Enjoy the benefits of a premium membership!
Upgrade Account
-
TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
True power output?
- Thread starter bassbrother666
- Start date
That means that this paper is mistakable
Link Removed
At the other hand
That looks like there is in any situation 3dB of headroom for the output stage that can't be used (abused) for loudness.
Either way, as the amplifier refuses to amplify 450 watt bursts it can't be rated like that, independent of the fact it probably could with APM not active. If there was a supply of 900 watt but 6dB of Headroom margin reserved by the APM the result was exactly the same.
The trace looks like as it is taken on note E2 (or in neighborhood of E2). Of course I'd wish to see a sample where signals are nore powerfull like E1 or B0. Whatever!
The signal for the RH450 seems to be reverse. There is no information available if the RH450 does invert the signal or the trace was taken inverted *(-1) to make difference more transparent.
At 3.26 the amplifier brand X shows a noticeable peak at (+) half-plane that is around +3.5 dB above the following peaks. A burst capability of 20 mSec would be long enough to amplify this peak. 20mSec would be even long enough for E1. Note, the signal is very asymmetric thus the "strong" peaks are only present at the (+) half-plane for amlifier brand X.
The signal of the RH450 is inverted AND somewhat delayed (around 12 mSec ?). The strongest peak appears around 12 mSec delayed at (-) half-plane. Difference in peak value to brand X amplifier: -1.5 dB.
The signal in total looks very different to the brand X amplifier. Most noticeable is a reduced crest (peak to rms ratio) which is noticeable at any point on time axis. Hard to evaluate the difference of crest in numbers but I'd assume around 3 dB at least. In consequence this means this signal does not only "feel" louder, this signal indeed generates 3 dB at least more of power even if peak levels at both amplifiers were just the same. Also the signal will sound noticeable different IMO.
Any processing of signals like that can't be managed with compressors (or limiters), a signal like that is processed by a different algorithm.
Burst capability, although the brand X amplifier could clearly benefit on burst if there was one the RH450 does not. The signal is nearly eavenly processed regarding peak level values. That does not mean that just the moment of a stroke to the string there was "all even" regarding to RMS. There is a remarkable improvement of RMS noticeable at 3.26-3.29 while peak level remains somewhat "unfazed". Note, amplifier X is totally different on this.
That's comparable to a tube amplifier that does noticeable compression to the signal peaks. All of the peaks then are nearly even and therefore there is no more benefit from burst capability.
At 3.34 - 3.52 there is a difference in peak level of around 2dB that lets assume more power at RH450 side.
Note: peak levels for brand X are at (+) half-plane, peak levels for RH450 are at (-) half plane due to signal negation.
But that's only half of the truth because the second half of every cycle looks totally different in comparision. That's why I suppose obvious different crest numbers for the signals as well.
In consequence this means, even if peak levels where just the same, the signal with lower crest number will produce more power (that's pure physik).
To give the amplifier brand X the same average of power it is neccessary to give the signal 3-6 dB more gain.
In this case the strong peak at 3.26 (very likely) was clipped (chopped off) which isn't a problem because it's inaudible. It's inaudible as long as the amplifier does not produce nasty artifacts like it happens when digital headroom is "clipped".
For the record,
I tried to make an objektive consideration based on the signals shown above.
Last edited:
Bob Lee (QSC)
In case you missed it, I work for QSC Audio!
- Jul 3, 2001
- 12,668
- 2,036
- 67
- Disclosures
- Former Technical Communications Developer, QSC Audio
So-called "RMS" <> continuous.
Sans APM the power amp of RH450 produces peak power of 800 watts to 4-ohm load. Continuous average power rating is therefore probably considerably less than 400 watts but for a 20 msec. burst signal the amplifier can output 225 watts.
With APM engaged soft clipping of signal transients reduces peak power BUT 20 msec. burst power rating increases to 264 watts due to resulting signal compression. -Continuos- output rating is similarly improved and turns out to be 236 watts.
IMO, this is really no different from what many other manufacturers are doing...
...as in they quote output power ratings specific to the power amp only, ignoring effects of the preamp. So, if power amp can produce, say, 500 watts at low level of THD but clipping from preamp signal processing really limits the output power to 200 watts (at same level of THD) the manufacturers will still quote "500". Very. Commonly.
Now someone just made a big deal about it.
For example, Ampeg SVT-3PRO is quoted to provide output power rating of 450 watts to 4-ohms (IIRC) at some minimal level of THD. But what they don't quote is that turning the "plate voltage" / "tube gain" dial can actually reduce the output power to meager 60 watts because of clipping and distortion introduced by the circuitry associated to that special effect.
