Amplification Best Approach for Efficient Active Speakers

What people are working on at the moment
vinylnvalves
Old Hand
Posts: 1074
Joined: Sat Jan 26, 2013 12:08 pm

#1 Amplification Best Approach for Efficient Active Speakers

Post by vinylnvalves »

I am at a point of not knowing, how to more forward wrt amplification for my active speakers.

I currently have a 3 way arrangement, utilising a JBL 2482 for the mids and the Beyma TPL for the HF.
Currently i am using a 6B4G PP amp with the JBL 2482's. The issue is the efficiency of 114db results in background noise. I have tried to optimise the gain structure utilising the Nadja DSP , only have 500mv output for this channel, this along with padding down the input into the power amp, gives a noise solution which is listenable but not HIFI.
On my Beyma TPL's i am currently using a little T amp (20watts), i tried using an El84 SE amp, but i wasn't man enough (100 db eff = 30 watts to capture dynamics correctly), and was noisy. The T amp IMO sounds as better, ( el84 amp Rodgers thing- so maybe more radiogram)
Putting the T amp in MF thats also quieter, than the 6B4G's. (6B4G amps - Guy Sargent custom monoblocks - with DC heater supplies)

The question is will SS always have a better s/n ratio than valves or is it implementation?

I was going to build a 4 channel chip amp, but have realised that they aren't the quietest solution, was only going this way as they are meant to sound like 2A3's and are an easy build, for someone is more at home with mechanical things.
For the MF i almost want a buffer with little gain, not the 26db the gain clones offer.

Whats course should i take, any recommendations?
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#2 Re: Amplification Best Approach for Efficient Active Speaker

Post by ed »

vinylnvalves wrote: On my Beyma TPL's i am currently using a little T amp (20watts), i tried using an El84 SE amp, but i wasn't man enough (100 db eff = 30 watts to capture dynamics correctly), and was noisy. The T amp IMO sounds as better, ( el84 amp Rodgers thing- so maybe more radiogram)
This doesn't sound right, the beyma's will capture all the dynamics you could want with 2 or 3 watts IME.

Are you using the correct xover/filter on them????.

with efficient speakers like these, JBLs/Beymas, the amp has got to be clean to avoid hummmmm. Any old valve amp will not do, IME you really have to pay attention to the power supply and heating with any speaker over about 96db.
There's nowhere you can be that isn't where you're meant to be
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#3

Post by Paul Barker »

Yes definately agree with all that.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
vinylnvalves
Old Hand
Posts: 1074
Joined: Sat Jan 26, 2013 12:08 pm

#4

Post by vinylnvalves »

I am using what i think is the correct x-over for the beyma's. Currently LR 4th order@1600hz, but have crossed them with the the JBL's as low as 1100Hz, both with zero phase shift (wonders of DSP).

Not sure if i agree with only requiring 3 watts for the Beymas. I like as a rule of thumb to keep 20db of dynamic headroom ( More for home theatre).
Are you basing this on the Beyma PX25 combination? As i feel Steves system runs out of steam, but then i only hear it in big rooms at show's.

Back to the original question, maybe it needs re-asking as,... is it easier to properly implement a SS or a valve amp.

I am nervous that i could convince myself that i need to go back to 6 mono -blocks this time with separate psu's. Previously I could not listen to music in the height of summer, as my listen room got to hot, and i ended up rotating the Altec bass drivers, when the Goop ran to the bottom, every month.
steve s
Shed dweller
Posts: 2829
Joined: Wed May 30, 2007 6:19 pm
Location: east yorks

#5

Post by steve s »

vinylnvalves wrote:I am using what i think is the correct x-over for the beyma's. Currently LR 4th order@1600hz, but have crossed them with the the JBL's as low as 1100Hz, both with zero phase shift (wonders of DSP).

Not sure if i agree with only requiring 3 watts for the Beymas. I like as a rule of thumb to keep 20db of dynamic headroom ( More for home theatre).
Are you basing this on the Beyma PX25 combination? As i feel Steves system runs out of steam, but then i only hear it in big rooms at show's.

Back to the original question, maybe it needs re-asking as,... is it easier to properly implement a SS or a valve amp.

I am nervous that i could convince myself that i need to go back to 6 mono -blocks this time with separate psu's. Previously I could not listen to music in the height of summer, as my listen room got to hot, and i ended up rotating the Altec bass drivers, when the Goop ran to the bottom, every month.
Thats a fair point about big rooms but i dont usually turn it up too loud, at home i can measure 110 db peaks at full wick , the other thing to bear in mind is the sound from my px 25 amps is different presentation to many amps and can make music sound less ballsey if you know what i mean.
My view is a T amp will is unlikely to do any speaker of that calibre justice.
The tube manual is quite like a telephone book. The number of it perfect. It is useful to make it possible to speak with a girl. But we can't see her beautiful face from the telephone number
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#6

Post by pre65 »

steve s wrote: My view is a T amp will is unlikely to do any speaker of that calibre justice.






i think T amps are old hat these days, but not all chip amps are the same. :wink:

I've been pleasantly surprised by a £13 TPA3116 chip amp only today. :D

http://www.ebay.co.uk/itm/400713053693? ... 1497.l2649
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#7

Post by IslandPink »

This is an issue I'm currently getting into, haven't got a full solution, but have some thoughts . Having taken delivery of a pair of Radian 745 Neo-Be's just the other day, I'm in the same situation as these are potentially 114db/w with the horn . There is certainly a fair amount of hiss audible at 'idle' .
In my situation I am not too concerned about going very loud, so I've been thinking of ways of burning gain to get the Radians down to eg. 97 or 98db/w to match the lower-mid driver ( cone ) . I've always intended to be using small SE valve amps for this part of the project and haven't heard anything from the chip amps I've heard ( TA2024 and TDA7297 ) that would dissuade me from this path .

