The new "Listening Room"

[andyr]

Yesterday, I finished a project which has been going on for a couple of months - that of changing from this:

 

 

 

 

I built the 'black box' 15 years ago, to house:

• a 3-way active XO
• 3x AKSA monoblocs
• 4x power supplies.

To the new setup - which is this:

 

 

 

 

The smaller case houses an AKSA 100w 'Soraya' monobloc - with an upgraded CRCRC power supply.  This drives the bass panel (which is currently just resting against - not bolted to - its braced stand ... and not connected).

 

The larger case contains a NAKSA 80 (stereo) PCB - and an EI power transformer.  This drives the (3 ohm) mid panel and (2 ohm) ribbon.  A 3RU Modushop case was needed, due to the height of this power transformer - which gives a generous heatsink for the output devices. 

 

Here are pics of the fronts and backs of the amplifier cases:

 

1.  Soraya monoblocs

 

 

 

 

2.  NAKSA 80 stereo amp

 

 

 

 

 

I'm keen to hear what the new PSes do to the sound of the bass panel amps - but, as posted above, I have to fix the problem with the bass panels, before I can have a listen. 

 

Andy

[andyr]

Well, @djb very kindly came round this arvo, to help me fix the 'connection' problem with my newly re-wired bass panels.

 

Luckily, "Plan A" - the quick solution - worked ... so after an hour and a half, we were listening to some vinyl.  First time for 3 months! 

 

What we had to do is simple:

• move the wires which previously were connected to a solder pad (where the previous (round) wires were soldered to)
• to connect to the new mini-spades, using right-angle spade connectors.

 

See below pic for the end result.  (Compare this to the pic in my Nov 14th post, above.)

 

 

 

 

The exercise hinged on whether I had enough wire spare, behind the panel, to allow for this.  Luckily I did! 

 

Andy

 

Edited November 22, 2019 by andyr

[ThirdDrawerDown]

Hi Andy, you'll be happy to have your LPs back into use! 

 

Because you are a tweaker, I must ask: any noticeable change in sound?

[andyr]

9 hours ago, ThirdDrawerDown said:

Hi Andy, you'll be happy to have your LPs back into use! 

 

Yes - absoloodle, Murray! 

 

9 hours ago, ThirdDrawerDown said:

 

Because you are a tweaker, I must ask: any noticeable change in sound?

 

 

Well it may be just confirmation bias ... but I felt bass had a bit more 'snap'.  For instance, my MoFi Joe Cocker album. 

 

But @djb thought the same (maybe he was just buttering me up so I'd give him another glass of red?  ).

 

Andy

 

[davewantsmoore]

On 30/08/2019 at 2:04 PM, Ittaku said:

The research showed that mono subs are fine, not that they're better. Where is the research that shows that shows that mono subs are better than stereo subs (and add to that - when positioned at the front wall?)

"Better" is terrible metric.

 

Really, what we need to look at is what do mono (or stereo) subwoofers do.    They will produce different results under different circumstances (one big one is whether there is stereo bass recorded)  .... which someone may like more or less, under different circumstances  ..... and you can either go on that .... or you can look at how they reproduce the recorded signal, and which one is a more accurate reproduction.

 

What you will find is that mono bass is almost always better at reproducing the recorded signal........ even if that signal is stereo.   ie. you should mono mix the the bass, if it wasn't already .... and send it to mono subs...... rather than letting it be "mixed mono" in room).    This applies more and more as the frequency gets lower and/or the room gets smaller.

 

 

 

 

[Ittaku]

17 minutes ago, davewantsmoore said:

What you will find is that mono bass is almost always better at reproducing the recorded signal........ even if that signal is stereo.   ie. you should mono mix the the bass, if it wasn't already .... and send it to mono subs...... rather than letting it be "mixed mono" in room).    This applies more and more as the frequency gets lower and/or the room gets smaller.

Again, you make it sound like fact. All I see is opinion without citations.

[djb]

All I really seemed to be there for was to entertain the dog by throwing balls?

 

The sound was sharper/crisper than when last I heard it.as the foil loosens a little the bass will sound even better < no buyer bias here.

take care of yourself next week Andy.

[davewantsmoore]

1 hour ago, Ittaku said:

Again, you make it sound like fact.

Physics is fact.... and it is uncomplicated.

 

If you "run the numbers" you will see, how two sound sources in a room behave a LF.   I don't know of any "paper" which really "spoon-feeds" such things.

 

 

 

In practise, it's possible that in some specific situations.... the results from different approaches could be quite different from each other.   This implies a few things.

 

There will be strong preference to one or the other.

One will be a lot closer to the "original signal" (ie. a more accurate reproduction).

