[Julf]

Specs for the bridged mono/one SMPS1200A400 or two SMPS600: 400/800/1200W @ 8/4/2 Ohm, safe to 1 Ohm but no rating.  With one SMPS600: 400/800 @ 8/4 Ohm, safe to 2 Ohm but no rating.

Where do those specs come from? What, in your view, is the limiting factor/parameter for going down below 2 ohm?

[James Romeyn]

Where do those specs come from? What, in your view, is the limiting factor/parameter for going down below 2 ohm?

Email from Bruno to me provided the following specs for single SMPS600 powering:

• One NC400: 200/400/600W @ 8/4/2 Ohm, safe to 1 Ohm but no power rating
• Two NC400: 200/400W @ 8/4 Ohm, safe to 2 Ohm but no power rating

I'm pretty sure the data sheet echoes the above specs.

Whether modules are bridged or not (see Jan-Peter's comments in this thread), maximum power per module = 600W into the speakers, requiring 600W of power supply current, yielding safe current to 1 Ohm minimum but no power rating. 

Take #1 above: doubling the modules (bridged pair) and doubling PS current (1200W) = 2x W across the board and impedance specs stay constant.  Ditto 600W PS powering bridged pair NC400 = 2x W across the board and impedance specs stay constant.

So PS current and number of NC400 determines minimum impedance into which NC400 is safe and minimum impedance into which the amp delivers rated power.  AFAIK this relationship is linear and unrelated to whether NC400 is single or bridged.   

I compared several times and can't seem to tell a difference between mono blocs each powered by SMPS600 and stereo powered by single SMPS1200A400 (about $400 lower cost).         

[Julf]

Email from Bruno to me

Thanks - I am curious, did Bruno actually use the word "safe"?

So PS current and number of NC400 determines minimum impedance into which NC400 is safe and minimum impedance into which the amp delivers rated power.  AFAIK this relationship is linear and unrelated to whether NC400 is single or bridged.

I agree with the "minimum impedance into which the amp delivers rated power", as the amps become current-limited. But I am curious about the "safe" part. What does "unsafe" mean? Amp goes in to oscillation? Amp gets damaged?

[James Romeyn]

Thanks - I am curious, did Bruno actually use the word "safe"?

I agree with the "minimum impedance into which the amp delivers rated power", as the amps become current-limited. But I am curious about the "safe" part. What does "unsafe" mean? Amp goes in to oscillation? Amp gets damaged?

Yes, IIRC Bruno used the word "safe."

It means you are, employing the late James Bongiorno's words to me, "tempting fate."  It means to employ a load with less than the rated safe spec is "unsafe" and hence outside the intended use/application of the device.  I presume all vital components have similar performance ceiling, meaning any number of immediate major component failures from one to most of the major components might fry.  Conversely, if the many failure systems all work perfectly, you might replace only a fuse.  That's where "tempting fate" comes in. 

The point is, specs like minimum load impedance can be confirmed, and must be confirmed if there is any doubt, and to ignore or exceed the specs is "unsafe" and to be avoided.  IOW, spend $X now on more PS or risk spending even more $ on blown amp/speakers later.  Like Dirty Harry asked, "Do you feel lucky?"

Professionally and legally and ethically, I presume, if/when any amplifier component burns in use, the OEM can claim such burn proves abuse.  IOW, it would be 100% up to the user to prove normal use, and by definition they would fail in legal claim unless there was huge class action with some huge rate of failure. 

When Kawasaki released a new liter class Superbike several years ago, a major motor failure soon occurred, which caused Kawasaki to replace a bunch of motors.  They didn't wait for the lawsuit because they'd have lost and incurred more damage in reputation.  As it turned out, no harm/no foul, all was soon tranquility.   

If such amplifier use burnt any loudspeaker component, beware that per professional and legal definition, a burnt loudspeaker component always and only defines abuse, whether or not the speaker company agrees to replace the burnt component, which they generally do not because to do so only promotes further abuse.  Usually charging such abuser for their abuse one time is enough to cause the intended behavior change.     

[Julf]

Yes, IIRC Bruno used the word "safe."

Interesting. This is the first time I hear that low load impedances might be "unsafe".