Of the 3 or more ways I'm considering reducing gain, two of these reduce noise too :
(i) Using the output transformers in non-intended tapping to get less output : ie. 4-ohm tap rather than 8-ohm . This will work OK in other ways for me as the transformers were designed for 2A3 but it would be nice to run 45's to get a touch more tranparency and save on some current from the power supply .
(ii) Series resistance ( eg. 8R ) at the driver which seemed to work OK without any sonic effect ( at that value ) as long as planar resistors ( N.I. ) were used . I've found the 'BI Industries' resistors from RS to be very transparent indeed, with no perceptible loss of dynamics, if values less than about 15R are used with an 8R driver .

Other than these, I think the issues are getting very clean filament supplies ( Andrew Lehane on this board has developed some very clean DHT units for output and driver valves ) and reducing whatever you can on other noise sources from EMI pick-up and non-optimum earthing . Some of this debugging will require a bit of measurement and characterisation of the 'content' of the noise , I reckon . One of the awkward sources of noise with these drivers is EMI getting into the speaker wires and causing 'twitter' - this happens with the amp turned off ! I've noted that the commercial ( Tellurium Q 'Ultra-Black' ) cables I recently bought are MUCH better in this regard than my previous home-made litz/cotton cables . Series resistance at the driver helps here too - but do consider the speaker wires, as this contribution is now almost nil in my system .
"Once you find out ... the Circumstances ; then you can go out"
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#8

Post by IslandPink »

Sorry, one more thing - I think there's definitely some milage in investigating small class-A jFET-based amps for this purpose ; but I'm not the guy to ask for circuits !
I'm not convinced any Mosfet-based solution ( I own & use - sometimes - a Pass-labs F4 ) is good enough in the treble to match a good small valve amp .
"Once you find out ... the Circumstances ; then you can go out"
vinylnvalves
Old Hand
Posts: 1074
Joined: Sat Jan 26, 2013 12:08 pm

#9

Post by vinylnvalves »

Before i went to the DSP, and was using more conventional approaches for the crossovers i would pad down across the speaker terminals, and that of course reduced the background noise. Maybe an attenuation bridge using 10W Caddocks would be transparent, and achieve the same result without needing to change my MF amp.

I had been discussing with John Wood some time back making me dual channel mono blocks, however he did not convince me that kT88 PP with a buffer driver stage was the right approach, as his approach was to have the sheer driving force there whether you needed it or not. (£3K+ was the other reason!)

Surprised that one thinks the Pass lab stuff won't be as transparent as a valve.

On paper the Hypex D class amps look very impressive (won't want hum in the logic driving machinery around. Its a shame they don't have something with around 15db gain.

The only SS amp i have heard driving horns was a "Le monstre" but my knowledgeable friend, says old technology - and you might as well use Blackburn geranium diodes! Putting that aside is it a nostalgia trip or are they work pursing.
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#10

Post by IslandPink »

vinylnvalves wrote: Surprised that one thinks the Pass lab stuff won't be as transparent as a valve.

.... old technology - and you might as well use Blackburn geranium diodes! Putting that aside is it a nostalgia trip or are they work pursing.
Pass F4 is not in same class as a good valve amp from 2-3k up, I know that for sure. It's spacious and musical overall but has a slight greyness in the upper treble that is easily beaten by valves .
Pass F5 ( Nick Gorham's ) is a bit better but still not there .

Germanium transistors - now that's worth a go . I have heard a Germanium amp by one of our guys ( MikeH ) that had a sort of transparency to the treble that I haven't heard from Silicon .
"Once you find out ... the Circumstances ; then you can go out"
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8867
Joined: Mon May 21, 2007 9:42 pm

#11

Post by Paul Barker »

Agree on f5 it's only failing is the same place in your situation it would have to be good. It's not like it's bad at all. It just isn't as good in that region as some other designs. Some of which are solid state.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
rowuk
Old Hand
Posts: 453
Joined: Sun Jun 01, 2014 2:50 pm
Location: Germany

#12 Efficient speaker and dynamics?

Post by rowuk »

The thread title was very interesting, the misunderstanding about headroom and dynamics was surprising.

If we look at the average distribution of energy for real music (electronic or acoustic), we see that there is only a relatively small band of frequencies that really need acoustic power. Generally we need the most output only up to about 250 to 300 Hz. Above that, the requirements drop VERY quickly.