Similarly, engaging "soft clipping" feature in (was it GB Shuttle amp series) enables a soft clipping limiter, which actually again considerably reduces both amp's peak power output and continuous power output at comparable THD levels. With the clipping limiter on the quoted output power ratings will not be met.
And these are just few examples of what manufacturers are commonly doing: Power amp is rated independently, actual performance may be heavily influenced by "preamp" circuitry, which gets conveniently ignored. Though admittably, many simply pass these effects "as nice tubey compression" insted of starting a riot because the manufacturer intentionally mislead them. ...And I don't even want to go to effects of permanently on "built-in" compressor/limiters implemented to power amps than operate at fixed, non user adjustable settings. Most of them work by principle of reducing input signal levels under sustained overloads, which similarly means "dynamic" reduction in amp's output power.
Is it really that big deal?
With APM engaged soft clipping of signal transients reduces peak power BUT 20 msec. burst power rating increases to 264 watts due to resulting signal compression. -Continuos- output rating is similarly improved and turns out to be 236 watts.
IMO, this is really no different from what many other manufacturers are doing...
...as in they quote output power ratings specific to the power amp only, ignoring effects of the preamp. So, if power amp can produce, say, 500 watts at low level of THD but clipping from preamp signal processing really limits the output power to 200 watts (at same level of THD) the manufacturers will still quote "500". Very. Commonly.
Now someone just made a big deal about it.
For example, Ampeg SVT-3PRO is quoted to provide output power rating of 450 watts to 4-ohms (IIRC) at some minimal level of THD. But what they don't quote is that turning the "plate voltage" / "tube gain" dial can actually reduce the output power to meager 60 watts because of clipping and distortion introduced by the circuitry associated to that special effect.
Similarly, engaging "soft clipping" feature in (was it GB Shuttle amp series) enables a soft clipping limiter, which actually again considerably reduces both amp's peak power output and continuous power output at comparable THD levels. With the clipping limiter on the quoted output power ratings will not be met.
And these are just few examples of what manufacturers are commonly doing: Power amp is rated independently, actual performance may be heavily influenced by "preamp" circuitry, which gets conveniently ignored. Though admittably, many simply pass these effects "as nice tubey compression" insted of starting a riot because the manufacturer intentionally mislead them. ...And I don't even want to go to effects of permanently on "built-in" compressor/limiters implemented to power amps than operate at fixed, non user adjustable settings. Most of them work by principle of reducing input signal levels under sustained overloads, which similarly means "dynamic" reduction in amp's output power.
Is it really that big deal?
If the APM could be defeated, I might have a different opinion. But it can't, so it's bogus.
BTW, I'd really like to see some proof of this claim about the 3 Pro. I used to have one and never felt like it cut power by engaging tube gain, and anyone can claim anything if they don't have to prove it.
I would say that yes, it is a big deal. It's lying. Blatantly.
Hook it up to an oscilloscope. Observe output signal / output power when you adjust the dial.
The "plate voltage" / "tube gain" control of the SVT3-PRO adjusts B+ of the power amp's tube-based voltage amp stage. So the control will effectively reduce plate voltage of that stage resulting into earlier clipping of the stage (the whole purpose). This also means lower headroom of the entire power amp because there is no mechanism to "make up" the signal swing lost to clipping in the voltage amp stage. The final stage is just a "buffer", a current amp with unity voltage gain that just replicates the output signal of the voltage amp stage to much more complex and lower load impedances.
Quoted output power rating is met when the voltage amp runs cleanly at maximum plate voltage setting. (Highest voltage swing) Output power rating will therefore considerably reduce when output voltage swing of the voltage amp is reduced by deliberate clipping (lowest voltage swing).
Hearing the difference of course is more an issue of listening the difference in headroom/distortion than actual difference in perceived loudness. What is the difference of an amp producing transients cleanly vs. lower power amp that soft clips transients but in average produces about the same output power. The lower output signal may even sound louder because in average its much more compressed. And you need to remember that people perceive loudness from average levels and have weak perception to details of transients. We don't neccessarily even hear if they clip (assuming they don't clip very nastily) or if they get produced with pristine clarity.
Which is basically the idea with the TC amps...
Last edited:
Assuming that what you're telling me is true (I don't have an oscilloscope or 3 Pro anymore), I would say that since the tube gain can be defeated, that means the 3 Pro is not the same as the TC, where its APM can't be defeated.
