Integrated Stereo Amplifier". Stereo = two channels. It has two sets of binding posts per channel, for things like bi-wiring or setting up a second pair of speakers in another room. Not uncommon, my Rotel RA-12 stereo amp had them too - but it still had only two amplification channels. Rotel was clear in pointing that out in the spec sheet, while Cambridge obfuscates that fact somewhat. At least that's my takeaway from the info on their website.

 

[diddley]

They wrote it somewhere that power would be halved, i've seen it in the manual.

 

[Snoopy]

They wrote it somewhere that power would be halved, i've seen it in the manual.

Yeah I remember that as well from the manual.

 

[diddley]

https://techsupport.cambridgeaudio.com/hc/en-us/articles/360002685557-What-would-the-power-output-of-the-CXA-be-if-I-connect-two-pairs-of-speakers-

 

[Kosimo]

I'm no EE so wouldnt have a clue looking at the internals. But yes, that is exactly what we are saying. A and B take from the same amp channel. If you select only A, you get 80w pc into 8ohm. Ditto only selecting B. Selecting both halves the output to each speaker/ terminal pair.

@Kosimo The Cambridge Audio website says "CXA 81 Integrated Stereo Amplifier". Stereo = two channels. It has two sets of binding posts per channel, for things like bi-wiring or setting up a second pair of speakers in another room. Not uncommon, my Rotel RA-12 stereo amp had them too - but it still had only two amplification channels. Rotel was clear in pointing that out in the spec sheet, while Cambridge obfuscates that fact somewhat. At least that's my takeaway from the info on their website.

I agree that things are a not clearly stated in the manual. But the cited post clears this completely.
However it is coherent with the fact that they call it bi-wiring. Now it is more clear.

The four capacitors are the smoothing capacitors for the + and - PSU lines. Each channel has a pair of capacitors, because the amplifier requires a + and - supply voltage for each channel - this is known as a 'split' supply.

I suspected.

As I explained in my previous post - each of the two* channels / amplifiers drives two pairs (4) of speaker posts.
These are switched as A and B.
*NOTE two NOT four channels / amplifiers (this amplifier only has two separate amplifiers - one for the left channel and one for the right channel).
Please understand this.

Yes now I see

Power output does not change with bi-wiring. The amplifier is still sending the same amount of power down each set of cables.

Yes, what I meant is that if they are linked it is equivalent to binding posts.

Yeah I remember that as well from the manual.

They wrote it somewhere that power would be halved, i've seen it in the manual.

Searched but not found.

https://techsupport.cambridgeaudio.com/hc/en-us/articles/360002685557-What-would-the-power-output-of-the-CXA-be-if-I-connect-two-pairs-of-speakers-

I never came across this post. This takes the last word on this aspect. @Snoopy you were right. I was mistakenly convinced it was a four channels amplifier.

Thank you all for clarifying this , I was the one doing bi-wiring without realizing!

There is just one problem, and now I am really confused: if it is equivalent to binding posts, why the hell I was able to recognize 14 times over 15 a difference using the two sets A and B with respect to just single set with binding posts?

 

[Jimbob54]

I agree that things are a not clearly stated in the manual. But the cited post clears this completely.
However it is coherent with the fact that they call it bi-wiring. Now it is more clear.


I suspected.


Yes now I see


Yes, what I meant is that if they are linked it is equivalent to binding posts.



Searched but not found.


I never came across this post. This takes the last word on this aspect. @Snoopy you were right. I was mistakenly convinced it was a for channels amplifier.

Thank you all for clarifying this , I was the one doing bi-wiring without realizing!

There is just one problem, and now I am really confused: if it is equivalent to binding posts, why the hell I was able to recognize 14 times over 15 a difference using the two sets A and B with respect to just single set with binding posts?

Pure speculation here , but I would suggest that something wasnt wired up correctly in one of the 2 scenarios. Your speakers have 2 sets of terminals and (when run with single wires) use a connector bar connecting top and bottom connectors? Was that in place for single wired operation and removed for bi-wiring? How many drivers do the speakers have?