Here is an excellent article on energy distribution.
http://sound.westhost.com/articles/fadb.htm
This means if we want to reproduce "live levels" with a 100dB/watt woofer array, a couple of watts will be enough - even for bigger listening distances. I have doubts that below 50Hz that the array is still 100dB/Watt, but it could be calculated.

When we talk about frequencies above 1000Hz, the power distribution is about 10dB down, with a 118dB midrange, we are now an additional 18dB below the power needed for 300 Hz. If 10 Watts is enough at 300, 1/10th of a watt is enough for the 118dB/watt midrange. The Beyma tweeter at 100dB also only needs a fraction of a watt to keep up with woofers as the power distribution curve is many more dB down at 10KHz.

The moral of the story is simple: for high efficiency systems, 20 watts is enough for the woofers, 5-7 watts is enough for the high efficiency midrange and 1-3 watts is enough for the tweeter. With this power, we have PLENTY of headroom. If we need headroom, then it will be where the greatest amount of power is necessary - bass and lower midrange. More power is not the solution here because the voice coils heat up and we get thermal compression. We either need a bigger array of bigger speakers, or a horn. For infrasonics, the only solution is cone area and horsepower. Having headroom below 50 Hz can mean 4x 18" woofers or 8x 15". If you get used to this, you will NEVER go back.

It is amazing that many respected audiophile speakers are missing "headroom" exactly where music demands it the most......
Whenever I feel blue, I start breathing again.
vinylnvalves
Old Hand
Posts: 1074
Joined: Sat Jan 26, 2013 12:08 pm

#13

Post by vinylnvalves »

Theres alot of information to take in here - and i am concluding there is no free lunch. Maybe there isnt one solution for MF and HF. What what makes me smile is the comments on the presentation of the different amps. Taking a step back from it - an amplifier is a wire with gain, and should not add or take anything away from the signal passing through it. I have come across "response contouring" before in AN equipment where the end of the frequency band are lifted for LF and dropped at HF to give that sound.

It looks from the comments that valves seem to be what is being recommended, do they have to be DHT's or can i get away with pentodes as triodes. I did see at one show an air cored OPT ( i think) on a >5K hz amp. I assume i would have to get transformers wound for the freq bands in question ( as its a waste to have lots of Iron work for the LF that won't get saturated)
Do i assume that all the proposals for the HF are SE amps and not PP, so is it the 2nd order distortion thats beguiling for HF?
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#14

Post by Nick »

Taking a step back from it - an amplifier is a wire with gain, and should not add or take anything away from the signal passing through it.
Given that the above is an optimistic fiction, and will never happen in the real world (at HF a wire isnt just a wire), we have to start from the position that the amplifier will have something of itself to add, so if we accept that, it makes sense that we consider the various aspects of amplifier with what they are being used for.
Do i assume that all the proposals for the HF are SE amps and not PP, so is it the 2nd order distortion thats beguiling for HF?
You could assume that, but its more the case that the 2nd harmonic as long as its low enough can be ignored, and the higher order terms become very important.

I also believe based on my own tinkering, though I have no hard evidence that the 2nd harmonic will also act as a mask for higher order harmonics.

I would suggest if you want to avoid the extra hassle of DHT that a SE triode el84 would make a nice HF amp in conjunction with an amorphous output transformer, and SE el34 would do well as the midrange amp, again with amorphous OP TX, then maybe a el34 PP for the LF.

Or you could consider OTL for the HF and mid.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
ed
retired
Posts: 5384
Joined: Thu Jun 21, 2007 4:01 pm
Location: yorkshire
Contact:

#15

Post by ed »

vinylnvalves wrote: Not sure if i agree with only requiring 3 watts for the Beymas. I like as a rule of thumb to keep 20db of dynamic headroom ( More for home theatre).
Are you basing this on the Beyma PX25 combination? As i feel Steves system runs out of steam, but then i only hear it in big rooms at show's.
this thread should explain why I say three watts should be adequate. Further down the thread is a list of amps I have used.

http://www.audio-talk.co.uk/phpBB2/viewtopic.php?t=5562
IslandPink wrote:Pass F4 is not in same class as a good valve amp from 2-3k up, I know that for sure. It's spacious and musical overall but has a slight greyness in the upper treble that is easily beaten by valves .
Pass F5 ( Nick Gorham's ) is a bit better but still not there .

&
I'm not convinced any Mosfet-based solution ( I own & use - sometimes - a Pass-labs F4 ) is good enough in the treble to match a good small valve amp .
what is it you are listening to at 2-3k up Mark?

The F4 has a completely different spread of harmonics to the F2 for instance, so it's a bit misleading to tar all mosfets with the same brush.

I tried this conversation with Steve but it was difficult to get my point across, which is that I found the beymas dull, simply because they don't address the frequencies I am interested in. The can't reproduce the instruments that I home in on, and at best they can only address the harmonics....

back to the original question, I believe the amp that you want for the particular ranges you have decided to pick on is the one that you like the best. I don't believe there is a SS for this and a valve for that...these categories are just subjects for arguments.... :)
There's nowhere you can be that isn't where you're meant to be
Post Reply