 

 

There is a reason why productions standards are to mono mix bass.    Allowing it to "mix in the air" at playback means what the listener hears will totally depend on their playback environment.    Unless the listener is using headphones, then it wont sum the way you predict.....  but, headphones bring their own paradox.

 

 

 

 

[andyr]

Well, 5 months have rushed by ... covering:

• my being admitted to the Epworth for a triple bypass at the end of November.
• recuperation for a month - unable to lift anything more than a couple of kilos (even a teapot was out!  ) and needing to go for a walk, for 40 minutes each day.
• then a bit more activity ... being allowed to start swimming again, 2 months after the op (you need to allow the sternum to knit together before stressing/torquing it).  As well as the daily walks!
• and now in lockdown for dog knows how long ... courtesy of emperor-for-life Xi Jinping! 

Before lockdown, Mark (' @frednork ') came over with his PC and mic, and kindly did an REW sweep in my room (after noticing that I had a high noise floor level, in an earlier post I had made).  He didn't find any issues with noise floor but did discover a peak around 50Hz which ideally shouldn't be there.  Which isn't noticeable when listening, btw!

 

So it would seem I have to:

• either figure out the source of this mains hum and remove it
• or use a PEQ dip in my miniDSP at 50Hz ... to cancel it out.

Anyway, as a result of Mark's session, I played with my EQ settings and driver output levels, to try and get a better-balanced FR.  I sent the REW file over to Mark and he pointed out I now had a massive distortion spike at 400Hz - so within the mid-range that I had boosted - and also in the frequencies covered by the subs.  He suggested the input signal to my miniDSP could be overloading the unit.

 

I found I was able to get rid of these high distortion frequency ranges by cutting the signal going into the miniDSP unit, by 6dB. So it would appear that overload was the cause of the distortion! 

 

Hopefully, this will remove the need for any more re-configuring of REW/miniDSP in the foreseeable future. 

 

So my next post should be about listening to the Class A stereo amps I have nearly finished constructing, in place of the NAKSA 80s which are currently powering my mids & ribbons.

 

Andy

 

[Steever]

Gee, not uneventful then.

I wish you a speedy recovery.

[andyr]

52 minutes ago, Winno said:

Gee, not uneventful then.

I wish you a speedy recovery.

 

Thank you, Winno.    I think I basically have recovered ... just wish I could keep on going to the pool 3 times a week. 

 

Andy

 

[andyr]

On 18/04/2020 at 2:37 PM, andyr said:

So my next post should be about listening to the Class A stereo amps I have nearly finished constructing, in place of the NAKSA 80s which are currently powering my mids & ribbons.

 

Andy

 

 

Well, I finished the first AKSA Nirvana Class A amp (to the stage that I could fire it up) a week ago and have been able to:

a. have a preliminary listen, and

b. do some measurements - with sig-gen, CRO and dummy loads.

 

I say "preliminary listen " because, with just 2 channels of Class A available, I could use them either on my mids (3.2 ohms) or ribbons (2 ohms) ... still using 2 channels of my existing NAKSA 80s to power the other drivers.

 

And the amp is not 100% complete - because I'm still waiting for my damn Binding Posts to arrive from the US (they are 3 weeks late  ).

 

@ghost4man, you'll be happy to know that this amp has no problems driving the 2 ohm ribbons.    And it sounded marvellous with the Alpha Nirvana on the ribbons, and my existing amps on mids and bass panels!    (I turned it up louder than I normally listen ... just to see if the amp got into trouble.  It didn't.)

 

The tests I did (with a dummy load) showed that the amp is outputting 34w into 2 ohms at clipping - and the current going through the resistor was 5.05a!  That kind of current would blow the ribbons apart - so I'm sure I'm only outputting a couple of watts, at the levels I listen to.

 

(The interesting thing about a Class A amp is that it outputs a constant current - so what doesn't go through the speakers ... goes into the heatsink.  So, yes, they do produce a bit of heat - I may have to keep my N80s for summer duty.)

 

So now I've got to finish the 2nd amp!  All the boards are soldered - it's simply a matter of drilling all the holes required to fit the pieces inside the 2nd case.  It'll be a boring task (as I've already had the fun of solving the "putting together " issues) but it should be quick - as this time I know what to do! 

 

So by mid-June, hopefully, I can have a real listening session - with ANs on mids, as well as ribbons.

 

Andy

 

[ghost4man]

9 hours ago, andyr said:

 

Well, I finished the first AKSA Nirvana Class A amp (to the stage that I could fire it up) a week ago and have been able to:

a. have a preliminary listen, and

b. do some measurements - with sig-gen, CRO and dummy loads.