Not questioning the Mrs' capabilities but if the bars werent replaced for single wire config, you would likely only get mid and tweeter drivers firing, missing the bottom end. Im not sure what happens if you leave connector bars in but run bi-wired.

 

[audio_tony]

There is just one problem, and now I am really confused: if it is equivalent to binding posts, why the hell I was able to recognize 14 times over 15 a difference using the two sets A and B with respect to just single set with binding posts?

How long are your speaker cables, and what gauge are they?

With very long cables, it may be possible to hear some subtle differences (this will depend on many factors though - in particular the load the speakers present, and how the amplifier reacts to that load, and the speaker cable gauge).

 

[Kosimo]

Not questioning the Mrs' capabilities but if the bars werent replaced for single wire config, you would likely only get mid and tweeter drivers firing, missing the bottom end. Im not sure what happens if you leave connector bars in but run bi-wired.

ahah but this would have generated macro-effects and I would have noticed that something was wrong. It must be something more subtle.

How long are your speaker cables, and what gauge are they?

With very long cables, it may be possible to hear some subtle differences (this will depend on many factors though - in particular the load the speakers present, and how the amplifier reacts to that load, and the speaker cable gauge).

I have used 3mt cables with 12 AWG, I have two identically sets of cables in two different setups. So I borrowed one from my setup 2 to make the test.

 

[RandomEar]

There is just one problem, and now I am really confused: if it is equivalent to binding posts, why the hell I was able to recognize 14 times over 15 a difference using the two sets A and B with respect to just single set with binding posts?

I remember from my Rotel that the second pair of binding posts was wired different internally. I think they had an additional pair of chokes (?) in there between the amp stage and the (B)-posts. You can see it in this image ((B)-connectors are on the top left). It was meant to compensate for the longer expected wiring on the (B)-set of speakers, as they were expected to be in another room. I suspect that this might affect the high frequency roll-off and could be handled in a similar way on your Cambridge amp.

Of course, there could also be something odd with how the test was performed, but I guess that's for you to investigate

 

[audio_tony]

I remember from my Rotel...

Which model Rotel do you have?

 

[Kosimo]

I remember from my Rotel that the second pair of binding posts was wired different internally. I think they had an additional pair of chokes (?) in there between the amp stage and the (B)-posts. You can see it in this image ((B)-connectors are on the top left). It was meant to compensate for the longer expected wiring on the (B)-set of speakers, as they were expected to be in another room. I suspect that this might affect the high frequency roll-off and could be handled in a similar way on your Cambridge amp.

It could be. But not knowing internal topology of CXA81 we cannot say if it is equivalent to join binding posts or not. From listening point of view seems to be
some differences but at this point I have no idea of the cause.

Of course, there could also be something odd with how the test was performed, but I guess that's for you to investigate

yeah I need to do some additional investigations

 

[Nikolaos]

@amirm I chose to bypass CXA81 DAC connecting it with XLR cables to RME ADI 2 DAC FS. What do you think would the best set of knob volume according to which set of RME db rate. At 9 o'clock ( actually the zero is at 7 o'clock where knob is at minimum) the RME minimizes at -75db (no sound output).

 

[terrys999]

@amirm I chose to bypass CXA81 DAC connecting it with XLR cables to RME ADI 2 DAC FS. What do you think would the best set of knob volume according to which set of RME db rate. At 9 o'clock ( actually the zero is at 7 o'clock where knob is at minimum) the RME minimizes at -75db (no sound output).

+7 0.0 vol.
This should give best quality.

I use adi with cxa 80. At those settings.

You could use. +13 -4.5. But not +13. At 0.0. That would be to hot for those amps.

 

[Nikolaos]

+7 0.0 vol.
This should give best quality.

I use adi with cxa 80. At those settings.

You could use. +13 -4.5. But not +13. At 0.0. That would be to hot for those amps.

+7 o'clock amp and RME....db ?
+13 o clock amp and RME -4.5db