Hello
As the subject reads, and this is a serious question, I will try to provide a specific example of amplifier differentiation to get away from the questions regarding generalizations, the question pertains to low frequency reproduction only.
Speakers:
B&C 15SW115 8Ohm in ROAR15 (series tuned 6'th order quarter-wave band pass design), rather high sensitivity in it's intended working range.
Current amp:
Lab.Gruppen IP450 (traditional class AB as far as I can tell, rated 100W into 8Ohm FTC).
more technical details here: Lab.gruppen IP Series IP 450 Specifications (Page 2 of 2)
Source material:
From high resolution FLAC material to spotify from Linux based PC via a HRT microstreamer host powered DAC.
Processing:
DBX Driverack 260, signal is LP filtered at 148Hz, LR48dB/Oct, no EQ applied in filtered range.
User case:
My user case is "hi-fi listening" at home, at times with a bit more more SPL (SPL is not the issue, i do not need more from the amp power wise).
Premise:
I rarely use more than 10W of output power from the amplifier.
Most often I use 1-5W (i guess).
Question:
Would I benefit in terms of sound quality, at levels of power up to 10W if I go for a higher power amplifier such as the Lab.Gruppen FP6400 class TD(?) (rated at 1280W into 8 Ohm FTC).
More technical details on the FP6400 here: Specifications - Lab.gruppen FP Series FP 6400 Specifications [Page 2]
Reason for the question:
A novice suspicion that the sound quality at low levels may be dependent on more than the specified power capability of the amp, the aforementioned novice guesswork would have these suspicions centered around power-supply stability, and capacity combined with the capability to handle back EMF.
Please let me know if there is any merit to my reasoning, or if I'm way off on this one, either one is most welcome, I ask since I recently came across a FP6400 for none-obscene amounts of money locally.
It is a strange and somewhat odd question I know, please let me know if further clarification is required and I will try to provide it.
As the subject reads, and this is a serious question, I will try to provide a specific example of amplifier differentiation to get away from the questions regarding generalizations, the question pertains to low frequency reproduction only.
Speakers:
B&C 15SW115 8Ohm in ROAR15 (series tuned 6'th order quarter-wave band pass design), rather high sensitivity in it's intended working range.
Current amp:
Lab.Gruppen IP450 (traditional class AB as far as I can tell, rated 100W into 8Ohm FTC).
more technical details here: Lab.gruppen IP Series IP 450 Specifications (Page 2 of 2)
Source material:
From high resolution FLAC material to spotify from Linux based PC via a HRT microstreamer host powered DAC.
Processing:
DBX Driverack 260, signal is LP filtered at 148Hz, LR48dB/Oct, no EQ applied in filtered range.
User case:
My user case is "hi-fi listening" at home, at times with a bit more more SPL (SPL is not the issue, i do not need more from the amp power wise).
Premise:
I rarely use more than 10W of output power from the amplifier.
Most often I use 1-5W (i guess).
Question:
Would I benefit in terms of sound quality, at levels of power up to 10W if I go for a higher power amplifier such as the Lab.Gruppen FP6400 class TD(?) (rated at 1280W into 8 Ohm FTC).
More technical details on the FP6400 here: Specifications - Lab.gruppen FP Series FP 6400 Specifications [Page 2]
Reason for the question:
A novice suspicion that the sound quality at low levels may be dependent on more than the specified power capability of the amp, the aforementioned novice guesswork would have these suspicions centered around power-supply stability, and capacity combined with the capability to handle back EMF.
Please let me know if there is any merit to my reasoning, or if I'm way off on this one, either one is most welcome, I ask since I recently came across a FP6400 for none-obscene amounts of money locally.
It is a strange and somewhat odd question I know, please let me know if further clarification is required and I will try to provide it.
Last edited:
The more "power", the more complex the circuits.
The more parts.
The more...
The more modulation of signal...-)
Your question: High power amp for low power needs - better result?
My answer: NO;-!
"A novice suspicion that the sound quality at low levels may be dependent on more than the specified power capability of the amp, the aforementioned novice guesswork would have these suspicions centered around power-supply stability, and capacity combined with the capability to handle back EMF."
Better;-)
Clamp, brace the drivers;-)
The more parts.
The more...
The more modulation of signal...-)
Your question: High power amp for low power needs - better result?
My answer: NO;-!
"A novice suspicion that the sound quality at low levels may be dependent on more than the specified power capability of the amp, the aforementioned novice guesswork would have these suspicions centered around power-supply stability, and capacity combined with the capability to handle back EMF."
Better;-)
Clamp, brace the drivers;-)
Hhmm... As if an alien asks to explain how human emotion erupts from experience.
Case 1:
Hiraga 12W amp on hungry electrostatic loudspeakers (~50W+ recommended): DEVINE!!!
Case 2:
Philips 2W tube amp on diy transmission line 4-way dynamic wall-filling speakers (200W):
DEVINE!!!
Luxman receiver mid eighties on telly side speakers including loudness and some treble-bass lifting: big OK!
Case finite:
Concert hall, live performance, orchestra, programmed: Daphnis et Cloë, Ravel
Ultimo reference period
Frankly I have no idea what you are after at.
Technology serves humans, humans abhore technology.
(It's anyone's choise between friction or agility.)
This is a very nice discussion anyhow.
Case 1:
Hiraga 12W amp on hungry electrostatic loudspeakers (~50W+ recommended): DEVINE!!!
Case 2:
Philips 2W tube amp on diy transmission line 4-way dynamic wall-filling speakers (200W):
DEVINE!!!
Luxman receiver mid eighties on telly side speakers including loudness and some treble-bass lifting: big OK!
Case finite:
Concert hall, live performance, orchestra, programmed: Daphnis et Cloë, Ravel
Ultimo reference period
Frankly I have no idea what you are after at.
Technology serves humans, humans abhore technology.
(It's anyone's choise between friction or agility.)
This is a very nice discussion anyhow.
Last edited:
As I understand your question: Is there any point (quality gain) in using a high power amplifier when you anyway only use it for lower power levels? No, your sound quality depends on the performance at the power level you use, not on the unused power reserves of your amplifier.
The only very hypothetical situation I can think of is if the music is highly dynamic such that you use typically 1-2W but suddenly have a transient demanding 150W for a short moment. Then, evidently this very unusual transient will require quite some power reserves of the amplifier.
It is easier to design an amplifier for moderate power levels that performs fantastic than a high power amplifier with the same performance at low power.
The only very hypothetical situation I can think of is if the music is highly dynamic such that you use typically 1-2W but suddenly have a transient demanding 150W for a short moment. Then, evidently this very unusual transient will require quite some power reserves of the amplifier.
It is easier to design an amplifier for moderate power levels that performs fantastic than a high power amplifier with the same performance at low power.
More power more difficult to get high quality of an amplifier.
I designing amplifier for several watts to several hundred watts class AB. Some class H using 130VDC PSU, but the stepper is not mine.
The most difficult to achieve in high power amplifier is enough slew rate and low distortion at high frequency. So, average high power amplifiers use for driving sub woofer.
But, all depend on the design of the amplifiers. Low power amplifier if the design is bad can not have enough slew rate, and have high distortion.
I designing amplifier for several watts to several hundred watts class AB. Some class H using 130VDC PSU, but the stepper is not mine.
The most difficult to achieve in high power amplifier is enough slew rate and low distortion at high frequency. So, average high power amplifiers use for driving sub woofer.
But, all depend on the design of the amplifiers. Low power amplifier if the design is bad can not have enough slew rate, and have high distortion.
Depending on the particular amp, the limitation of the SQ might be in the PSU. In which case using a step-down toroidal output transformer would give you a performance lift - in particular getting you better subjective dynamics by virtue of giving the amp an easier load to drive. With a 100W amp and only 10W required that's a 3.2:1 step down.
The big amp has modestly "worse" THD and DIM specs.
At >1W. Nothing said about lower power.
Those are some serious LOUD speakers. I can't imagine using much of 100W in the home (I have done 600-seat theaters with such speakers and less amp). You might be fine with a good clean 10-Watt amp. (A guitarist would blow you away with this speaker on a 5.6 Watt Champ amp.)
These are excellent LOUD PA amplifiers. PA amp performance at *LOW* levels (such as you need) varies; some fine some a bit hoarse.
There is the factor that a *healthy* FP6400 for 'non-obscene amounts of money' may have solid re-sale value in the local rock-shop.
At >1W. Nothing said about lower power.
Those are some serious LOUD speakers. I can't imagine using much of 100W in the home (I have done 600-seat theaters with such speakers and less amp). You might be fine with a good clean 10-Watt amp. (A guitarist would blow you away with this speaker on a 5.6 Watt Champ amp.)
These are excellent LOUD PA amplifiers. PA amp performance at *LOW* levels (such as you need) varies; some fine some a bit hoarse.
There is the factor that a *healthy* FP6400 for 'non-obscene amounts of money' may have solid re-sale value in the local rock-shop.
Attachments
Thanks for all the replies, if I condense all your answers they all seem to point towards a big NO, the bigger amplifier will not improve anything with regards to sound quality compared to the smaller amp if only using them at very moderate levels.
I was thinking that a speaker like this (200g moving mass, 4,5inch Voice coil, 1,8mH (Le), 30 (BL), would pose a bit more challenging for the for the smaller amplifier (IP450) to control, back-emf wise, especially if housed in a series tuned QW enclosure as it the case here, and that the bigger amp (FP6400) would then be better suited for such a task even if used at very moderate levels?
I was thinking that a speaker like this (200g moving mass, 4,5inch Voice coil, 1,8mH (Le), 30 (BL), would pose a bit more challenging for the for the smaller amplifier (IP450) to control, back-emf wise, especially if housed in a series tuned QW enclosure as it the case here, and that the bigger amp (FP6400) would then be better suited for such a task even if used at very moderate levels?
... Wait a moment ...
But what about the Beta-Droop of the output power transistors? The gain of the output devices normally tends to "sag" as a function of increasing collector current.
In the case of a low power amplifier, which is being pushed to its power limits, the "sag" of the gain ('beta-droop') may become an observable quality issue. My reasoning for this is that a low power amplifier has a small amount of paralleled output transistors. Therefore, when approaching the upper limits of its power capability, such output transistors will (imo) be working in high current, "large signal" conditions, and hence susceptible to beta-droop.
On the other hand, if we have a somewhat "beefier" amplifier, in terms of power capability, this is most probably achieved by a). higher rail voltages, b). increased number of output transistor pairs, c). driver with stronger capability, so as to control such increased number of transistor pairs.
Now, if we use such a beefy power amplifier, for the purpose of achieving a similar level of volume as in the case of the low power amplifier, then the following situation (imo) emerges:
a). The signal is within a much lower percentage of the total output voltage capability of the device (a lower percent of the maximum possible output voltage swing, considering the rail voltages). Therefore, this probably does not represent "large signal conditions" for this amplifier, which is normally, by design, capable of delivering a much higher output swing.
b). Since there is an increased number of output transistor pairs, the average signal current, being handled by each such transistor, is significantly lower, as would be the case of the low-power amplifier. Therefore, I would assume that the beta-droop of the output devices is not so painful, as the current (per transistor) is not in its extremes.
c). A lower fraction of the driver capability is actually being used by the output devices, due to the lack of beta-droop. The driver is therefore not "hard driven" in terms of what needs to be provided to the output transistors. Would that not enable a more relaxed mode of driver operating point / operations, and generally a higher impedance as seen by the voltage amplifier stage, just prior to it?
Independently of all of the above, a beefier amplifier has a beefier power supply with bigger capacitor bank and bigger short term current pulse capability, which would probably translate to more punch and authority in the base line.
I am most probably wrong in this reasoning, but I tend to sympathize with the concept as presented in post #1. If indeed in error, please explain why.
I would imagine that the complexity of a power amplifier of say twice the power of a benchmark amplifier would not be that much higher, so as to deteriorate sound quality, if the more powerful unit is correctly designed.
But what about the Beta-Droop of the output power transistors? The gain of the output devices normally tends to "sag" as a function of increasing collector current.
In the case of a low power amplifier, which is being pushed to its power limits, the "sag" of the gain ('beta-droop') may become an observable quality issue. My reasoning for this is that a low power amplifier has a small amount of paralleled output transistors. Therefore, when approaching the upper limits of its power capability, such output transistors will (imo) be working in high current, "large signal" conditions, and hence susceptible to beta-droop.
On the other hand, if we have a somewhat "beefier" amplifier, in terms of power capability, this is most probably achieved by a). higher rail voltages, b). increased number of output transistor pairs, c). driver with stronger capability, so as to control such increased number of transistor pairs.
Now, if we use such a beefy power amplifier, for the purpose of achieving a similar level of volume as in the case of the low power amplifier, then the following situation (imo) emerges:
a). The signal is within a much lower percentage of the total output voltage capability of the device (a lower percent of the maximum possible output voltage swing, considering the rail voltages). Therefore, this probably does not represent "large signal conditions" for this amplifier, which is normally, by design, capable of delivering a much higher output swing.
b). Since there is an increased number of output transistor pairs, the average signal current, being handled by each such transistor, is significantly lower, as would be the case of the low-power amplifier. Therefore, I would assume that the beta-droop of the output devices is not so painful, as the current (per transistor) is not in its extremes.
c). A lower fraction of the driver capability is actually being used by the output devices, due to the lack of beta-droop. The driver is therefore not "hard driven" in terms of what needs to be provided to the output transistors. Would that not enable a more relaxed mode of driver operating point / operations, and generally a higher impedance as seen by the voltage amplifier stage, just prior to it?
Independently of all of the above, a beefier amplifier has a beefier power supply with bigger capacitor bank and bigger short term current pulse capability, which would probably translate to more punch and authority in the base line.
I am most probably wrong in this reasoning, but I tend to sympathize with the concept as presented in post #1. If indeed in error, please explain why.
I would imagine that the complexity of a power amplifier of say twice the power of a benchmark amplifier would not be that much higher, so as to deteriorate sound quality, if the more powerful unit is correctly designed.
Consider differences in the operating class and operating conditions. The big sucker is class TD - the smaller one is not. With the tracking output stage it can be biased closer to optimum without running up a stupid amount of idle dissipation. So the output stage in the bigger amp probably isn’t paying a penalty here, and may have an extra pair or two just to make peak Ic. I wouldn’t expect crossover distortion artifacts to be any worse with the FP. However, the entire front end is definitely more optimum in the smaller amp. I know from designing and building big 2kW+ amps that YOU CANNOT get away with nice low Cob VAS transistors with 300 or more volts between the rails. They blow up from second breakdown because they have to pass 10s of mA at full voltage. Designers have to resort to large die, slow devices which were really intended for switching power supplies. Using 8 amp devices just to get 50mA Is/b capability at 350 volts is not uncommon. It’s what you have to do, there’s no getting around t. It will not prevent decent performance, but building A 5 ppm THD20k amp at those voltage levels is just out of the question. Likely the differences between the two amps is there. The bigger one will also have some artifact of the downconverter action above a few 10’s of watts. I doubt one would ever hear it, but it will show up in measurements if you know where to look.
Thank you all for your help, I take it that there is no easy way to deduce if the FP6400 is less load sensitive than the IP450 in the very low power range of up to 10W given the information I provided, fully understandable, and at this level both amplifiers are hardly sweating at all, and they are both 2Ohm capable if that is any help.
Still my gut feeling (does not account for much in this case I know) tells me that since the common user case for the FP6400 is to power a couple of 218's per channel at clip level, providing several kW of power maybe even into low impedance such as 2 Ohms, and therefore designed for this.
That is pretty far from the delicate (in comparison) IP450, whose intended purpose I have yet to figure out, studio monitors, small fixed installs, HF sections of more powerful speakers perhaps, completely different design criteria in other words, and it does sound just fine on top speakers.
Given these rather different natures of their purpose one is inclined to think that the FP6400 would be less sensitive to back EMF and more stable or more in control of the load so to speak, but this is just a guess of mine and nothing in their spec's would let you know or hint at there is such a difference.
I guess we will never be able to figure this out from the specs, the only way to know for sure if there is a noticeable difference in low frequency sound quality (control) at low levels is to simply test it side by side and listen, shame, it would have been good to know if there at least was a likelihood that the FP6400 would do a better job.
Still my gut feeling (does not account for much in this case I know) tells me that since the common user case for the FP6400 is to power a couple of 218's per channel at clip level, providing several kW of power maybe even into low impedance such as 2 Ohms, and therefore designed for this.
That is pretty far from the delicate (in comparison) IP450, whose intended purpose I have yet to figure out, studio monitors, small fixed installs, HF sections of more powerful speakers perhaps, completely different design criteria in other words, and it does sound just fine on top speakers.
Given these rather different natures of their purpose one is inclined to think that the FP6400 would be less sensitive to back EMF and more stable or more in control of the load so to speak, but this is just a guess of mine and nothing in their spec's would let you know or hint at there is such a difference.
I guess we will never be able to figure this out from the specs, the only way to know for sure if there is a noticeable difference in low frequency sound quality (control) at low levels is to simply test it side by side and listen, shame, it would have been good to know if there at least was a likelihood that the FP6400 would do a better job.
Last edited:
I would expect the amps to behave similarly in terms of “control” as long as the global feedback, or at a very minimum, local feedback in the output stage, is in control of the amplifier. A loss of control (and therefore damping) can occur at clipping, absolutely will occur during current limiting, and may occur if a difficult load forces the amp to run out of open loop gain. If these conditions never happen with your load and power levels you would be hard pressed to hear any difference at all.
The most common reason for differences in “control” is whether or not you’re clipping. Clipping can occur more often than you realize - the average power does not necessarily need to be high. Some amps will go from sounding very clean below clip to positively awful - the worst offenders being common collector output stages. With some amps, you can’t even really tell there’s clipping until it’s already gotten dangerous for your tweeters. That’s the kind of amp you want on a subwoofer, because the output impedance remains low (tenths of ohms) even when clipping. I suspect the both the Labgruppen fall into that category. I have amps that fall into both, and the difference is night and day at war volume.
The most common reason for differences in “control” is whether or not you’re clipping. Clipping can occur more often than you realize - the average power does not necessarily need to be high. Some amps will go from sounding very clean below clip to positively awful - the worst offenders being common collector output stages. With some amps, you can’t even really tell there’s clipping until it’s already gotten dangerous for your tweeters. That’s the kind of amp you want on a subwoofer, because the output impedance remains low (tenths of ohms) even when clipping. I suspect the both the Labgruppen fall into that category. I have amps that fall into both, and the difference is night and day at war volume.
When the power supply runs out of gas, either voltage or current, the amplifier clips. Switchers have a nasty habit of going into current limit - usually they don’t have the reserves to put out a half sine much lower than about 200Hz at full peak voltage into 2 ohms. The LGs are better about this than most (will do 20 Hz in burst mode).
When an amplifier does clip, one of 3 things happens. The worst case is the output becomes a current source - having a high output impedance. This happens in current limiting and when a common collector stage clips. All control over the speaker is lost until feedback returns. If a good common emitter stage clips (Vce reaches minimum) the output impedance is dominated by the power supply impedance. If it’s low, you still have control, if it’s high you don’t. Most PA equipment does this. The better the supply, the better behaved, and basically you get what you pay for. The last possibility is if the output followers never saturate when the front end clips. Many hi-if amps were designed this way. The downside is you have to waste a couple volts of potential output swing. And to be the most effective, waste a bit of open loop gain with local feedback in the VAS (lower output impedance, but better linearity and load driving ability). The output impedance will stay near zero (dominated by paralleled emitter resistors) even if all global feedback control is lost. Amps like this can clip pretty severely with typical pop/rock music and you won’t even hear it will your tweeters are toast.
When an amplifier does clip, one of 3 things happens. The worst case is the output becomes a current source - having a high output impedance. This happens in current limiting and when a common collector stage clips. All control over the speaker is lost until feedback returns. If a good common emitter stage clips (Vce reaches minimum) the output impedance is dominated by the power supply impedance. If it’s low, you still have control, if it’s high you don’t. Most PA equipment does this. The better the supply, the better behaved, and basically you get what you pay for. The last possibility is if the output followers never saturate when the front end clips. Many hi-if amps were designed this way. The downside is you have to waste a couple volts of potential output swing. And to be the most effective, waste a bit of open loop gain with local feedback in the VAS (lower output impedance, but better linearity and load driving ability). The output impedance will stay near zero (dominated by paralleled emitter resistors) even if all global feedback control is lost. Amps like this can clip pretty severely with typical pop/rock music and you won’t even hear it will your tweeters are toast.
At the other end of the beta-drop discussion, which is very true and a major ingredient when designing a solid main amplifier, in daily use it boils down to this discussion:
How to spot a HIGH CURRENT amp design?
Currently I'm designing a intertwined balanced and bridged Hiraga with four legs on each side of the speaker terminals, running at 2.5A each leg.
Only ment to ignite the aformentioned cannons during a wintercampaign, when it's cold enough to prevent the audience from a hellish frying when this dragon stretches it claws.
How to spot a HIGH CURRENT amp design?
Currently I'm designing a intertwined balanced and bridged Hiraga with four legs on each side of the speaker terminals, running at 2.5A each leg.
Only ment to ignite the aformentioned cannons during a wintercampaign, when it's cold enough to prevent the audience from a hellish frying when this dragon stretches it claws.
Beta droop certainly can contribute. At high currents the open loop gain drops, increasing distortion and increasing output impedance, possibly enough to be a problem. In an extreme case, the VAS can’t supply enough current, the output stage becomes a current source (Beta times the VAS current) with a high output impedance. Muy bad.
Amps can be made to be beta insensitive, but it usually takes an output triple to do it. Some people are scared of them. They always have poorer phase margins, and most implementations have a significant thermal lag in the bias compensation. I’m pretty sure the LG’s use triples.
Amps can be made to be beta insensitive, but it usually takes an output triple to do it. Some people are scared of them. They always have poorer phase margins, and most implementations have a significant thermal lag in the bias compensation. I’m pretty sure the LG’s use triples.
Just to clarify, we are not considering clip or even near clip behaviour, and when I think about it, maybe the question should have been described better, not that your answers has not been great and very educational, at times perhaps a bit above my head being a speaker guy more than a amp guy 
Lets look at it from the other perspective, -What factors (or design criterias) of an amplifier influnces the loose term "cone control"? and perhaps related (not sure) -when does a load in terms of back emf (reactive current and voltage) become a problem for the amp and why?
I used to think that it was the specified damping factor, I later learned that it is more of a marketing figure, previous to that I was lead to belive it was the current capacity of the amp (Harman Kardon made a big deal about this) now i tend to think more in terms of how stable the powersupply is and sensitive groudplane is for disturnaces, but I'm far from sure that I'm any closer still...
I asume this is not at all easy to answer, and i hold out little hope that it can be explained in an easy to understand fashion (more related to my understanding than yours I'm sure) and just to remind you the two amplifiers I'm want to compare in this fashion are the labgruppen IP450 and the labgruppen FP6400, two extremes, in order to understand which of them might be better at low levels (<10W) low frequency reproduction (up to 200Hz) given an asumed rather tricky load (back emf wise).
Lets look at it from the other perspective, -What factors (or design criterias) of an amplifier influnces the loose term "cone control"? and perhaps related (not sure) -when does a load in terms of back emf (reactive current and voltage) become a problem for the amp and why?
I used to think that it was the specified damping factor, I later learned that it is more of a marketing figure, previous to that I was lead to belive it was the current capacity of the amp (Harman Kardon made a big deal about this) now i tend to think more in terms of how stable the powersupply is and sensitive groudplane is for disturnaces, but I'm far from sure that I'm any closer still...
I asume this is not at all easy to answer, and i hold out little hope that it can be explained in an easy to understand fashion (more related to my understanding than yours I'm sure) and just to remind you the two amplifiers I'm want to compare in this fashion are the labgruppen IP450 and the labgruppen FP6400, two extremes, in order to understand which of them might be better at low levels (<10W) low frequency reproduction (up to 200Hz) given an asumed rather tricky load (back emf wise).
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.