 

I say "preliminary listen " because, with just 2 channels of Class A available, I could use them either on my mids (3.2 ohms) or ribbons (2 ohms) ... still using 2 channels of my existing NAKSA 80s to power the other drivers.

 

And the amp is not 100% complete - because I'm still waiting for my damn Binding Posts to arrive from the US (they are 3 weeks late  ).

 

@ghost4man, you'll be happy to know that this amp has no problems driving the 2 ohm ribbons.    And it sounded marvellous with the Alpha Nirvana on the ribbons, and my existing amps on mids and bass panels!    (I turned it up louder than I normally listen ... just to see if the amp got into trouble.  It didn't.)

 

The tests I did (with a dummy load) showed that the amp is outputting 34w into 2 ohms at clipping - and the current going through the resistor was 5.05a!  That kind of current would blow the ribbons apart - so I'm sure I'm only outputting a couple of watts, at the levels I listen to.

 

(The interesting thing about a Class A amp is that it outputs a constant current - so what doesn't go through the speakers ... goes into the heatsink.  So, yes, they do produce a bit of heat - I may have to keep my N80s for summer duty.)

 

So now I've got to finish the 2nd amp!  All the boards are soldered - it's simply a matter of drilling all the holes required to fit the pieces inside the 2nd case.  It'll be a boring task (as I've already had the fun of solving the "putting together " issues) but it should be quick - as this time I know what to do! 

 

So by mid-June, hopefully, I can have a real listening session - with ANs on mids, as well as ribbons.

 

Andy

 

Andy,

 

Sounds fantastic. Can you provide photos of the internals to the amp. That is a decent amount of current isnt it. The amp must have a stiff power supply for those results.

[andyr]

6 minutes ago, ghost4man said:

Andy,

 

Sounds fantastic. Can you provide photos of the internals to the amp. That is a decent amount of current isn't it. The amp must have a stiff power supply for those results.

 

In due course, Ozzie - when the BPs are fitted!    (Otherwise, some people here might comment that I am a cr*p amp builder!  )

 

The (dual-mono) PS is very special - specially designed for Class A amps.  Absolutely no noise (with my ear at the drivers).

 

You can find details on these DIYAudio threads:

• PS:  https://www.diyaudio.com/forums/group-buys/336479-slb-smooth-butter-active-rect-crc-cap-mx-class-power-supply-gb.html
• amp:  https://www.diyaudio.com/forums/solid-state/344540-alpha-nirvana-39w-8ohm-class-amp.html

 

The initial design was for a Class A amp suitable for 8 ohm spkrs.  But Hugh then tweaked the circuit for 4ohm use (more bias current) - and I was the first one to build this version.  He said it would be able to drive 2ohm ribbons - and I'm very happy to report that it certainly does! 

 

 

Andy

 

[ghost4man]

26 minutes ago, andyr said:

 

In due course, Ozzie - when the BPs are fitted!    (Otherwise, some people here might comment that I am a cr*p amp builder!  )

 

The (dual-mono) PS is very special - specially designed for Class A amps.  Absolutely no noise (with my ear at the drivers).

 

You can find details on these DIYAudio threads:

• PS:  https://www.diyaudio.com/forums/group-buys/336479-slb-smooth-butter-active-rect-crc-cap-mx-class-power-supply-gb.html
• amp:  https://www.diyaudio.com/forums/solid-state/344540-alpha-nirvana-39w-8ohm-class-amp.html

 

The initial design was for a Class A amp suitable for 8 ohm spkrs.  But Hugh then tweaked the circuit for 4ohm use (more bias current) - and I was the first one to build this version.  He said it would be able to drive 2ohm ribbons - and I'm very happy to report that it certainly does! 

 

 

Andy

 

Hi mate just had a quick read. I can only surmise your design has been rejigged because looking at the thread - and I hope I have not misinterpreted - seems to say:

"19.1W into 4R, clipping at negative 13Vpk. This is not designed for 4R; it's specifically for 8R with dips to 6R. For big power, say >50W into 4R, I would change outputs and rejig some of the output stage values to maintain stability, THD and power into lower loads."
 

Which seems to suggest to me that Hugh has been very inventive in his design. Clearly it works!!
 

Edited May 26, 2020 by ghost4man

[andyr]

8 hours ago, ghost4man said:

Hi mate just had a quick read. I can only surmise your design has been rejigged because looking at the thread - and I hope I have not misinterpreted - seems to say:

"19.1W into 4R, clipping at negative 13Vpk. This is not designed for 4R; it's specifically for 8R with dips to 6R. For big power, say >50W into 4R, I would change outputs and rejig some of the output stage values to maintain stability, THD and power into lower loads."
 

Which seems to suggest to me that Hugh has been very inventive in his design. Clearly it works!!
 