The NC400 data sheet does contain a section with "Absolute Maximum Ratings", with a warning that irreversible damage might occur as a result of exceeding the specified limits. Load impedance is not among those parameters (it is only in the "Recommended Operating Conditions"). The only load-related parameter is Peak Output Current, and the data sheet specifies that it is "Guarded by current limit at 24 A" - clearly the current protection circuitry is there to protect against excessive output current (caused by a load impedance that is too low).

It means you are, employing the late James Bongiorno's words to me, "tempting fate."  It means to employ a load with less than the rated safe spec is "unsafe" and hence outside the intended use/application of the device.  I presume all vital components have similar performance ceiling, meaning any number of immediate major component failures from one to most of the major components might fry.  Conversely, if the many failure systems all work perfectly, you might replace only a fuse.  That's where "tempting fate" comes in.

OK, so you don't actually know what the supposed failure mode would be, and what the actual limiting factor is?
 

The point is, specs like minimum load impedance can be confirmed, and must be confirmed if there is any doubt, and to ignore or exceed the specs is "unsafe" and to be avoided.

See my comment above about "Absolute Maximum Ratings" vs. "Recommended Operating Conditions".

[James Romeyn]

I think we're going in circles, on the same page, etc. 

Anything can fail.  Current limit protection comes under the heading "anything." 

The sole purpose of the amp's current protection is to protect components from burning as a result of current exceeding a certain threshold, beyond which a component's risk of frying rises exponentially.  All these items don't matter except at the margin.  In the middle of the curve or below they are irrelevant. 

The lower the load impedance the higher is current demand.  If you believe current demand and load impedance are universally unrelated I strongly disagree.   

When one clearly crosses the line (beyond gray area) re. minimum load impedance, and for whatever reason the amp's "current limit" protection simultaneously fails for any time period no matter how brief (they only must occur at the same time), what is the obvious or implied risk?  Burn component and that acrid smell of money flying out one's wallet with which I'm all too familiar. 

We're back to "tempting fate."

The less are my personal finances at risk the less interest I have in persons ignoring the advice.   

[Julf]

I think we're going in circles, on the same page, etc. 

Quite. Maybe the easiest way to clarify things would be if you could post the actual email you got from Bruno?

Anything can fail.  Current limit protection comes under the heading "anything." 

Anything can fail - but assuming anything can fail there is nothing that can be considered "safe". Everything can fail as a result of something else failing.

The lower the load impedance the higher is current demand.  If you believe current demand and load impedance are universally unrelated I strongly disagree.

Where do you think I have stated that current demand and load impedance are universally unrelated? I specifically stated that peak output current is a load-related parameter.

When one clearly crosses the line (beyond gray area) re. minimum load impedance, and for whatever reason the amp's "current limit" protection simultaneously fails for any time period no matter how brief (they only must occur at the same time), what is the obvious or implied risk?  Burn component and that acrid smell of money flying out one's wallet with which I'm all too familiar.

Would Hypex really specify a limit for the load impedance based on "this could cause damage to something, *but only if the protection doesn't work as intended*?

We're back to "tempting fate."

That still doesn't address potential failure modes / limiting factors.

[Julf]

Yes, IIRC Bruno used the word "safe."

I actually checked with Hypex. They don't see any reason why connecting two modules to the supply would suddenly raise the 'safe' load impedance to 2 ohm either.

[James Romeyn]

I finally built two bridged mono.  I estimate it's one clear step up in performance vs. regular SMPS600 mono or SMPS1200 stereo.  Everything that's good about the regular amp is improved, even the bass, if that's possible. 

I have earthed all chassis prior to this bridged amp.  The only thing that solved fairly bad AC Mains hum in the bridged amp was to lift the chassis earth.  Appreciate comments if you have any light to shed on the issue of AC Mains noise, bridged vs. not bridged, chassis earthed vs. not.

[gstew]

James,

Thanks for the update. Pix would be great!

Did you use one SMPS per channel or one per module?

This may get me off the mark to convert mine to a bridged set!

Greg in Mississippi

[James Romeyn]

Howdy Greg,
I used one SMPS1200A400 per two NC400, in Siliconray's "RE2507-NC" chassis.

I plan more thorough testing by Tuesday, including direct A-B with regular mono SMPS600.

So far it seems like there's less noise modulating with the music, less grunge, it has more density like the best tube amps (I'm not saying it equals the best tubes in this regard, just more of this effect than regular mono). It sounds calmer, lower stress. 

The stock mono is already pretty good in all these aspects, it just seems like the bridged amp is better.

SMPS1200 is the only power supply to use unless cost is no concern.  Two 1200s cost slightly less than one 600.   

[Roger A. Modjeski]

Keep in mind that when you bridge an amp each channel thinks its driving a load of half of what you put on it. To the bridged amp an 8 ohm load looks like 4.  and 2 ohms looks like 1.

Very few amps will like being bridged into a two ohm load because it looks like one ohm.

[James Romeyn]

Roger, what you posted I have read is the case with stereo amp used as bridged mono.  But IIRC Hypex principal Jan-Peter said it's not the case with bridged NC400, but that may apply only when bridged PS current is doubled vs. stereo.   

My 93 dB AudioKinesis Zephrin 46 speakers are unique in that each channel comprises two 16 Ohm separate loudspeaker loads: a "Main/On Axis" section and "Late Ceiling Splash" section.  The two sections can wire parallel (8 Ohm) or series (32 Ohm).  Bridged NC400 makes 100W @ 32 Ohm, so I'm anxious to audition it both ways. 

I shall also audition bridged NC400 powering the 16 Ohm "Main/On Axis" section, with my receiver powering the 16 Ohm LCS section.   

[Roger A. Modjeski]

Roger, what you posted I have read is the case with stereo amp used as bridged mono.  But IIRC Hypex principal Jan-Peter said it's not the case with bridged NC400, but that may apply only when bridged PS current is doubled vs. stereo.   

My 93 dB AudioKinesis Zephrin 46 speakers are unique in that each channel comprises two 16 Ohm separate loudspeaker loads: a "Main/On Axis" section and "Late Ceiling Splash" section.  The two sections can wire parallel (8 Ohm) or series (32 Ohm).  Bridged NC400 makes 100W @ 32 Ohm, so I'm anxious to audition it both ways. 

I shall also audition bridged NC400 powering the 16 Ohm "Main/On Axis" section, with my receiver powering the 16 Ohm LCS section.

I use the term bridged to mean that one channel is driven 180 degrees out of phase and the load is connected between the hots. Does he do something different?

The supply current is doubled.

[Julf]

But IIRC Hypex principal Jan-Peter said it's not the case with bridged NC400, but that may apply only when bridged PS current is doubled vs. stereo.

In view of the previous exchange in this thread, can you share with us Jan-Peters exact words?

[Julf]

The only thing that solved fairly bad AC Mains hum in the bridged amp was to lift the chassis earth.

Do you have a schematic of your earthing scheme?

[James Romeyn]

In view of the previous exchange in this thread, can you share with us Jan-Peters exact words?

I'll search his posts here and quote him if I find it.

Again, the speaker can be wired parallel for flat 8 Ohm above the bass range, 6 Ohm minimum, worst case moderate phase angle.  8 Ohm = 93 dB per 1W input (2.83V).  32 Ohm = 87 dB per 1/4W input (2.83V).     

I presume SMPS1200A400 powering Bridged NC400 = 100W @ 32 Ohm.  IIRC someone I trust stated NC400 does not change distortion spectra into different impedance loads, but I personally checked and same as every SS amp I know of, it's distortion spectra degrades (albeit not a lot) as load impedance decreases.  This of course implies the opposite, being more pleasing distortion spectra as load impedance rises (up to a point I suppose, that point being unknown to me...there are certainly very few speakers with high load impedance to test this). 

Supposing Bridged NC400 had reasonable to more than reasonable magnitude of head room powering the 32 Ohm load described above.  Does anyone know why performance might improve powering the 8 Ohm load? 

It's a completely different animal, but just for reference, I auditioned over long period my analog chip amp powering both the 8 Ohm and 32 Ohm loads (into these loads the amp makes 100W and 25W, respectively).  Powering the 32 Ohm load it was immediately obvious that treble quality improved, with a large portion of that shimmering magic of the best SE tubes.  But after time, overall I preferred the 8 Ohm load for its power advantage.  It just sounded more "grounded."  As usual, still no free performance lunch.         

