advertisement


Amplifiers known to be reliable for <4ohm speakers

I recall being in a dealer’s in Windsor (or Eton) and I recall them having an integrated amplifier (c£5k) by a brand called ‘Storm’ which had a dial where you could adjust the ohms
 
My Bel Canto Ref1000Ms are stable into a 2 ohm load according to the manual (they deliver 1000 wpc into 4ohms).
 
Oo, that’s interesting. I’m not technically minded so it would be interesting to know more!
 
... IIRC (and I could be wrong) ATC use a Hafler like "swinging PSU, grounded output stage" topology and (more importantly) again IIRC they use old style lateral mosfets, which are the closest to an indestructible output device you can get. They like to run at least 100mA quiescent so do run warm yes. ...
ATC's amplifiers are called "grounded source" so, yes I do assume they use Jim Strickland's 1984-patented "trans*nova topology" as did Hafler and Acoustat. I don't know about the MOSFETs (but I have found the ATC folks at shows to be free with information when asked so it could be confirmed I am sure).

The design is used in studios world-wide so I assume it has to be robust. To me, substantial heatsinking is a key part of that. I posted an ATC photo link earlier and here's a link to @Jim Audiomisc's Armstrong 732 showing me that it too is likely to be robust. That's the sort of construction I would look for in a robust amplifier.
 
ATC's amplifiers are called "grounded source" so, yes I do assume they use Jim Strickland's 1984-patented "trans*nova topology" as did Hafler and Acoustat. I don't know about the MOSFETs (but I have found the ATC folks at shows to be free with information when asked so it could be confirmed I am sure).

The design is used in studios world-wide so I assume it has to be robust. To me, substantial heatsinking is a key part of that. I posted an ATC photo link earlier and here's a link to @Jim Audiomisc's Armstrong 732 showing me that it too is likely to be robust. That's the sort of construction I would look for in a robust amplifier.

FWIW I tried to arrange that the current gain of the output drivers+output devices would fade at about the current that was OK from a device safety POV. Sort of soft limit effect. But set this high enough not to be problem drving even awkward speakers . I then tweaked the PSU so the rails would drop under persistent high current, which eased both the SOA and output device heating.

In effect, aimed to make use of the peak/mean stats of music. Get the maximum performance I could when playing music into nasty loads at ear-bashing levels.

Result was - barring idiot reviewers - no need for a hard current limiter or protection circuity. Just some power line fuses. Note all the values in the "Popular HiFi" review (lower part of; https://www.audiomisc.co.uk/Armstrong/reviews/finale/1982.html) didn't blow the (5 amp) power line fuses. Despite the reviewer in the HFN review cluelessly blowing one and not knowing that they'd done. Despite having the circuit diagram, etc. (sigh).

In reality, of course, most people will buy a powerful amp and use it happily at just a few watts per channel. 8-]
 
>SNIP<
https://www.audiomisc.co.uk/Armstrong/reviews/finale/1982.html) didn't blow the (5 amp) power line fuses. Despite the reviewer in the HFN review cluelessly blowing one and not knowing that they'd done. Despite having the circuit diagram, etc. (sigh).

In reality, of course, most people will buy a powerful amp and use it happily at just a few watts per channel. 8-]
Good grief, I actually read all of that!
And I am one who uses a >250W amp at vanishingly low levels.
 
I asked ATC direct and their reply is below

It would be safe to assume that the SIACD will run a little warmer driving a 4 ohm speaker but given that you state you will use them at low to moderate levels, I don’t think you will have any issue.

In addition, although the W8 are 4 ohm rated, from this review https://mcusercontent.com/d54a38bd8...9b-7090-e1ce6c17a590/W8_Hi_Fi_News_Review.pdf
it appears that the minimum impedance points are relatively high in the frequency range where there is less energy in a musical signal.


I don’t really understand all of what it means but it sounds reassuring :D. Why would impedance drop in the higher frequencies? Would it be more common in the bass region trying to push a bigger cone?

What do I know………..:eek::D
 
The load that a speaker imposes on an amp is a much more complicated question than most people realise. It’s not just sensitivity and impedance you need to consider; have you ever wondered what was being referred to when people talk about resistive loads, capacitive loads, inductive loads and phase angle? Check out this great article: https://www.stereophile.com/reference/707heavy/index.html
 
It's my one reservation about using a class-A amplifier. Even though on paper, 30WPC is plenty for a pair of benign 90dB/w/m loudspeakers, I don't think it'll be as truly effortless as a much more powerful amplifier - even at sensible listening levels where barely 1W average is called for. Bring back power meters, I say.
 
Well, 30W into 90dB/W/m maxes out at around 105dB/m peak. Allowing for 20dB peaks, that's an average of 85dB at one metre. At the listening position at three times the distance, that's a pretty average level. 120W is more accommodating with additional 6dB headroom.
 
The load that a speaker imposes on an amp is a much more complicated question than most people realise. It’s not just sensitivity and impedance you need to consider; have you ever wondered what was being referred to when people talk about resistive loads, capacitive loads, inductive loads and phase angle? Check out this great article: https://www.stereophile.com/reference/707heavy/index.html

Add to the above factors like the speaker impedance being temperature and often signal level dependent! Speaker designers essentially "throw over the wall" all these problems and leave it to amp designers to cope. Then the amp gets the blame if it struggles.
 
Better not tell @foxwelljsly that! :D
Pardon? :)

It depends on the size of your room and how far you are from the speakers. 107dB+ 1m from the speaker is a lot less at 5m. But it's still quite loud and more of a party trick than something you'd do all the time.

My 405 was way off that when it exploded. Honest.
 
Well, 30W into 90dB/W/m maxes out at around 105dB/m peak. Allowing for 20dB peaks, that's an average of 85dB at one metre. At the listening position at three times the distance, that's a pretty average level. 120W is more accommodating with additional 6dB headroom.

I measured 107dB peak 1m from a speaker using an amp rated at 150w WPC into 4r close to full wick (quad 520) and speakers rated at 92dB/w @1m and a constant 4r (Kef 104/2). The speakers are rated at 112dB peak at 1m.
 


advertisement


Back
Top