 

Yes, that comment was by someone who built the original '8R' version - designed for an 8ohm load (with no claim to drive even 4 ohms, well).  He's also someone who is bold (stupid?) enough to take someone's circuit and "improve" it, by changing component values. 

 

Hugh then created (at my urging  ) a 4R version - which he said would be able to handle a 2 ohm load.  The same circuit & PCB is used - but some resistor values are different (to the 8R version).  Plus, the DC rails are lower: +/-20v instead of +/-27v (mine are actually +/-21.8v).

 

Andy

 

[ghost4man]

7 hours ago, andyr said:

 

Yes, that comment was by someone who built the original '8R' version - designed for an 8ohm load (with no claim to drive even 4 ohms, well).  He's also someone who is bold (stupid?) enough to take someone's circuit and "improve" it, by changing component values. 

 

Hugh then created (at my urging  ) a 4R version - which he said would be able to handle a 2 ohm load.  The same circuit & PCB is used - but some resistor values are different (to the 8R version).  Plus, the DC rails are lower: +/-20v instead of +/-27v (mine are actually +/-21.8v).

 

Andy

 

Yeah I had a sneaking suspicion the only way to do this would be by lowering the rail voltage. 

 

Well done ANDY. I am intrigued by your amplifier build. 

 

Interestingly when Maggie's first came out I believe they were getting coupled with valve amplifiers so it is not like they can't be. 

[andyr]

Well, today - at bloody long last - A/Post delivered a parcel which was shipped from west coast USA on April 28th!    (Somehow US Post managed to lose it for a month!   )

 

This contained the binding posts that were needed to finish off my Class A AKSA "Alpha Nirvana" amps - destined to replace NAKSA 80s on my mid panels (3.2 ohms) and ribbons (2 ohms).

 

First amp was completed this arvo - second one will be, hopefully, tomorrow morning. 

 

Then I can finally - it's been a 6 month odyssey! - listen to my 'new' system. 

 

Pics below:

 

Front panel:

 

 

 

Modushop case with their 10mm front panel, handles and feet.

 

The red LEDs light up when clipping occurs on each channel; the green LEDs show that the '+' and '-' DC rails are working.

 

Back panel:

 

 

 

Yes, alas - I stuffed up the position of the amp boards (the 4 hex-head bolts on each side) so wasn't able to use the holes that I had asked Modushop to drill. 

 

Internals:

 

 

 

This build uses 'active cooling' for the output MOSFETs ... rather than (passive) heatsinks.  So the 2 MOSFETs, each channel, are clamped to a Dell CPU cooler - held horizontally, with a 92mm Noctua fan under each CPU cooler (and a 92mm hole in the bottom panel).  Each CPU cooler has to get rid of 130w of heat.

 

The heatsinks by the side of each CPU cooler are for the DC rail transistors in each PS board - which are in the middle of the case, at the top of the pic.  (The heatsinks are over-specced!  )

 

The amp boards, themselves, are mounted on the back panel; underneath them - mounted on the bottom panel - are spkr protection boards.

 

The black discs next to the mains power switch are thermistors which provide a slow start-up to the amp (so there's no 'thump' on turn-on).

 

The white board just above the chassis earth bolt is the fan control board - to reduce the +ve DC rail's 22v to 12v.  This voltage reduction is done with 120 ohm/12w Mills resistors.

 

As I said - hopefully there'll be listening tomorrow! 

 

Andy

 

Edited June 23, 2020 by andyr

[Luc]

Them coloured lights and the black colour make it look like the replacement for the  Redgum Son of  a Gum. Which I can't remember it's name but  yeah, they look a bit like this.

33 minutes ago, andyr said:

(Somehow US Post managed to lose it for a month!   )

Trump+virus probably.

[andyr]

Need to delete this post.

 

 

Edited August 17, 2020 by andyr

[andyr]

Haven't started on the amps yet - too busy listening to the beautiful new sound from my Topping E30s! 

 

Andy

 

Edited August 17, 2020 by andyr

[metal beat]

Happy for ya Andy!!  congrats.  nothing better than an upgrade that actually is an upgrade 

[Bunno77]

Great idea using those heatsinks and the Noctua fans are so quiet

[andyr]

2 hours ago, metal beat said:

Happy for ya Andy!!  congrats.  nothing better than an upgrade that actually is an upgrade 

 

Indeed! 

 

Andy

 

[andyr]

2 hours ago, Bunno77 said:

Great idea using those heatsinks and the Noctua fans are so quiet

 

Thanks, Dan. 

 

Yes, I just didn't want the bulk of 5RU cases!    The CPU cooler/fan solution seemed to me to be elegant ... rather than 'brute force'.

 

Andy