It appears Bridged NC400 has good power reserve into 32 Ohms, with better perceived distortion spectra vs. typical 8 Ohm load.  A 100W amp with a 1200W power supply...the proverbial iron fist in a velvet glove. 

[Roger A. Modjeski]

Roger, what you posted I have read is the case with stereo amp used as bridged mono.  But IIRC Hypex principal Jan-Peter said it's not the case with bridged NC400, but that may apply only when bridged PS current is doubled vs. stereo.   

My 93 dB AudioKinesis Zephrin 46 speakers are unique in that each channel comprises two 16 Ohm separate loudspeaker loads: a "Main/On Axis" section and "Late Ceiling Splash" section.  The two sections can wire parallel (8 Ohm) or series (32 Ohm).  Bridged NC400 makes 100W @ 32 Ohm, so I'm anxious to audition it both ways. 

I shall also audition bridged NC400 powering the 16 Ohm "Main/On Axis" section, with my receiver powering the 16 Ohm LCS section.

If your only load option was 32 ohms then bridging would make sense.

Again, if bridging is hooking up the load between the hots you will get the same power using just one channel as you will using two, so why use two. One channel is 100 watts at 8 ohms, isn't it?

From Hypex website (technology ucd)

Reasons for choosing full bridge:
dot_or Voltage swing. At the current state of MOSFET technology, diode recovery problems become prohibitive for devices of VBRDSS>150V. If more than, say 50Vrms of output swing is required, either current-steering diodes must be added, or a full-bridge configuration selected. We are aware of people building class D amplifiers using 200V MOSFETs without using current-steering diodes, but these designs have serious EMI and efficiency problems because of this.
dot_or DC operation. In non-audio applications where DC output voltages and currents are required, bus runaway (pumping) pretty much rules out half-bridge designs.

Not reasons for choosing full bridge:
dot_or Pumping (in an audio application). The storage capacitance needed to produce a suitable DC supply is already sufficient to render pumping effects pretty much a non-issue. Much is made of the pumping problem by people who have full bridge amplifiers to sell.

Reasons for choosing half bridge
dot_or Economy. For up to a few hundred watts, half bridge power stages are the lowest cost solution, and are also the most compact. Otherwise put: for the vast majority of applications, half-bridge amplifiers are the default choice.

Not reasons for choosing half bridge:
dot_or “Better performance or sound quality”. Given equal care in design, there is no difference in attainable performance in full-bridge or half-bridge amplifiers. No sonic differences have been found caused by the power stage arrangement.

[James Romeyn]

It's not my amp.  It belongs to a client.  Actually, one NC400 makes 200W @ 8 Ohm.

Again, I presume, all things being equal, distortion spectra improves with increasing load impedance, which If this is correct, might justify Bridged 100W @ 32 Ohm over single NC400 @ 8 Ohm (latter makes twice the power).

I'm wiring the series load right now while my mountain bike patch dries. 

[James Romeyn]

I made, well, a lot of NC400 single mono/SMPS600 and stereo/SMPS1200.  For reasons relating to cost and performance above the bass range, some time ago I "downgraded" from single NC400 to my old standby analog chip amps.

Brand new, no hours, Bridged NC400 making 100W into this 32 Ohm load is a revelation.  It simply blows away mono NC400, not even close.  It's a completely different animal.  Of course YMMV, depending on the load, room, personal taste, ad infinitum. 

I am not at all looking forward to downgrading from Bridged NC400 to the analog chips now. 

It should take only a few minutes to switch to 8 Ohm load to directly compare.  Considering with peaks above 100 dB there was no audible distortion into 32 Ohm load (100W), I predict worse performance (less pleasing distortion spectra) into 8 Ohm, but we'll see.

Remember that the higher the load impedance the better is loudspeaker audible performance, and vice versa.  This is because the higher the load impedance the smaller is the ratio of the wire's contribution to the entire load (wire + speaker). 

Wire is effective 9AWG, about 12' total.