Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[John Roberts {JR}]

In your statement above you have once again correctly noted increased output can be acheived using higher voltages. This is a statement true of any amplifier design. What needs to be considered, and of course you already know this, is that all components in the design must be capable of handling those higher voltages. I see this problem on a regular basis when rebuilding/refurbishing/blueprinting amplifiers that have been "rebuilt" by less than qualified people. Many times lower capacity resistors or lower voltage capacitors are used in critical circuit points which will lead to failure.

The second point I would make is that regardless of the design or type, an amplifier should be designed to replicate the preamplifier signal without altering that signal. The amplifier should be a linear and transparent design. Higher voltages can be an aid to reducing distortion, or they can lead to increased gain and higher levels of distortion, take your pick. Generally speaking power amplifiers tend to run best when designed to use 70-80% of the components capabilities. In other words, don't push those transistors or tubes past that point in your design and all should be well.

Today, with the exception of guitar amplifiers, almost all power amplifiers used to replicate audio are designed around solid state components. My criteria for amplifier choice has always included the distortion ratings at 4 and 8 ohms. This IMO tends to point towards overall quality of the components and care in design. Next would be the FTC ratings, manufacturers reputation, ease of service, and customer support. After that it's end user preference based on the job at hand, then finally the understanding that you get what you pay for.

While I won't get into a full blown discussion about amplifier design philosophy, and I won't discuss tube design ( because I have never designed one), but the science of voltage tolerance for components is pretty mature.

I capacitor that is rated for 100V damn well better work at 100V. There are multiple reasons for derating that voltage, since you don't want an amplifier that is safely within the capacitor tolerance release it's smoke during transient high line events. So conservative design leaves some headroom for that, and perhaps a little for JIC. It is a commonly held ASSumption that the capacitor maker, also builds in some headroom so his 100V caps will all work at 100.1V. I knew one well respected amp designer who would cheat a little on his cap specification to use up this cap maker's headroom too. He got bit pretty bad in one design, which he pretty much blamed on one brand of capacitor, who apparently was playing by his his thin headroom rules too. I hope that broke him of that practice.

Solid state power devices have fairly well defined failure mechanisms, but some failure modes, like secondary breakdown may not be well understood by all amp designers. Many of the popular modern amps use the "exact" same complement of power devices, so amp models that push the power output relative to their similar peers may allow for less headroom (pure speculation on my part that should be confirmed by looking into the BOM). 

Distortion or linearity in solid state power amps generally tracks linearly with current output, so 4 ohms distortion is dominated by the 2x current output of 8 ohm power (and 2 ohm is that much worse).

I am not aware of any mechanism that would cause power amps to run better using 70-80% of components available maximum output. I don't know much about tubes, but do believe they have a more gradual overload characteristic, so your rule of thumb may make sense with them.   

When repairing "hot rodded" product all bets are off, as prudent engineering is rarely  practiced by hot rodders, who by definition are pushing some envelope or the other.

JR

[John Roberts {JR}]

Crown has a series (the CTS) that on several of their models they are rated for the same wattage into an impedance range of 4:1.  This range does vary on different models.  For example on one model 4-8-16 ohms are all the same power.  On another 2-4-8 are the same power.

So they do have something along the lines you are talking about going on.

I vaguely remember a series of Crown install amps (driven rail IIRC) that switched around some transformer windings in series-parallel to optimize PS voltage for load impedance..  The driven rail amps were not exactly hifi for other reasons, but definitely a cost effective approach for install. (I may be thinking of different amp series than you... tho).

Speaking of obscure approaches, IIRC QSC had one amp, where you could connect all the outputs devices from two channels in parallel to make one high current output.

Amp design is pretty mature so many tricks have been tried. I have a few that still haven't but they probably never will as class D technology takes over the world.

JR
 

[Ivan Beaver]

Speaking of obscure approaches, IIRC QSC had one amp, where you could connect all the outputs devices from two channels in parallel to make one high current output.

The Crown MA/MT series of amps had that unique "parallel" configuration.

When you switched the amp into parallel mode-it did not parallel the inputs.  You would also use a large piece of wire (23ga) and tie the two amplifier "hots" together.  NOTE TO READERS-DO NOT DO THIS on any other amp model-unless you like a trip to the repair shop.

The same drive circuitry was used, only the outputs were now in parallel-which allowed for a higher current capability into the load.  The voltage swing was not any greater (as if it was in the bridge mode).

This did NOT give any greater power/output into normal impedance loads (because of the voltage swings)-but did give the capability to drive a 1 ohm load.

Sometimes you don't have a choice.

ALSO HOWEVER-I have heard from some reliable sources (I have not actually measured or checked it) that this mode "starts to fall apart" sonically above 5Khz.  From what I understand it gets "grainier" or not as clear as when driven in the normal modes.

[David Parker]

Amplifiers don't "react" to their load impedance.

Mac

I used the best wording I could come up with. It did spur on the discussion I was looking for. Maybe I could go on the peavey forum and they would answer my question. I still believe that QSC's GX series was designed with the same thing in mind, and possibly used similar technology to get there. I remember a 4 channel amp several years back where you could bridged mono the separate amps into pairs to make it work as a stereo amp. Probably no similarities.

[Mac Kerr]

Maybe I could go on the peavey forum and they would answer my question.

What part of your question has not been answered? The 4080HZ is a higher powered amp, but gives up 2Ω capability, the 4000 can drive a 2Ω load, but is only half the power of the 4080HZ at any given load impedance.

JR's explanation was pretty complete.

There have been more than one 4ch amp that could be bridged into a reduced channel count but higher channel power. What does that have to do with 2 different amps with different output specs?

Mac

[Marty McCann]

The CS-4000 and CS-4080HZ are very similar under the hood.

We needed a model that could actually put more power to our 8 Ohm high power Black Widow loudspeaker models.

So the CS-4080HZ started with a PS that could have a higher voltage swing into an 8 Ohm load.  This of course also resulted into a higher voltage swing at 4 Ohms as well.  However at this point (4 Ohms) we had to engage the current limiting circuitry to keep the smoke inside.

HZ indicated Higher Impedance (Z) operation.

[David Parker]

The CS-4000 and CS-4080HZ are very similar under the hood.

We needed a model that could actually put more power to our 8 Ohm high power Black Widow loudspeaker models.

So the CS-4080HZ started with a PS that could have a higher voltage swing into an 8 Ohm load.  This of course also resulted into a higher voltage swing at 4 Ohms as well.  However at this point (4 Ohms) we had to engage the current limiting circuitry to keep the smoke inside.

HZ indicated Higher Impedance (Z) operation.

Thanks for the info. Wondering if the QSC GX series uses some similar technology?

[ProSoundWeb Community]