I used a 3-Pro on various stages for years and can't say I ever had issues getting heard.
I don't think you need a mechanism to compensate for the lack of power with tube gain activated, since the use of the knob implies a desire for clipped tones, hence less need for clarity.
A compensation system I really liked however on the 3-Pro was the Bright switch, AKA brand new string tone for free.
I don't think you need a mechanism to compensate for the lack of power with tube gain activated, since the use of the knob implies a desire for clipped tones, hence less need for clarity.
A compensation system I really liked however on the 3-Pro was the Bright switch, AKA brand new string tone for free.
May be that's done for good reason. It's not uncommen with class-D power module that the unit runs hot if the average power exceeds a certain number that is atypical to generic audio program material.
IIRC the early ICEpower module were designed for save operation with 1/8 average at most in long term consideration (continous). In consequence that means they expected that "generic" program material does never undercut a distinct crest number in long term considerations.
You may ask what is "generic" in this context?
It can be considered as an average (like the receive of a radio station program) where lots of different signals must be amplified over a very long time. The program material contains speach, different kinds of music like country and pop songs and hard rock.
What does this all mean to the subject?
A signal like shown above for the RH450 is atypical regarding crest number thus it could cause more average than reasonably expected. Those power module were designed for "generic" applications rather than MI amplifiers.
If a power modul were designed for 1/6 in long term consideration then a strain with 1/3 caused by a rather atypical signal would overheat the modul over time.
I'd expect the RH450 signal in this 1/3 area because peak levels are obviously tweaked even at sustained swing of string. For a power modul that is designed to saveley amplify 1/6 average over time the consequence would be fatal.
The signal is very efficient at it's rms average and therefore able to strain the power modul with 1/3 average over time. The modul would gradually heat up till the moment the thermal protection is trggered.
Besides the (very likely redundant) burst capability there is theerfore additional need to restrict the signal to a distinct level regarding rms average. This could be managed by protecting 3dB of headroom for the output stage. Thus the signal can't exceed 1/6 of average, the average the power amp modul was designed for.
With an option to deactivate this 3dB protection there is also inevitably need to deactivate the processing of the signal itself. In this case the amplifier is able to amplify 450 watt, but just the same time get's lost of the processed signal characteristic. And instead the amplifier will gain more power at the output, the amp will then feel less noisy, the amp will produce less average power although output power as such is increased.
Looks like a scenario full of contradictions, but it's physics at it's best.
Ya, well I couldn't tell you why it can't be defeated...I can only tell you that it can't. And not that I'm saying there's anything wrong with that...I'm just saying TC should have been honest about it and not try to BS people.
Exactly.
If you market an amp as being 450 watts, it should be able to do that at peak, at least. Marketing the RH450 as being 250 watts RMS would have been fudging it a little, but acceptable, IMO. But marketing it as being nearly twice as powerful as it actually is? That's not really okay, in my book. Marketing it as 450 watts is using a number that has no connection to any part of the amp's performance. Not RMS power, not peak power, nothing. It's just a number they pulled out of their ass at random.
In this case (adequate power rating) TC had to provide additional information to the user like this:
"Although the power rating for the R$450 is determined in respect of conventional approach for power determination. TCE has to notice that this amplifier may easily produce considerably more average of power than you might expect from a solid state power amplifier in this class. We highly reccoment to ensure your loudspeaker system provides sufficiant power handling to prevent potential damage from loudspeakers.
TCE is not responsible for damaged loudspeakers if this is not taken into account by the choice of loudspeaker cabinets"
Better?
TC is hiring political speechwriters to write their ad copy now?
IMHO that's nonsense because things like RMS power and peak power have nothing to do with the subject and totally ignore that there are some more different but essential things important
Like it or not, this system we have for rating amplifiers is what we've got, and most everyone else seems to play by those rules.
No, nothing at all, to the contrary.TC is hiring political speechwriters to write their ad copy now?
Although ment as something like a joke but, addressed to all of the users of the RH450 it could be fatal to consider this amp as a poorly 230 watt amp. Regarding power handling of cabs this amplifier should be rather considered as a thing in the 400-500 watt classification to be fine.
Like a watt is watt and tube watts are not different?
300 svt watts are just the same as 300 solid state watts?
In perspective of conventional power rating 1% THD the svt hangs off around 100 watt or 150 watt at best. If I give the svt a second chance with 5%THD than may be 200 watt at best.
Last edited:
